7^71 Oisxmzlh 7H&1 % S8

KAERI/CM-9 13/2005 £l§tiILW PE-68201-RZ-Z001

7^71 OISXMZLH 7H&1 % S8 Development and management of proton accelerator user program

XH5 ^ iS^|7||o| ^USOII LHti SAt 1J![

Proton irradiation effect on hydrogen bonds in materials and biological systems

°i ^ *} 7) 4 ^ ^ 7] # £ A> <9 £ 2006. 4 444^44^ 4

H> r(ti ^444 4 ^_Jl4#

4 3)

04. a a a %

4 4 4% 28.

4^ 444 : : : : : :

44 aL^tfltJ-H 444 444 4e 444 444

4^44 2:4 o]^}Zg_3.^

3 l

4) #7^4^ ”

4 4

44M4S. 7^

44 # 1

1=7 44

44444. 444

^ZMl

"(7l]&

4444 KAERI/ 1111 414414 2005.7.11006.3.31 11 44 (1111) / (411) CM-913/2005 1444 m 4 1 4 H 14414 1144 4444 #1 14411 14 11 1& % 1111 4^4%! 1% 117} &7} J&3} 14414 14411 414444 !44^g.cz.;@ 7]]# ^ ^rl #:6 1 14: 35,000 41 1111 1111 14114 114 14 : 6 1 41: 0 44 411444 14 : 0 1 144 1: 35,000 44 14441 3.4 l%a 4444 4 A%44% 144 41411 ^4444 414#14 4141: 41414441: 4 4 % 4 14441: 14114: -8.4(1444# 44A2. 7^1 50044^) iLJLllh 1. ^ R]^- 1^4&* &442.& 11%^ m !4%%4 44% 41 1#& 4#%7] 44 4 4% %4 %1 ^7^4 4^%%! 4% 117} 44414.

44 4 til ZZ-,4 44 45-i 4% 42- 4%: KH2PO4 41 41, urea, squaric acid, SnCl2-2H20, 2-4 4414 41 44 44 a-HH 4^4% network^ 44 414 4% 4^4%, native defect 44 44 1^ #4^4 4 0]* #4^4 l7i] * 4dhi}% defect % 4 4 4 % 444 2:44 4^%% defect 3l4 If: X-ray 44 1 42:1 4 44-4 2 4 44 AC/DC proton conductivity 44 1 ^ @94141 rH NMR), IR spectroscopy 4E 11 (KH2P04 41 41, lr4, graphite 4' 44 4 tii ?. 4* 4% As., 44 11 #71144 44% -a.. |%1 4% 114 1% H: KDP dl 1 4H, Proton Conductor, Carbon nanotubes 15. %2. 4 %} - EPR, NMR 144 4% 41 414 &4 2:1 41 4 &4 41 2. 1414 * 114 1 &4 14 !&! KH2PO4 11 41, 2-1114 41, carbon nanotube 414 14 4 11 4% 14 14 444 7]]7ii (SCI 44 25%, 41441& 15%, 414412 10%) * 11%} 2:4% KDP, TDP 1&4 41 414 4% AC proton conductivity, XRD, NMR, EPR, IR, Raman 4 444 114 42. 41, 114 14% &44 XRD %4&44 444ml #7}, VSM, EPR %4&44 paramagnetic impurity 141 1144 41 41441 % (Material Research Society 2005 Fall Meeting, Boston), (ICAMD) IS Physical Review B 4 41 %44! H 12: agefcwm ti&M* &N% bM'aol 5-Lee'and x Kim' 'Pmton ~beam

111'^ ll7}&4, 4^4%, 4441, 441, 44^1 proton-beam irradiation, hydrogen bond, magnetic resonance, ferroelectricity, (4 571] 14)11 1 T1H2P04, KH2P04

- 2 - SUMMARY

* Aims and details of the project

To control a nano structure and develop relevant technology, understanding of hydrogen bond under various environments is necessary. For this purpose we conducted the following:

* Study of hydrogen bond networks in the solid state (KEpPC^-type materials and others)

* Study of proton-beam irradiation effects and hydrogen bond defects 1. X-ray diffraction patterns 2. Structural change analysis 3. AC/DC proton conductivity measurements 4. Hydrogen nuclear magnetic resonance (XH NMR) 5. IR spectroscopy

* Examination and understanding of irradiation effects on various materials

Nuclear magnetic resonance, electron paramagnetic resonance and luminescence study of graphite and polymer systems

* Results of the project

* Analysis of the physical properties of proton beam irradiated samples such as KH2PO4 type materials, graphite, and polymer systems (25 SCI papers)

* Data (AC proton conductivity, XRD, NMR, EPR, IR, Raman, etc.) for the proton beam irradiated KDP, TDP: . lattice expansion after the proton beam irradiation . paramagnetic impurities confirmed by VSM, EPR after the proton beam irradiation

* Presentation at (Material Research Society 2005 Fall Meeting, Boston, ICAMD)

- 3 - CONTENTS

Chapter 1: Abstract of the project ...... 6

1. Title...... 6 2. Purpose and necessity of theproject ...... 6 3. Details and range of the project ...... 8 1) Goal s ...... 8 2) detai Is ...... 9

Chapter 2- Development of the technologies at home and abroad ...... 10

1. Report of the present condition of technology (State of the Art Report) ...... 10

Chapter 3: Detailed results ...... 11

1. Strategy and systems ...... 11 2. Result ...... 13 1) Details ...... 13 2) Results and discussion ...... 13

Chapter 4: Achievements and contribution to related field...... 16

1. Achievements and self-evaluation ...... 16 1) Achievements ...... 16 2) Self-valuation ...... 16

2. Methods ...... 17 3. Summary ...... 17

Chapter 5: Application plans of the results ...... 18

1. Application plans of the results ...... 18

Chapter 6: Scientific technology abroad obtained during the project ...... 19

1. Foreign scientific technology abroad ...... 19

Chapter 7- References ...... 20

Chapter 8: Attachment ...... 21

- 4 - ii ~ 1 *}

*ll i & 997W2i*M 6

1. 4 4...... 6 2. #97f44 #3) Rt si.a.3...... 6 3. #97M44 4§- # #4...... 8 1) # V-7}| u>»| y| .ih # 8 2) #4'.M # V-7F-!>■!';■ ih * 49 ...... 9

*n 2 3 fulfil p\^m ##...... 10

1. 7|-Iff! # ,‘91 V. (State of the Art Report) ...... 10

*113 3 297^911 uis % m\...... ii

1. 9# 3 4 ^ 4711...... 11 2. #97^4^4...... 13 1) #9-4-8-...... 13 2) #9^4...... 13

*11 4 3 =Si^£ DA 9.'@g0l0||o| 7led5 ...... 16

1. #97:4 %4 49s ...... 16 1) #94 'll %-92| V14.V...... 16 2) 97] #7}# 4# 9 4 4# ^g- 49 s4 4# 44#7} 44...... 16

2. #999 4#...... 17 3. WS...... 17

*11 5 4 997h 41212.' 18 i. #97:4^44 4-8-44 ...... 18

*11 6 4 9944434*1 435- 5hs 2-47199 y ...... 19

1. #971144343 93# *114 44# #3 S...... 19

*11 7 3 #2ig#...... 20

*11 8 3 3¥*rs...... 21

- 5 - n\ 1 g jyl7H |*4*ll S I 7HS

1. 4 4

7H2 ^ 4444 ###44 4% 444 #4 24

2. ^4;#^ 4^ ^ #_&-4

4# 7]## 2# #4 (#24, #2, ## #Xl]# 7]# 44# 4-442 2147]

44 # x3 4 # 4244 4 42a ] #24 4 4 4 444. 44 42 27] 44 #4# x># 24 4442 #44 4 4# #444## 44, 4-44 #4 #44 2#4# 44#4

44, protein folding, 242 drug delivery 44 44 44# 4442 444 4 4# 4 4# 2444. #4, #444 (Fig. 1)4 44 #4 (self assemble)# 444 4242 4

444 44 4 supramolecular chemistry, #41 tflh| protein folding 7] 4 #4 4442

4# 424 44# 4# 422 #4 4 44. 2422 42#2# #4422 4442 ^44 4^44# ##4 441 4## ^#e}7] 444# 4# 444 44 44 #414 # ###44 4 4 444 ##444.

- 6 - 1 ?FV4

X H 0 I! X—H.

X—h:

The xvhofe pafette H -w O-G- II ■ of hydrogen 6onds

Fig. 1. ##

- 7 - 3. #77#^ ^ ^

1). 9#7]]#a] 4 99#

4 44a: #99 9 24#- 44 4=, 44 zlb]jl 99 #*!]#-# 444 4 299°ll 414 494 nt 49, #99 9# EPR (electron paramagnetic resonance) dosimetry

KH2PO4 4] 9 #9, urea, squaric - 4444 #44- 9 (~MeV)4 acid, SnCla^HaO #: proton 1 94 44 #299# 444 444 conductivity 419 9 429"94 4 #299 44 44 4

- #44 4 24-#- #4 4 a, 44 a#2 Tgxl] ^ol] #3]]&H^ 4#4 44444 294 44 - powder X-ray# 4# 9 4 4 42 49 44 7fl4, #44 9# EPR 9"# 44 (Treor 90 program) (electron paramagnetic - 344 999 x-ray# 4#9 #29# resonance) dosimetry # 99 99# 94 9 4a 44 99 - #44 444 444 4a2M] - j7 7ii a4 4a 244 999 99 9 # #44 #299# 4492 44, 7H 9 (Hetero correlation) 29 4 444 44 44 44 4# - al# ##7] 2^4 # 4: magic angle 294 7fl4 spinning 7]#-# #4# #99 4 - #2999 #44 4492 44## 4# modification, 44444 tunable - 449 4 a# nmr 4a# #4 4 4 technology 7]] 4 # 4##4 944## 9# #4 ^ 4 - 21417] 4-414 44444 99 4142 44 4#4 4a94 #4 7fl# 99 4 - 44# #9 f #29## interbond, 9 intrabond# 44 #2 44 44 - #99 499 4144# #9 r #4 #9 4S-94 9 proton ionic conductor 7] 99 #4 4#

•KH2PO4 49 #9, urea, squaric - 20 MeV-60 MeV# #99 9 acid, SnCh^EpO, organic/inorganic 2:441 4 9 damage# 499 hybrid systems, supramolecule #: 97199 9 EPR dosimetry 1, 24 94 944 4a (#4-#, # 394 494 #9 defect 944 9a). - ~MeV #99 94] #4 #2 9## 444 4144 #29# •DNA, proteins: DNA 4 #4942 4# 49 # zipping/ unzipping dynamics, protein# folding/unfolding #2

- 8 - 2) #44 4 4#

441^7 41 *444# * g #*;## 4S- 51 4 4 S7p|# (#4) •KH2PO4 51 # if4, urea, squaric acid, SnCl2-2H20 #4 ## 31514 541 43:44 network# 4# # 451 4# 4444 4 native defect51 44 4# 4 44 44 4 4 #4 -4## 43:44# 4^4#5 44 -#44 2:^1- 4 4 44 44 #44 4 444 #44 ^Ai- 4^: setting 4 #44 5:4 station 44 *l| 14# 5 # #3:## * 541# *4, 4*3 (30 %) : 43:^44 4##4 44 44, 35,000 44 4##445-4 44: 10*10 mm, 4 25> 2: 4* 2#, #4 #* 4: 4 MeV, 1015 ions/cm 2 44541 4444 2# #44 2:4 •Proton 4 Hellium 4 # 21# #- graphite 4 a} 7142: 4# 44, sample station : 10*10 mm, 4 2:4 54: 4 MeV, 1015 -1018 ions/cm 2 ##514 Hellium 4# ^4 -X-ray #4 4 4^4# 44 - 4 3: 4 # •#5151 4# AC/DC proton defect 44#conductivity #4 4 4# ^471# 4 (:H NMR), IR spectroscopy #5 *l| 244 5 41# #44 Z: 44 51#44 * 4# (KH2P04 51# #4, urea, 35,000 (70 %) ^1-4 #### squaric acid, SnCl2-2H20 *), - 4 * ^ 44 defect S4 #. *1*4 #44 444# 4 modification, 43:444 tunable technology 7fl4 •2445 4# 44544 kh2po4 514 #4, urea, squaric acid, SnCl2 •2H20 ## 454# defect 51 4# 54# 4# #5 0151 rfl -5#4^ ^ # 4tt4 simulation# 4*3#. 4*4 4*51## *l| 34ti5 -#*1#4 4^4 414* ## (100 %) 4 * 4 proton conductivity 514 444 #* ^ 4 35,000 simulation *3 ** - 21414 4414 445144 454 44 #544 #4 514 44 44 - #44 #*# 5N4* #4 4 3:4 44 #5444 proton ionic conductor 444 3:5! 7(14

- 9 - n 2 5 ^lh 2.| as

1. 7] (State of the Art Report)

- #4 44444 4% ^4 ## l) 444 1#: • Ionized radiation biology (4 444) - #4 41#4444 #44&44 4# 44 44 44 : Armed Forces Radiobiology Research Institute (AFRRI, 44) - 4A4 ^4# #-§-4 44#44#4- 3^ ##4# 44 :44444444 - 44 4444 4444 444 44 44 : 443)144 444 (44)

• ERR dosimetry (4 ft 441) - Alanine/EPR therapy level reference dosimetry service : NDRL Radiation Chemistry Data Center (Notre Dame Radiation Laboratory, 44) - Alanine-EPR Industrial dosimetry : Victoreen Inc. & Micro-Now Instrument Company Inc. (44)

• X-ray 544 44 3.4144 damage 444 EPR (4444) - CaS04 44 4444: NDRL Radiation Chemistry Data Center (44) - 4A4 ^44 44 DNA4 44# 4^ 44 : Saarland university 4 J. Huttermann 3# (44)

2) 444 1#: • 4"44 &44 44 4 44 damage 44 (4444)

- 444 44 4 4 444 &4A 41: 44 4414

. ^A}^ 4 #4 4 4 (4444) - 4# 44 #444 4 #4 41## ### #442. #44 ^4

: 444 Is #44444 4 44 44

3) ^44 444#444 44 4444 # 1414 44144 #44 &4 444 4# 4# 444 4444 2-4 4 3, ionized radiation (X-ray, x-ray Hi 4# damage 1l44 444 4#44 44.

10 - 1. ^

4 *4 #4 4% #41 44* ri4f 444 4% 444# 4*4 #4 4444 ri4f # o]g|]^-o_g_^| ### <^thg- ^^47] ## 4e #1# 44&1 Ti 11 iM #4# 44 *44 7.]} v]s{ »s] *4# ti]ji7 7l- 7|-*44. 444 a. #^-3]-4 44 #

44 1* m*## network# 7>4 jl *4*# 4=*1# 444 zi ### 4##7l 4 44 4-8-1 444. 54 4^44 444 4444 444 44 4444 4 *# 44 4 4444 44 roadmap 4 44 *4 #4 4 4 4 44 444 4# 444 444 T^if 44&44 # #44 4# 444 *444 444 14 4# 4=*1# network4 414 4- damage &4 * 4* o) # 4-8-1 EPR dosimetry 7l * * #4 4 4# W 4 4 444. 4=*11* # 4= #* 4=#* #4 *4. 4444 #!*&4 *4*, *e*4 #44 x-ray 7} 74^44 4444 #1*&4 44#44 #114. *4*& #*114 4 4 111 44* 44144 4414* 4=* 4#* * # ## 4=44 1* *1 #4=4 1 4= #**&, 444 44 it #4-4 4444 4*44444444 NMR group 44 44* 444 4*44 444 4 4*# 447} 1*1 #*& 4444.

#41 #444

\ I &J\\ II &J\\ III B3II

Material Chemical Biological System System System

— MeV id #94 a* OIS ~ MeV 5 KDP 41, Organic/inorganic DNA tts 4 #*91 #44 CN urea, hybrid materials, (zipping/unzippin y^XI- 4S 1 4# squaric acid, supramolecular g), proteins - 4*9#4 4 71# ('02 ~ '05) SnCI2-2H20 structure (folding/unfolding) proton conductivity t|4

KDP til, urea, Energy : squaric acid, 4 MeV 3 SnCI2-2H20, -4 444 #44 1* o|S Dose : 1013~ 1023 organic/ DNA, proteins 4 4*9 It defect §9 # ions/cm 2 inorganic hybrid bond damage St ## S Size : ~ 1 um materials, supramolecular ('05 ~ '08) structure DNA, proteins, KDP til, urea, 20 ~ 60 MeV squaric acid, Ld SnCI2-2H20, - # 444 #£4 a oil 4 th organic/ bond damage Stt 444 #£X|- 4S inorganic hybrid 4459 (EPR) dosimetry ('08 ~ '12) materials, supramolecular structure Nano & Bio materials ; ^ # station 51=1

Hydrogen bond structure o|sH

>-4 in SCaleOjl A-j f ^^1 manipulation

Nano & Bio Technology f si W^j sj-M)

12 - 2.

i) 1. Tilt defect 444 1ft 44 4 5At

* KH2PO4 (KDP), TIH2PO4 (TDP) ; la 4 14111 #1 * 4=41 32 At 524 ; Energy : 0.5 - 1 MeV, Dose : 1013 - 10l0 ions/cm 3 2. 441ft defect Hi 44

* 4=47} ^4 4 #4 X-ray , 4444 e' 4 dissipation e" #4 4 4&H 44 4521 44 AC/DC proton conductivity #4 44 ^4444 (4l NMR), EPR, VSM 44 #4 Raman, IR spectroscopy, luminescence 44 4'll (KH2PO4 4 s 44, graphite, 3.4 7jMl) 11 44 44 MM ^ 4414 #4 #a

2) 4411

* 4=44 4 44 44 4&4 KH2PO4 Ml# #1, 2-4 4 MM #1, carbon nanotube #11 714 #4 4 4ft #4 14 441 MM (SCI 44 251, 4MM-4#a 151, 4MM-#la 101)

* 4=4 7j ^41 KDP, TDP 4&1 #4 444 1ft AC proton conductivity, XRD, NMR, EPR, IR, Raman 4 111 HI #4 41, #47} 141 ^44 XRD 11441 Him! 47t, VSM, EPR 11-4.41 paramagnetic impurity 441 4414 14 4 Ml 1411 (Material Research Society 2005 Fall Meeting, Boston), (ICAMD) it5. Physical Review B 4 4 Ml 141M) 14 1#: S. H. Kim, K. W. Lee, J. W. Jang, C. E, Lee, J. Y. Choi, K.-S. Lee, and J. Kim,

"Proton-beam irradiation effect in TIH2PO4" Phys. Rev. B 72, 214107 (2005).

13 - T'

36c aeo Temperature (KJ

3.4 1 TDP4 444 4^ 4444 4 s

s ait' Qeloro ?HnC Z1 Bb < C ■ U - ■ ■ £ S'1 jShlC ■ ------U------Sj IIEHIE* rl - ", 44EHr v > Kin' am mu " 4kd ' mil one"

1 3:1': After

* - 3fi3 K ^ 1 jiid M

.J 3jC :ic s am i to n z.4y.\ti nam ^.mh' z- (kn)

34 2 TDP4 444 ^4 44 4^ 4443 tijs.

14 - ■ before r Alsr i ]€'-■ / . v 10 z

1CS □ □ 1 ->'r 1C '--TrJ"L —|----- 1------1------1------1------1------1------1------1------1------r 300 330 3^0 330 300 400 Temperature (K)

zl^ 3 TDPS] ^33 til3.

15 - *11 4 5 S! &0^oto||£| 7|oiE

l. 4^ T=ME

1) 447m 4a4 443

4 a 4 4 3(%) 4 4

•KH2PO4 44 #4, 244^1, graphite 44 4 °] 3.4 4" 4 4 4444 network# 44 #4 4 4 4 4444 4 native defect0!] 4 4 4 -444 ^4 4 4344 31 4 4# 44 44 #4 #44 44-44 44 4 100% -44=4 43444 44433 444 4 44 4 44 44 4 4 444 setting 4 44 4 34 station 44 : 43444 4444 44 44, 4&4 44: 10*10 mm, 4 34 34: 4 MeV, 10la ions/cm 2 °] 44 4 44 4 34

2) Aj-7] 47} s] 4444 4# 4 a 4434 4% 44^7} 44

3@44 444 4 4 4 7}

444 4 34 4 #4 444 100% 4a 44

1250% 44 4a 44 44 SCI ^4 24 (SCI 44 254)

16 - 2. <2^4^

e| defect a m ? ef defect Ef 415 t)gt 5Au|

?J$n 3:At^ #39 7 £AIA|5.: KII^CKDP). T> l*PO» (TDP) - &5&5JI °5i J3- ■ X-ray . PRC ^ & Y?S?t ■ ^IZCHI ITk5 AC/DC proton conductivity =3 j Energy: o.t? 1 Mf*y. Doac : I O'1 - l(r» ion^cm^ ■ ^L± MtLSS (V «ilR>, FFTt VRMg=? £iJ ■Raman, Ft RpficfmRnnpy 5: El (K]-y=f)^ 1 = SS, urna, zmVPE- add, SnC^ 2H=0

JM a ^U£l

f £ ^ LH ^ sj # 71| t e 711 12 16 28 t ^ ^ S. 14 10 24 2 0 2 ^ *\ ^ ^ 0 0 0 = ^ZL2H## 0 0 0 7|#0|g-W^ 0 0 0

17 - n \ 5 g #stii si

1. ‘9T L7fliil4

1) ~MeV #43 A 4d41# # ## proton conductivity# Ml # 7] #: proton conductivity# ### 2.M| # #5_#^1 7] ##

2) Supramolecule ^ organic/inorganic hybrid system #1#-: #### network# ## supramolecular chemistry# organic/inorganic hybrid system Ml # ##![## # #5] °] #.

3) ~MeV #^7} ### ### DNA# ### DNA zipping/unzipping 0] # protein folding/unfolding 7] # ##.

4) (EPR) #tb #A34 til# dosimetry: ## ## 4 # # dosimetry# ### #######. ### #, ####, ### kl# ##, ^ #^##4 ##, ### ## # ^H# ### Ml ## 7}#.

- ### ### #4 Al# (neutron, helium, y-#) 4 ## 4# + #### - ### #Ml #### # ### 7]]zll# ## 7]# ## - 4##M1#MH 4d#Ml 4# ^## ^ ### #43 ##4 - ## 7## #4 ### Ml# 7l# # 7# 7l# ## ## - 7H## 7l## 7l ### ###### #2-ti>o]# -§-## #### ##

- 7}>H4 4 4### membrane electrolyte energy device cell Ml#

18 - n\ 6 9 3 ^7H H49

1. 3?;H44i44 4444;]*ia

• Ionized radiation biology Lf4]) - #4 4# LM] 3]# ^9- : Armed Forces Radiobiology Research Institute (AFRRI, h| N") - o]2n] ^Af# #-§-% 2^

- Mid')] "0# : ^ (4^)

• ERR dosimetry (-#-§-^4]) - Alanine/EPR therapy level reference dosimetry service : NDRL Radiation Chemistry Data Center (Notre Dame Radiation Laboratory, h| N") - Alanine-EPR Industrial dosimetry : Victoreen Inc. & Micro-Now Instrument Company Inc. M N")

• X-ray 5AH1 ft 2.41 ## damage leafy] EPR - CaSCL 41 H w E><9n z': NDRL Radiation Chemistry Data Center (h| N") - DNA^f sM# : Saarland universityS] J. Huttermann 2# (N"i=)

19 - * ii 7 s- amse

1. R. Blinc and B. Zeks, Ferroelectrics 72, 193 (1987). 2. J. Seliger, V. Zagar, R. Blinc, and V. H. Schmidt, J. Chem. Phys. 88, 3260 (1988). 3. K. Hanazawa, M. Komukae, T. Osaka, Y. Makita, M. Aral, and T. Yagi, J. Phys. Soc. Jpn. 60, 188 (1991). 4. N. Yasuda, S. Fujimoto, and T. Asano, Phys. Lett. A 76, 174 (1980). 5. Y. Oddon, A. Tranquard, and G. Ppe, Acta Crystallogr. Sect. B 35, 542 (1979). 6. R. J. Nelmes and R. N. P. Choudhary, Solid State Commun. 38, 321 (1981). 7. M. Ichikawa, K. Motida, and N. Yamada, Phys. Rev. B 36, R874 (1987). 8. M. I. McMahon, R. J. Nelmes, W. F. Kuhst, R. Dorwarth, R. O. Piltz, and Z. Tun, Nature (London) 348, 317 (1990). 9. I. V. Stasyuk, R. R. Levitskii, and A. P. Moina, Phys. Rev. B 59, 8530 (1999).

- 20 - * ii 8 g

- 21 PHYSICAL Rt/VlhW B 72. 214 107 (2005)

Proton-beam irradiation effect in TIH2PO4

S. H. Kim, K. W. Lee. J. W. Jang, and Chcol Eui Lee + Depaitment nj Physics and Institute far Nano Science, Karen University, Scon/ 1.16-713. Korea

,1. Y. Choi Department of Computer Science and Institute for Nano Science, Korea University, Seoul 136-7 IS, Korea

K.-S. Lee School of Computer-Aided Science. Inje University. Cimhae 621-749. Korea

.1. Kim ’ School of Life Sciences and Biotechnology. Korea University. Seoul 136-713. Korea (Received I November 2004; revised manuscript received 26 September 2005; published 6 December 2005)

We have investigated the hydrogen ion beam irradiation efleet on TIH1PO4 (TDP.) showing an anttferro- electric phase transition and a terroelaslic phase transition. The polycrystalline, sample was irradiated by I MeV hydrogen ion beams to a dose of H)15 ions/cm 2. The x-ray diffraction pattern shows a lattice elongation along the hydrogen bond direction. According to ‘H rotating frame nuclear magnetic resonance measurements, the activation energy increased above the anliferroelcctric phase transition after proton beam irradiation from 0.41 to 0.57 cV, indicative of hampered proton motions. The temperature-dependent ”P high resolution nuclear magnetic resonance isotropic chemical shift indicated a displacive change in the PO* tetrahedra. and the Iinewidth showed a broadening after the proton irradiation.

DOl: lO.HO.WhysRevB 72.214107 PACS number!s): 64.7(1 Kb. 76,60.-k. 77 90. + k

I. INTRODUCTION other way of modifying the hydrogen bond geometry. Most TlH2PO.| (TDP) is closely related to the KH,PO, (KDP)- mediation studies in the hydrogen-bonded ferroelectrics lype crystals, which are interesting hydrogen-bonded materi have been concentrated on the transient defects induced by als undergoing structural phase transitions accompanied by ionizing-radiations such as x-ray and ultraviolet ray (UV ray), where the defects are closely related to the optical ferroelectricity or anliferroeleclricity. In these crystals, it is properties .11"14 On the other hand, light ion beam irradiation known that protons in file double well potentials on the hy effects have rarely been studied,13 4 Nature of the transient drogen bonds are involved in a phase transition accompanied defects and their relaxation processes are fairly well studied, by displacements in the heavy atom (K,P,0) structure. One for cases of apparently very dilute defects, indicating close outstanding phenomenon is the proton-deuleron “isotope ef relation to the modification of the hydrogen bonds .11"13 The fect” that raises the transition temperature by about 100 K hydrogen bond lengths were reported to change due to the and decreases the pressure dependence of the transition tem defects, particularly the hydrogen vacancies.14 While the perature. Traditionally, the prolon-deutcron isotope effect has pressure is a macroscopic tool, the irradiation method may been explained by the tunneling model, in which the marked be used as a microscopic tool to modify the hydrogen bond. increase in the transition temperature was attributed to a At room temperature, a fairly stable defect, caused by an change in the tunneling frequency by the mass change, and ionizing radiation and observed by electron spin resonance the decrease in the transition temperature with increasing (ESR). is POjfT which can originate from the potassium or pressure was attributed to an increase in the tunneling hydrogen deficiencies.11 Proton beam irradiation was re integral .1 Recently, neutron Compton scattering experiments ported to induce the defects similar to the hydrogen vacan have shown that the protons arc self-trapped in one or the cies from x-ray irradiation, giving rise to an optical transition other of its equivalent positions, and jumped from position to in the color centers.16 A helium ion irradiation study also position through the phonon-assisted tunneling." However, revealed that hydrogen atoms arc depleted in KDP by the the tunneling model has mainly been concerned with the irradiation and that a large amount of depletion can break the proton-deuteron mass difference and does not take into ac crystal structure into an amorphous slate.1’ The suggested count the effect of the hydrogen bond geometry. mechanism of hydrogen depletion by light ion irradiation In order to vary the hydrogen bond geometry, isolope may diffei little from the stabilization mechanism of the tran substitution 14 as well as high pressure studies have been car sient defects In ionizing radiation. Hydrogen was reported to ried outs '" It is worthwhile to note that the transition tem hardly he depleted In prolon irradiation, presumably due to perature did not shift by dculeron substitution when the hy the small ion size. Since the defects due to the hydrogen drogen bond lengths were retained.’ The geometrical aspects ion irradiation ai room temperature arc reported to be related of the hydrogen bonds were considered in the proton lattice to the hydrogen vacancies, only a very small number of hy coupling model ,'8 and redistribution of the atomic positions drogens may be depleted by the irradiation. ,l’ The hydrogen hy the electric liekl-induced structural changes was bond length can be local!v modified in the hydrogen ion suggested ."11'1 Yet, the irradiation methods may provide an irradiation while retaining the crvslnl!i 11c structure, and hs-

10984)12 1 /2(H>5/7.7 2 i 72 14 10714 l'S25.(l( 1 21 - | 1 27 >5 I lie \i: 1 .Ts.sni I ’hr ova! Socict'. KIM ei ul. PHYSICAL REVIEW B 72. 2141(17 (2005) drogcn bonds with various bond lengths may be realized in a ------before sample. Therefore, the modified hydrogen bond by the irra ------alter diation can be expeeted to affect the proton ordering and the proton motion in the hydrogen bond network. TDP undergoes two major phase transitions, an anti ferro electric phase transition at 77=230 K and a ferroelastic phase transition at 7"'=357 K.IS 211 The room temperature phase is known to be paraelectric and ferroelastic while the low- temperature phase is anliferroeleetrie. The high-temperature phase is known to be paraelectric and paraclastic. TDP has a monoclinic primitive cell at room temperature with a- 1.4308 ran, 6 = 0.45 IS nm, i =0.6516 nm, and )3=9L76° 21'22 TDP has three different crystallographic ’ hy drogen bonds as determined by x-ray and neutron diffraction, and the crystal structure of TDP illustrating three inequiva lent H sites can be found in the literature.11022 The two shorter bonds, 0.243 nm and 0.245 nm, respectively, are cen- 26 (degree) trosymmctric and form zigzag chains along the c axis. Hy FIG. 1. The XRD pattern of TDP before (solid line) and after drogens of these bonds are at special positions at a center of (dotted line) the hydrogen ion irradiation. inversion and undergo an order-disorder phase transition through the phase transition temperature Tr. The longest bond, 0.25 nm, is asymmetric along the b axis and the pro shown in Fig. I. By using a conventional ‘TREOR 90 ” code, the lattice parameters were obtained from the XRD pattern, tons are at a general position both above and below Tt ,1S as shown in Table 1. The lattice parameters before the irra In this work, we have studied the hydrogen ion irradiated TDP in comparison to the virgin (unirradiated) TDP. While diation are consistent: with previous works .22 After the irra diation, the lattice constants increased along the b and c axes, proton irradiation was also done on KDP, ESR measurements in the hydrogen bond directions, with decreasing monoclinic showed that the effect of the irradiation is much greater in angles. Our results appear to be similar to the electric field TDP than in KDP. Thus TDP was chosen for extensive in vestigation in this work. In view of l he hydrogen bonds, TDP induced structural changes, in which the P atom is displaced has very short bond lengths, R = 0.243 nm and 0.245 nm. and a redistribution of the hydrogen over the two previously whereas the transition temperature 7, = 230 K is very high, equivalent sites in the hydrogen bond lakes place .910 distinct from other KDP-type ferroclcctrics. Proton irradia The rotating frame spin-lattice relaxation curve obviously tion effect in this system was also expected in the structural deviated from the simple single-exponential function, and was fitted to a stretched-exponential form below T'c, changes and proton dynamics. The main purpose of this work is to study the effect of irradiation in view of the proton M(r) = M,,exp[-(;/T,,(I) motion in the hydrogen bond networks. Above T'., the rotating frame spin-lattice relaxation curve was well fitted by a double-exponential function. The spin- II. EXPERIMENT lattice relaxations were very similar to our previous measure A polycrystalline TDP sample was irradiated with 1 MeV ments at all temperatures.23 A very fast decay with a time 10~6 hydrogen ions to a dose of 1012 ions/cm 2. The irradiated constant of order of s. which can be attributed to direct samples were kept at room temperature for several days, for relaxation to the paramagnetic centers, was also observed but the transient defects such as H and HP04" to relax into stable its fraction was too small to determine the corresponding forms, before carrying out the measurements. A BRUKER time constant. The temperature dependence of the rolating- MSL 200 spectrometer was used for the 'll NMR (nuclear framc spin-lattice relaxation rate 7^ is shown in Fig. 2. The magnetic resonance) rolating-framc spin-lattice relaxation temperature dependence of the rotating frame spin-lattice re time measurements at the frequency of the rotating frame, laxation time can be taken to consist of the dipolar contribu tu,/2ir=55.6 kHz. The rf heating of the sample wax ob tion of proton motion (7jiZ) and paramagnetic impurities served not to exceed I K. The high resolution ’’'P magic (Tlp) terms, i.e., angle spinning (MAS) NMR measurements lor TDP were 7T>7w + 7p,- Cl made at the Larmor frequency of 243 MHz at the spinning frequency of 7 kHz. The chemical shills were measured rela tive to a 21 P standard, a solution of !! .('()• TABLE 1. Monoclinic lattice parametria for TlfLPO., before and after tin- proton irradiation.

111. RESULTS AND DISCUSSION a (A) h (A) c (A) A (degree) hi the ferroelastic phase, the crystal structure of TDP be Before 1 4.3 1D v 0.0(13 4.5l3± 0 00l G 5 13 ±0.00 1 01 .79 ±0.02 longs to the monoclinic space group P2,/u.21,22 After the After 14.3 15 + 0.003 4.320 ±0.0(11 6.523 + 0.001 91 .73 ±0.02 irradiation, the XRD pattern shifted to smaller angles as PROTON-BEAM IRRADIATION EFFECT IN T1ELP04 PHYSICAL REVIEW B 72. 214107 (2005)

t r1 T

c After

u □□

lEfTTO 43-13-“ ■ Before n After 1TG. 3. Temperature dependence ot' the proton motion contribu tion (77,/) obtained by subtracting the. paramagnetic impurity con 2.5 3.0 3.5 4.0 4.5 5.0 tribution (77') from the spin-lattice relaxation rates in Fig. 2 before 1000/T(103K"1) (solid symbol) and after (open symbol) the irradiation (in a linear scale!. FIG. 2. Temperature dependence of the rotating-frame spin- lattice relaxation rates in TDP before (solid symbol I and after (open After the irradiation, the activation energy for the proton symbol) the irradiation. The data were obtained by the slretched- motion shows a marked increase, which may severely hinder exponenlial fit below T'r and by the double-exponential fit above T'r. the hydrogen motions. Using the data and the activation en 7), and T’. indicate the antiferroelectric and the feiToelaslic phase ergy Ea obtained from the Arrhenius tits, the correlation time transition temperatures, respectively. The solid lines are the Arrhen was obtained. After irradiation, the correlation time at 300 K, ius fits, and the horizontal dashed lines indicate the temperature t=5.02X 10“’ s, is much longer than that of 1.74 X UP3 s independent paramagnetic impurity contribution below 77. Inset, before the irradiation, reflecting the slowed proton interbond the exponent n of the stretched-exponential til. jumping motions as well as slowed hindered rotational mo tion of the H/PO4 groups .25 While the activation energy for Before and after the irradiation, the rotating frame spin- H,PO.( rotation can be controlled by the cation size,24 our lattice relaxation rate shows a sharp divergence correspond work shows (hat it can also be controlled by the proton beam ing to a critical fluctuation near the antiferroeleclric transi treatment. tion temperature Tc. In the low temperature anti ferroelectric In an effort to further identify the effect of proton beam phase, 77p, showing a temperature independence which can irradiation on the TDP lattice, we performed the lngh- be ascribed to the paramagnetic impurity contribution, be resolulion 31P NMR. Figure 4 shows the temperature depen comes much larger after the irradiation. The ratio in the mag dence of the 31P NMR isotropic chemical shift. After the nitudes of ('/YpXhcr/f'^Xcfore 7 bclow may be attributed to the paramagnetic density introduced by the irradiation. The nuclear spin diffusion toward the paramagnetic centers gives rise to an increase in the spin-lattice relaxation rate, whose magnitude is determined by the concentration of the paramagnetic centers.24 The 77,shows an anomaly at the antiferroelectric phase transition temperature 7,., above which it increases with temperature indicating activated pro ton motions. In order to obtain a clearer picture of the relax ation rates due to the proton motion in the intermediate tem perature range before and after irradiation, in Fig. .3, the data in Fig. 2 is shown in a linear scale after subtracting the paramagnetic contribution (the horizontal dashed lines in Fig. 2). Thus, it is found that below 7, proton motions are not activated enough to contribute to the relaxation rate, the paramagnetic impurities accounting for most of the relax ation. whereas above it they are activated enough to make a 150 200 250 300 350 400 significant contribution to the relaxation rale. Temperature (K) As T’ is approached from below. 17' shows an .Arrhenius type of temperature dependence. 77 , ~ r_1 = r 4 expf-Tg, / FIG. 4. Temperature dependence of the high-resolution "‘P kn T). The straight lines in Fig. 2 correspond to the /Arrhenius NMR isotropic chemical shift in TDP before and after the proton- fils. The activation energy before and after the irradiation beam irradiation. T and 7( indicate the antiferroelectric and the was obtained to be A. = 0.41 eV and 0.57 cV. respectively. fern-elastic phase transition temperatures, respectively. KIM el ai PHYSICAL REVIEW B 72. 214107 (2005)

proton activation energy, as the H2P04 tetrahedral reorienta- ■ Before tional motion and the interbond proton motion are known to o After 700- lake place synchronously .21 Thus, the proton motion giving rise to the rotating frame relaxation can be identified to be 600- the interbond proton jumping motions as previously TTTT observed .21'26 £ 500- In summary, we have studied the effect of the hydrogen ion irradiation on a hydrogen-bonded ferroelectric system, ® 400- TIH1PO4, before and after the proton irradiation. The x-ray diffraction pattern revealed lattice elongation along the b and 300- c axes, in the hydrogen bond directions. The *H rotating- frume nuclear magnetic resonance showed an increase from 0.41 to 0.57 eV in the activation energy of the proton mo 150 200 250 300 350 400 tion. induced by the proton irradiation. Besides, 3IP high Temperature (K) resolution NMR measurements indicated that proton irradia tion gives rise to a displacive change in the P()4 tetrahedra FIG. 5. Temperature dependence of the higli-resolution J,P linked with the hydrogen bonds. NMR linewidth in TDP before and after the proton-beam irradia tion. T,. and T[. indicate the antiferroelectric and the ferroelastie phase transition temperatures, respectively. ACKNOWLEDGMENTS proton irradiation, the isotropic chemical shift moved toward low frequencies below the ferroelastie phase transition tem This work was supported by the Korea Science and Engi perature T[. the difference in the isotropic chemical shift neering Foundation (RO1-2005-000-10798-0 and Proton before and after the proton irradiation being Atrisn~- 16 Hz Accelerator User Program No. M202AK0I002I-04A1101- below the ferroelastie temperature. The change in the isotro 02110) and by the Korea Research Foundation (Grant No. pic chemical shift for the 3IP nucleus indicates displacive KRF-2004-005-C00060/D00057 and Brain Korea 21 Project change in the structure. Besides, the increase in the ’'P NMR in 2004). The authors thank Dr. H-J. Woo at the KIGAM linewidth after the proton irradiation over the entire tempera for the proton-beam irradiation. The measurements at the ture range, shown in Fig. 5, can be associated with that in the Korean Basic Science institute (KBS1) are acknowledged.

Corresponding author. Electronic address: rsce!<7korea.ac.kr l4C. S. Liu, N. Kioussis. S. G. Demos, and H. B. Radousky, Phys. 'Electronic address: [email protected] Rev. Lett. 91. 015505 (2003). *R. Blinc and B. Zcks. Ferroelectrics 72. 193 (1987) |ST. Som. M. S. Navati, and V. N. Kulkarni, Nuel. Instrum. Meth 2G. F. Reiter. .1. Mayers, and P. Platznian. Rhys. Rev. Lett. 89. ods Phys. Res. B 179, 551 (2001). 135505 (2002). U'V. T. Kuanyshev, T. A. Belykh. I. N. Ogorodnikov, B. V. Shulgin, 3S. Tanaka, Phys. Rev. B 42. 10488 (1990). M. K. Satyhaldieva. and M. M. Kidibaev. Radiat. Meas. 33, 503 "’S. Koval, J. Kohanoff, R. L. Migoni, and E. Tosatti. Phys. Rev. (2001). Lett. 89. 187602 (2002). I7S O. Kucheyev and T. E. Fetter. J. Appl. Phys. 95. 8475 (2004). 5M. I. McMahon. R. J. Nelmes, W. F. Ktihst. R. Dorxvarlh, R. O. 'Cl. Seliger, V, Zagar. R. Blinc. and V. H. Schmidt, J. Chem. Phys. Piltz, and Z. Tun. Nature (London) 348. 317 (1990). 88.3260(19888 61. V. Stasvuk. R. R Levitskii. and A. P. Moina. Phys. Rev. B 59. K. Hana/awa. M. Komukae. T. Osaka. Y. Makita, M. Arai. and T. 8530(1999). 19 7 A. Bussmann-Holder and K. PI. Michel. Phys. Rev. Lett. 80. 2173 Yagi. J. Phys. Soc. Jpn 60. 188 (1991). (1998). JIN. Yasuda, S. Fujimolo. and T. Asano. Phys. Lett. 76A, 174 8 N. Dalai. A. Kiymaehyov. and A. Bussmann-Holder. Phys. Rex 11980) Lett. 81. 5924 (1998). 2 Y. Oddon. A. Tvanquanl. and G. Pepe. Acta Crvstallogr., Sect. B: vS. J. van Reeuwijk. A. Puig-Molina, and H. Graafsmu. Phys Rex. Struct. Crystallogr. Cry si. Chem. B35. 542 (1979). B 62, 6192 (2000). 22 R. J Nelmes and R. N. P Choudhary. Solid State Commun. 38. 10S. .1. van Reeuwijk. A. Puig-Molina. and H. Graafsma. Phys Rex. 321 119811 B 64. 134105 (2001! 21C. H. Lee. K. W. Lee. C E Lee. and K. S. Lee. Phys. Rev. B 55. 11 S. D. Setzler. K. T. Stevens. I. E. Halliburton. M. Van. N P. I 1088 11997); C. E. Lee. C. I I. I.ee. J. H. Kim. and K. S. Lee. Zaitseva, and J. J. DeYoreo. Phys. Rev. B 57. 2645 ( 19981. Pins. Rev. Leu. 75. 3309 1.1995). I2N. Y. Garces, K. T. Stex'ens, L. E. Halliburton, S G. Demos. H. -4A. Ahragam, The Principles of Nuclear Maauctbm (Oxford Uni 8 . Radousky, and N. 1*. Zaitseva. J. Appl. Phys. 89. 47 (20011. versity Press, New York, 1983). p. 378. 13 M. M. Chin la. N. Y. Garces. L E Hallihuvlon. S. G. Demos. T. 2?M. Sharon and A. K. Kalia. J. Solid State Chem. 21. 171 (1977). A. Land, and H. B. Radousky. .1. Appl. Phys. 94. 6450 (20031. :,’R. Blinc and J. Pit's. J. Chem Phys. 54. 1535 (1971). Available online at www.sciencedirect.com SCIENCE DIR E CT * solid state communications ELSEVIER Solid State Communications 136 (2005) 63-66 www.el sev i e r. eom/t oc a te/ssc

Quasi-persistent photocurrent in the UV-irradiated MEH-PPV conjugated polymer

Kyu Won Lee, Kyu Huyn Mo, In-mook Kim, Cheol Eui Lee*

Department of Physics, Korea University, Anam-dortg, Sunghuk-ku, Seoul 1)6-7Id, South Korea

Received 18 May 2005; received in revised form 6 July 2005; accepted 7 July 2005 by A. Pinczuk Available online 25 July 2005

Abstract

We report the quasi-persistent photocurrent in the MEH-PPV conjugated polymer, induced by UV-irradiation in air. It is attributed to the irradiation-induced defects, which also act to accelerate its decay. © 2005 Elsevier Ltd. All rights reserved.

PACS. 78.70.-g; 78.55.Kz; 78 66.Qn

Keywords: A. Polymers, elastomers, and plastics; D. Photoconductivity and photovoltaic;

1. Introduction molecules such as iodine, C6(l. or fC=0 groups [7-11]. Positive charge carriers are known to be responsible for the There has been a great deal of interest in the light- enhanced photocurrents. While some workers suggested that emitting diodes (LEDs) based on the conjugated polymers the photoinduced charge carriers are produced in the because of their potential applicability to large-area flat dissociation of excitons (or polaron pairs) by the acceptors panel displays operating at a relatively low voltage, along [9.11.12], other authors suggested that the ultrafast electron with the promise of low cost and easy fabrication [1 ]. The transfer to the acceptors, leading to a charge separation, is poly(phenylenevinylene) (PPV) conjugated polymer, which responsible for the enhanced photocurrent [7.8,1.7], is one of the most studied polymers in the PLEDs (polymer Quasi-persistent photocurrent, or a very slow non LEDs) due to its excellent luminescent and mechanical exponential relaxation of the photoinduced current, was properties [2.3J, has been known as a hole-transport observed in PPV and some of its derivatives [I J.14J. The material, with its hole mobility being one to three orders ongoing debate whether an energy band model or an exciton of magnitude higher than its electron mobility [4,5]. Thin model is appropriate for these polymers has been extended film devices made with poly[2-methoxy-5-(2-ethylhexy- to the origin of the quasi-persistent photocurrent 11 1,14). A loxy)-p-phenylenevinylene] (MEH-PPV). one of the PPV slow dispersive diffusion of the photoinduced bipolaron derivatives, were found to be dual function polymer according to the band model [ 14] as well as a slow devices, i.e. light-emitting and photodetecting diodes [6], recombination of the positive charge carriers with defects Photoconductivity of the polymers is known to he according to the exciton model [ 1 1 | were suggested as an significantly enhanced when they are doped with acceptor origin of the quasi persistent photocurrent. In the AL/MEH- PPV/1TO thin film devices, the photocurrent was enhanced upon exposure to air, but it showed a decrease under forward * Corresponding author. Tel.: +82 2 3290 3098; fax: L 82 2 027 bias and illumination through the 1TO [ !5|. Accordingly, it 3292. was discussed that the exposure to air gives rise to an E-mail address: rsccl(O'korca ac.kr (C.E. Led. increased photocurrent by (he light absorbed at the Al

0038-1098/$ - see front matter <3 2005 Elsevier Ltd. All rights reserved. doi:I0.1016/j.ssc. 2005.07.006 64 K.W. Lee el at. / Solid Stale Communications 136 (2005) 63-66 electrode and that the enhanced dissociation of excitons by 3. Results and discussion exposure to air is the origin of the enhanced photogeneration of the charge carriers [15]. Contrasting to PPV and some of The photocurrent and its decay showed some sample its derivatives, quasi-persistent photocurrent has not been dependence but the overall behavior showed a similar observed in MEH-PPV. In this work, new results of the dependence on the UV-irradiation time. Fig. 2(a) shows a quasi-persistent photocurrent in a UV-irradiated AL/MEH- representative photocurrent response to switching the light PPV/ITO thin film are reported. on and off. As was reported in a previous work [15], the photocurrent decreased by about 30% after about a 400-s UV-irradiation. After a successive 1800-s UV-irradiation the photocurrent did not change, just becoming noisy. While 2. Experiment the rise of the photocurrent was completed at the instant (within 1 s) of switching the light on regardless of the UV- The MEH-PPV conjugated polymers were purchased irradiation time, the decay of the photocurrent after from ALDRICH. The solution of MEH-PPV dissolved in switching the light off slowed down with increasing UV- chlorobenzene was spin-coated onto the ITO glasses of irradiation time. The instantaneous rise and slow decay of 1 cm X 1 cm to a thickness of about 400 nm. In order to the photocurrent in MEH-PPV is in sharp contrast to the avoid oxygen contamination prior to the UV-irradiation, so slow rise and decay in PPV and some of its derivatives [11, that the photo-oxidation effect alone can be separated, argon 14]. While the decay of the photocurrent in UV-irradiated gas was injected into the spin coater. The current was MEH-PPV may not be as marked as in PPV and some of its measured using a Keithley 2400 source meter at 1-s derivatives, it may still be called a quasi-persistent intervals. A 300-W xenon lamp with a UV-filter was used photocurrent. as the UV source. The photocurrent was measured under a Fig. 2(b) shows the photocurrent decay obtained from 3-V forward bias and UV-illumination through ITO in air. Fig. 2(a) just after switching light-off. The decay curve was The films were kept in the dark before the measurements that were carried out at room temperature. The photo luminescence (PL) spectrum was obtained one day after a 30-min UV-irradiation in air. The PL intensity decreased by about a half and the spectrum was blue-shifted by about 7 nm as shown in Fig. I. The UV-irradiation-induced defects are presumably the carbonyl groups acting as the PL quenchers and breaking the Tc-conjugation [ 10].

----- Before ...... After (x 1.85)

Wavelength (nm) Fig. 2. la) A representative photocurrent response to switching the Fig. 1. Photoluminescence spec Ira before (solid line) and after light on and off as a function of time, (b) Photocurrent decay just (dotted line) a 30-min UV-irradiation. The dotted line was after switching the light off. The solid lines correspond to measured multiplied by a factor of 1.85. currents and the dotted lines are fits to Eq. (I). K. IV. Lee et al. / Solid Slate Communications 1S6 (2005) 63-66 65

obviously not a single-exponential function. In PPV, a total irradiation time. The numerical symbols correspond to stretched exponential form of the photocurrent decay was the irradiation sequence. The numerical symbols of 1-4, 14, described according to a band model, but a logarithmic and 15 correspond to the transient irradiation time of 100 s. decay form was also noted to be consistent with the According to Fig. 3, it is quite clear that the parameters are experimental data in the long-time region [14]. In a PPV principally dependent on the total irradiation time rather derivative, the photocurrent decay was fitted by a double than on the transient irradiation time, indicating that the exponential function 111 |. The form of the photocurrent photo-induced defects do not simply disappear from the decay changes according to the time scale. The photocurrent polymer during the dark periods. The numerical symbols of decay in MEH-PPV consists of at least two time scales, with 14 and 15 with the transient irradiation time of 100 s, a transient decay within 1 s and a quasi-persistent irradiated 3 h after the total irradiation time of 5850 s, photocurrent in the long-time region. The quasi-persistent presumably with sufficient sample cooling period, show photocurrent follows a power-law decay in the long-time slightly decreased photocurrents, suggesting that our results region. The photocurrent decay in the long-time region was may be quite temperature-dependent. In fact, the time-scale fitted by of the quasi-persistent photocurrent indicates that the UV- irradiation-induced defects, presumably the carbonyl I = ar" (1) groups, must act as very deep traps. This would make the charge transport strongly depend on temperature, so that the where / and t are the current and time, respectively. UV heating effect due to a long-time irradiation cannot be Fig. 3 shows the parameters a and n in Eq. (I) obtained completely ruled out. from the fits to the data as a function of the transient In the inset of Fig. 3(a), the current measured 1 s after irradiation time, or the irradiation time since the last dark switching the light off shows a similar irradiation-time period, the insets showing the same data as a function of the dependence to that of the current amplitude a, which is the current at 1 s obtained from the long-time fit of the data. A (a) 12 13 smaller value of a indicates a more dominant transient decay 11 within 1 s after the switching light off. The value a 10 10 : 1 \i increasing with the UV-irradiation time indicates that the 9 UV irradiation renders a stronger decay of the quasi- 8 Total irradiation time (s) 7 100 1000 persistent photocurrent. As a larger value of the power n 6 34 indicates a faster decay, the power n increasing with the UV- 5 irradiation time indicates that the UV-irradiation gives rise 1 to a faster decay of the quasi-persistent photocurrent. Thus, 1(7- 2 the UV-irradiation-induced defects give rise to a dominant 1 and faster decay of the quasi-persistent photocurrent. The quasi-persistent photocurrent in the UV-irradiated

100 1000 MEH-PPV was observed to be affected by the film thickness, and showed some change in the MEH-PPV (b) films kept in vacuum for a long time (several weeks). As the 12 ” defects in MEH-PPV may be affected by several conditions, 0.6- further systematic studies are being carried out in order to 14 elucidate the dynamics of the photogenerated charge 9 15 8 Total irradiation time (s) carriers. Still, our work has unambiguously demonstrated 0.5- 7 100 1000 that the occurrence and the decay of the quasi-persistent 6 0.7 photocurrent in MEH-PPV strongly depends on the presence 0.6 0.4- 5 of the UV irradiation-induced defects. 4 0.5 In summary, we have studied the photocurrent response I .■ in the UV-irradiated MEH-PPV. The photo-induced defects 0.1 0.3- 1 give rise to a quasi persistent photocurrent as well as to its 0 3 - accelerated decay. too 1000 Transient irradiation time (s) Acknowledgements Fig. 3. (a) The amplitude a and (b) the power n as a function of transient irradiation time. Inset shows the same data as a function of total irradiation time. The open symbols in the inset of (a) are This work was supported by the Ministry of Science and currents measured at 1 s after switching the light off. The numerical Technology (National Research Laboratory and Proton symbols correspond to the irradiation sequence and errors are Accelerator User Program No. M202AK010021-04A1101- comparable to the symbol size. 02 110) and by the Korea Research Foundation (Grant No. 66 K.W. Lee el al. / Solid Stale Communications 136 (2005) 63-66

KRF-2004-005-C00060 and Brain Korea 21 Project in [6] G. Yu, C. Zhang, A.J. Heeger, Appl. Phys. Lett. 64 (1994) 2005). The measurements at the Korean Basic Science 1540. Institute (KBSI) are acknowledged. [7] C.H. Lee, G. Yu, D. Moses, K. Pakbaz, C. Zhang. N.S. Sariciftci, A.J. Heeger, F. Wudl, Phys. Rev. B 48 (1993) 15425. [8] L. Smilowitz, N.S. Sariciftci, R. Wu, C. Gettinger, A.J. Heeger, F. Wudl, Phys. Rev. B 47 (1993) 13835. References [9] H. Antoniadis. L.J. Rothberg, F. Papadimitrakopoulos, M. Yan, M.E. Galvin, M.A. Abkowitz, Phys. Rev. B 50 (1994) 14911. [1] J.H. Burroughes, D.D.C. Bradley, A.R. Brown, R.N. Marks, 110] M. Yan, L.J. Rothberg, F. Papadimitrakopoulos, M.E. Galvin, K. Mackay, R.H. Friend, P.L. Burns, A.B. Holmes, Nature 347 T.M. Miller, Phys. Rev. Lett. 73 (1994) 744. (1990) 539. [11] E. Frankevich, A. Zakhidov, K. Yoshino, Y. Maruyama, [2] R.H. Friend, R.W. Gymer, A.B. Holmes, J.H. Burroughes, K. Yakushi, Phys. Rev. B 53 (1996) 4498. R.N. Marks, C. Taliani, D.D.C. Bradley, D A. Dos Santos, [12] E. Hendry, M. Koeberg, J.M. Schins, L.D.A. Siebbeles, J.L Bredas, M. Logdlund, W.R. Logdlund, Nature 397 (1999) M. Bonn, Phys. Rev. B 70 (2004) 033202. 121. [13] D. Moses, J. Wang, G. Yu, A.J. Heeger, Phys. Rev. Lett. 80 [3] P.K.H. Ho, J.S. Kim, J.H. Burroughes, H. Broker, S.F.Y. Li, (1998) 2685. T.M. Brown, F. Cacialli, R.H. Friend, Nature 404 (2000) 481. [14] C.H. Lee, G. Yu, A.J. Heeger, Phys. Rev. B 47 (1993) 15543. [4] H.C.F. Martens, J.N. Huiberts, P.W.M. Blom, Appl. Phys. [15] M.G. Harrison, J. Griiner, G.C.W. Spencer, Phys. Rev. B 55 Lett. 77 (2000) 1852. (1997) 7931. [5] B.K. Crone, I.H. Campbell, P S. Davids, D.L. Smith, Appl. Phys. Lett. 73 (1998) 3162. JOURNAL OF APPLIED PHYSICS 98. 074316 (2005)

Hydrogen storage capacity of different carbon nanostructures in ambient conditions Jae Won Jang and Cheol Eul Lee31 Department of Physics and Institute Jor Nano Science, Korea University. Stand 136-713. Korea Chan lek Oh Department of Physics, Yonsei University. Seoul 120-749. Korea Cheol Jin Lee Department of Nanotechnology, Hanyang University. Seoul I.13-791, Korea (Received 14 April 2005: accepted 24 August 2005; published online 14 October 2005,1 The hydrogen storage capacity of bamboo-shaped mulliwalled carbon nanotubes (BS-MWNTs) in ambient conditions was studied by means of the volumetric method, and compared with those of single-walled carbon nanotubes (SWNTs) and mulliwalled carbon nanotubes (MWNTs). The BS-MWNTs, whose herringbonelike structure was characterized by the transmission electron microscopy, showed the greatest hydrogen storage capacity with about 0.4 wt % at atmospheric pressure. The SWNTs showed a comparable hydrogen storage capacity, whereas the MW NTs were insensitive to the pressure change. Our work indicates that the herringbone carbon nunotube structure is more capable of hydrogen storage than the herringbone graphite nanofiber structure. © 2005 American Institute of Physics. [DOI: 10.1063/1.2076433]

I. INTRODUCTION MWNTs with open-ended graphene sheets, also implying the effectiveness of the herringbone structures for hydrogen stor Because of the scientific, technological, and economical age. In this study, we measured the hydrogen storage capac potentials, hydrogen storage in carbon nanostructures such as ity for bamboo-shaped mulliwalled carbon nanotubes (BS- nanotubes and nanofibers has attracted considerable atten MWNTs) as well as SWNTs and MWNTs, using the tion. Hydrogen storage capacity of more than 6.5 wt % volumetric method. The BS-MWNTs are named for the bam (63 kg Hj/nv1') is needed for car applications ,1 which has boolike compartment structure of the inner tubes, resembling been a goal in the research of hydrogen storage using the the herringbone structures. As the walls of BS-MWNTs at carbon nanostructures. Hydrogen storage in carbon nano the compartments have "open-ended" graphene sheets, the structures has been reported by various workers. In 1997, BS-MWNTs with more compartment structures can be ex Dillon et aid reported excellent hydrogen storage of single- pected to possess a better hydrogen storage capacity. walled carbon nanotubes (SWNTs), which was estimated to range between 5 and 10 wt %. Chambers et alf reported the hydrogen storage abilities of various carbon nanostructures: II. EXPERIMENT II. 26 wt % for tubular graphite nanofibers (GNFs), For a comparative study of the hydrogen absorption ca 67.55 wt % for herringbone GNFs, 53.68 wt % for platelet pacity, three different kinds of CNT samples were prepared. GNFs, and 4.52 wt % for graphites. In contrast to these re The SWNTs (ARC-SWNT-AP, IIjin Nanotech) and the markable values, much lower hydrogen storage abilities of MWNTs (ARC-MWNT-IC, Iljin Nanotech), with diameters the carbon nanostructures, such as SWNTs, mulliwalled car of ~ 1.4 and ~ 10 nm, respectively, were synthesized by the bon nanotubes (MWNTs) and GNFs, have also been arc-discharge method. On the other hand, the BS-MWNTs reported. 4'* A possible reason for the discrepancies is the were synthesized by the vapor-phase growth. 10 Figure 1(a) different hydrogen storage measurement methods. The shows a transmission electron microscopy (TEM) image of temperature-programmed desorption (TPD) method,' the hot the BS-MWNTs. manifesting herringbonelike compartment extraction method/ ’ the volumetric method, ' and the gravi structures, with a diameter of — 100 nm. Our previous study metric method* are generally applied to study the hydrogen revealed the open-ended graphene sheets near the compart storage capacity. Of these, the volumetric method is suitable ments on the outer wall of BS-MWNTs.11 In Fig. 1(b) is a for room-temperature measurements/’ making it important high-resolution TEM image manifesting the open-ended (dis from the viewpoint of practical applications. Tor practical continuous) graphene sheets and herringbonelike structure of applications, hydrogen storage in ambient conditions would a typical BS-MWNT compartment. be very important, which motivated the present work. H\drogen adsorption at room temperature was achieved According to Chambers ct ah.' the herringbone carbon by a specially built apparatus, schematically shown in Fig. 2. fibers show the best hydrogen storage capacity, and Chen <•/ Hydrogen adsorption was evaluated by the following proce a!9 reported a hydrogen storage capacity of 13 wt N in the dure. The CN IS were filled in an open-top glass tube (with an inner diameter of 2 mm and length of 8 mm), and a glass -lT;lccti"nnic mini: rscclO'-korca sc kr tube filled with ("NTs was placed in the sample chamber

0021-8979/2005/98I7'/074316#/S22 50 98. OTasig.i vi 2005 American Institute of Physics .TM ..a I * K* H i f'l-w »•;. j/mv

pi>i h .o •'« . >_

1.( n:r 4V: *c» >.i: inn sv:i r ' 1 • al / X'.VK n i'T' i *;<

4/10 ^ 1' ;/i:1 MWf' f \ i: ' ... • x kW.VMIX i (i- 1 ?/n' C \ 1 '

■i xio-.v; i'k vi 'v •»! Iiyili^xr iir.f •; i:i.lx: i:* I:.;: Iiy:l t'nirv u ;:.;miIv I..** hyJiouen •.inid;vs » i. ur n r- \ w I, - *.* i 'A, w.ml „c ^;iriuroj «.e me hr .môi:-: <:n:r.nes. Ik l:rlr.- .*'V: Vt. : np.i: iv .»* - .sVV\ I’ = : î| I. V'- ~ .^:il ; i***5vili Hi. imiK: Vvi :miik : f Miv6 «.:•> xmn.lc. I.lz.i! th;.i v; six- 3S-MVV.NI5. -.nil (: >u\»i u; .'.n' H.'.f-. 4i4»!xw6 ,h;:v :ii.l h:r h .c,vsii At sikv ..:: f. .IS I Jin . ! v.m v.uiK:i. L"u . xkvyvii *'ord^c c :;:ju :y i:l ikv •>uv\\iiv:# .SV/.VP MVa:>' ;.-.fvoi'y :r' $WN7s >ysx:n ^ vvIJy vjûnu dc.^.'sed r.i •0"? >ir for l i \b 1 I .'ml X .VJW.N h i.lSV vC- we. I >Hfc 0l;iC: -vv. ' * K:rtl, hydv.gcii was vI:.-.iki:u in ilr^ ujmiK'ii :Mily. j j .lx Tftv hv.-liuêi' S.O's.LV -.hv*s .1 line* II /rvdfx- kli )u*ô:C pr.v.virr i w;i% ."lie ::nin:iii nl hydhn^n >: :i: "i SXVVT-. -ikl tiS-MWV . ....Ik *:..■• I K >jV/N ;X mCw -.i: r i? iwvni-. „ vn* n.*n ok:n;r .',1 rv.m iw it r on HfiVitx: jv;|y u* 'v ir Xi; %xlk v. i^v liix %vrk i.) villi ii- cx|:crii:;caiiji . ini' A I ix.i iikx .im. uf >.y i n • nr:iy.< v/iii: pmû.v i:; v ^ p. exs vi piv/.f j:vv * xi? '/* i$ rkc v<;\ ivt of :h > r.-.yif voi r, 7 rhc mniwinie •:* : MK.l A<; nr.-lln^ 10 >inn nic hyrlroêii ?»:•»• •jic svmciu j.il X :l>: >^:.l-^D;sCv i<.enL in riic i*ext *ic.\ me i»n.O •* i iL'fx* \ iv < f'.HAIs: I /: i :•;« :«l'v' V := MHn I , -v i- na.uylC v wOs evened ..i d :l ~ f:C£U f? P? •**& ilCOt.i v«l. col. V.viC hvXs^s I: <10;ec n LN A I k iKU.be: -i v <.-ik %:> ui . . • -.2:11v -JcciXL'C^ u.-.-J.'-x: l':iv ;i;.:iv.kr iixrvjtxx'4 irJivinin; I rim ;h< v.l;i.i«ir M:ii k %/," Kxu:i ii-v.tf: h: C'NTs ;.< n l,Li:i :#.% liy •iiO^.'i sici*.;ce Tu > Sxvn Tx cm ;« • v ir:-:Jc :nom hyj n- yvr. Ii . i Mw.VI:i. Oi. :lx :X ,%i .lOnv, K:S MWNJV ru%>;-%<. iv«* '.vci*. .ihl* ll* : x.* l:y:l:*::v*ji £..< »s. ylvnriu^- wi:l~ '.v.v^r I ;in.ci-r^ :h-.r I m e m MWN^:: mi < n.l Mix ihv • i ::1 ml ;i r/1 ivl* x«r<^>N IK' SI iKI'i'f I :K V4 V-;;v^ I) i f )iVLXH - :C. / C ' :k i.iriCftt*: III. HtrSULTS AND DISCUSSION cl I ydiOuCii v.tuaJC is uunv ii.xiim ::vc i:: .i.Cv .jil*. N wv -.an zii .kv .n ^.>s-s-: i vin ;?i !:VLfi.:»n s.v:- .'iil- ui :iujiAS i>l Ikx Iiv^Ci. it :Uv C.NJ samples pi: :v.p;ii:ilitx r in Tattle I. fmrit wh ;:V. Ih>. Ir.xlmirn slnm;:% c/p.iriiy '.v.v- c^’i:i.l/.k.l. Mk-.rv hl«i ./ c

CO D? 0 I Cfi 00 •2 Fit*:' 0(a*rl

’ lu s.l»: i :* .i.r c;i. I il». .«?•' u c:l !• I .’c V. :l*n' : II. i i • >»• ruv . i| . • .ii i ; - . Xin- *' •: .zt:i

parison with that of SWNTs. The ratio of the storage capacity 2005). One of the authors (C. J. L.) was supported by the of BS-MWNTs to that of SWNTs is shown to be 1.1 in this Center for Nanotubes and Nanoslruclurcd Composites at work. In fact, while the herringbone!ike CNTs have been Sungkyunkwan University. reported to store more hydrogen than SWNTs do at room temperature, 1’ it has been reported that the herringbone GNFs M. S. Dresselhaus. K. A. Williams, and P C. Lklund. MRS Bull. 24, 45 store less hydrogen than SWNTs do. with storage capacity I'lWU I. ratios of 0.57 (Ref. 5) or 0.33 (Ref. 17). Therefore, we can ‘A. C. Dillon. K. M. Junes, T. A. Bckkedahl, C. H. Kiang. D. S. Bethune, and M J. Kaleen. Nature (London) 386. 377 (1997), conclude that the herringbone CNT structure is better for 'A. Chambers, C. Park. R. Terry. K. Baker, and N. M. Rodriguez, J. Rhys. hydrogen storage than the herringbone GNP structure, which Chem. B 102,4253(1998). can be explained by the fact that the herringbone GNFs are JM. Hirscher cm!.. Appl. Phys. A: Malvr. Sci. Process. 72, 129 (2001). devoid of inner room in contrast to the case of the BS- lM. Ruse he!, M. U hlemann, O GucHeisch. A. Leonhnrdt. A Graff, Ch. MWNTs. Taschner. and J. Link, Appl. Phy.s. Lull. 80, 2985 (2002). °H. Kajiura. S. Tsutsui, K. Kadono, M. Kakuia. M. Ala, and Y. Murakami, In summary, we have investigated the ambient-condition Appl. Phys. I^u. 82, 1105 (2003). hydrogen storage capacity of BS-MWNTs and compared 'C. Uu, Q. H Yang, Y. Tong. II. T. Cong, and II. M. Cheng. Appl. Pliys. with those of SWNTs and MWNTs. BS-MWNTs, with the Lett. 80, 23X9 (2002). herringbone structures with open-ended graphene sheets at T). I-. Quinn, Carbon 4(1. 2767 120021. the compartments as revealed by the transmission electron l/Y. Chen. D. T. Shaw, X. D. Bui. E. G. Wang, (’. Lund, W. M. Lu. and D. D. L. Chung. Appl. Phys. Letl. 78. 2128 (2001). microscopy, were shown to possess a hydrogen storage ca lUC. J. Lee. S. C. Lvu, H. W. Kim, J. H. Lee. and K. L Cho, Chem. Phvs. pacity slightly greater than that of SWNTs, and considerably Leu. 359, 115 (2002). greater than that of the herringbone GNFs. "c. J. Lee and J. Park. Appl. Pliys. Lett. 77. 3307 (20001. '-G. G. Tibbetts. G. P. Meisner. and C. H. Oik, Carbon 39. 2291 (2001). ACKNOWLEDGMENTS 1 'S. M. Lee el ,il. S)inh. Met. 113. 209 (2000). mA. Ziitlel. P. Sudan. Ph. Mauron. T. Kiyobayashi, Ch Emmenegger. and I. This work was supported by the Korea Science and En Schlapbach. Ini. J. Hydrogen Energy 27, 203 (2002). gineering Foundation [Proton Accelerator User Program (No. l'M. Rzcpka, P. Lamp, and M. A. de la Casa-Lillo, J Phys. Chem. B 102, 10894 0998). M202AK010021-04A1101-02110) and RO1-2005-000- lo Q. Wang and J. K. Johnson, .1. Chem. Phy.s. 110. 577 (1999). 10798-0] and the Korea Research Foundation (Grant No. 1 'H. G. Seim me l, G. J. kearley. M. G. Nijkamp. C. T. Visscr, K. P. de Jong, KRF-2004-005-C00060 and Brain Korea 21 Project in und E. M. Mulder. Chem.-Cur. J. 9, 4764 (2003). Available online at www.sciencedirect.com

SCIENCE solid state communications ELSEVIER Solid State Communications 135 (2005) 683-686 www.clsevier.com/locate/ssc

Mechanical cutting of bamboo-shaped multiwalled carbon nanotubes by an atomic force microscope tip

Jae Won Janga, Cheol Eui Lee3’*, Cheol Jin Leeb

‘Department of Physics and Institute for Nano Science, Korea University, ! Anam-dong, Sungbuk-ku, Seoul 136-713, South Korea h Department of Nanotechnology, Hanyang University, Seoul 133-791, South Korea

Received 10 June 2005; accepted 10 June 2005 by A. Pinczuk Available online 5 July 2005

Abstract Mechanical cutting of the bamboo-shaped multiwalled carbon nanotubes (BS-MWNTs) was achieved by using an atomic force microscope (AFM) tip. Nip- and pen cap-like shapes were observed in the BS-MWNT cut sections. By close inspection of the cut sections, we were able to identify the mechanism of the mechanical cutting in BS-MWNTs. © 2005 Elsevier Ltd. All rights reserved.

PACS: 81.07.De; 61.46, + w; 81.40.-z

Keywords: A. Nanostructures; B. Thermal chemical vapor deposition; C. Scanning probe microscopy; D. Mechanical property

1. Introduction (AAO) templates 11 I ] and impactive cutting by steel balls 112], are also reported. Because of their excellent physical properties [1-4], the If one can easily tailor the carbon nanostructures, carbon nanotubes (CNTs) have attracted much attention and carbon-based ‘bottom-up ’ nanotechnology will be more have been intensively studied. The cutting of CNTs has been feasible. For that reason, various nanostructured forms of attempted for the purpose of building block materials in the carbon have been studied and reported, e.g., carbon nano-scale and functionalizing at the CNTs ends. Thus, nanoscrolls [13], carbon micro-trees [14], carbon nano several methods of CNTs cutting have been reported, the cages [15], and carbon-glass beads [16]. The bamboo chemical etching being a typical method for shortening the shaped multiwalled carbon nanotubes (BS-MWNTs), CNTs [5-7], however, it usually disrupts the CNTs walls. named by the bamboo-like compartment structure of the While electrical cutting using a scanning probe microscope inner tubes, are now among the familiar and popular (SPM) tip has also been reported [8,9], it is not adequate for nanotubes. The BS-MWNTs can be obtained by nitrogen mass processes. Recently, mass-cutting methods by means doping 117] and plasma enhanced chemical vapor depo of the lithographic processes were reported [ 10]. Besides, sition (CVD) methods [18]. Besides the compartment various mechanical cutting methods of CNTs, such as the structures, the BS-MWNTs possess an additional unique sonication cutting of CNTs in anodic aluminum oxide structural feature: Both continuous and discontinuous graphite sheets on the outside walls. Thus, they have attracted much interest and can be expected to be applied as building block materials in the nano-scale. For example, * Corresponding author. Tel ; 82 2 3290 3098; Fax: +82 2 923 their improved mechanical stability by the bamboo structure 2682. will be suitable for use as supporting blocks in the nano E-mail address: [email protected] (C.E. Lee). scale. Here we report nib-shaped carbon nanolubes, which

:^>luiriKl hy r-y*;liur.iLul ^ullir.v ofihK iiiinhcin . cbr-v in^hxliin of ihr e:i «cciM>a, w?

2. £a?pei1uMW

JfiC BS-MV/N'II V.:,c yowi: :^«JX VI: I:* bV cai&ly£tt Jv^vS-XC/j SiO;. iu:»*XfiiU'i w.:ls CYD iirtlidds •>U U lU- v ’.ctir.v. V>«xt 1 K JAbK) iiiv^crimrinn nftnc?V0*s '.v-v.t dr.p'ird or ii?My n/i?i-M pyr^lyrir ni-nphre ittO?Cr> xnbrrftro: The mrch- .iri::ul ~ur.hp \vn« ::nrrieii nul by cntrurA rr»::i>. AP/1 v:iih conio.il S' rip

3. lltxulfe dfcCuttiOu

hi^-. 2 show- b JISM'A’NI cul l;> an Abtt L.p Ita? images hflcit ;xii: a:\v- I>ik cul air vs-i:in uiikjut acnapes ilx v.:i teuton* T %nd d Ivvk.i^vUfilip.u il pc:i vap wliv.'Uv. lliv vul sivLOi: 1 2' anJ M reicttMc a r.ib ji api>5ni$ a> il ihcy an- «•< a :. U .1 p.M CO I Of cacti Oil# if These cl?S<&ICiiSr r. rvi fCftl.tt.i nrty Ivt aivomrcd for hv rhr- hrr.nVolikr vuviV.fO. rvoifctfl 'n ikcTfiM inm^ (Ti<; 11 4'- n;% mb-if5 fCulV/CE rirc shown in rhe .-.ud;ng 0: n:;ilLe>.!lkr: cacban nanou:oc (MWNTs) l:< 11. i %. lh<*v wr ly hnt (APM) :ip. In H e je 0 2 re Nl.y. rr.-.^e o!if: :h: •"•illi f iiiih ^.*. I:,:!i:i -.ll

xmi I V lv'> Ififtll AAV. Art. hi rii:;l !>• "I ft iZ '7 A'»", 111.- CdlAf hi •V xin: I Ih: dAv.'iwi:l . iii.'^i liA.1 A vt :1k Clll xrv >J:rx,vrt

l ie-1 r: I e l.xi.o- :a <: Ti;$e of uie ^xrpr. :i •: ift:«b • I ^ 1. 1l:M ,ti re .ttrrr^v-inuli -.ilkL (irhâ ;:i :cr a Iai a r K:i.:.|xvm "*:.x;i,ix* 2 »V '• iii. •" Sc»"v ftot* '.'c-viui'v • rt.-.v. ii ' * ?•* • %W:, f«1 as ik;(ip is scanning she vpr^r-JeO pur (?) in ;;ig. : Wcrpi eon puc "kf fricfrir. fnK.es iv.-.iirring v.tcn Ins tip :;jii< r.vt BS-fc-WST 'rtr. v.irr r p,\ in B in th* cmtiiLT pLot l' y. *l(h> rhiic to- ur.V.rr :"riv:i<:r. fcrcc is aim.is! the n;< I nr. ya«ii<§'«:vî IiiCUOji :o /X\ Whin <• MW.V. iOil- i.:i Hi. Kl>5Xi Will: :Oi;iiuCi2i.Ril? wi.iaci liiC iriCliCci lui-X is ^vi>" laiuo Ic'J.cl |. >lii.v il Lwumvv aimi ’.lvi ’lie MW*V|' znuvito i>a (he- HOl'CJ w:ih uxommer.iVJiite XnilaCl #x: ea.llcd y.vuUi^' .X Oi '.lie vtliix haivd. I..\ bS-MXV.Nfs Ix-.vtr îMp ike V.iXlS in: Cl.-. vulS.Lv ;>3lU near l:i: uur:ip4itnx;il aliuUuiva. so Ilial d CvuiuCiixjiaiv vu.iU.t will; HOl'G h*:cly happerg. h n a BS-MV/NI dory /101 CXpC/iCoX <. ccai-*;ikc fail witi: •jairtnicttiu.-a.c CvnU-.C <$i; iao H0?G. bU :Jv eel oy n force small w chofricuo.i fvicc tiucunMig win Ik Lb.- AIM I::: ix pasciig iwki il. lie CulLin^;uivchiriisui oi US-MWN1 > etii^wulaiixd lodiJlvi Ir0ni4l-.il u.’ M WN l a by lit eu?i:partiii6iil atiuSU-i: Cûtwl :riSyXliO:i c- UiC wall ilruUU.-v near a BS MW'ST i\ini|Xistir>f!il jOvtals a clmig* in Uik wxIS lliit &iiex^ i-ur l o conipirlrrenl joint i.o capering of the Jiivnid grapkem nSft-L:. oeermpanird ky d'Mpp-annnr of ibf. nulsirlr grnphpnc shselx (Pip,. 4u>J Dceio lb* Un-r-il emifig-jraliec. ; ibnniT cull*i %v*ill ^raixiiu: r\wx>.\ (I ' :. uoiiriufteU In (IA upyir LKiJip«i/Ii:ivi:l. would bivak :noi% easily Uiuri ll •. •J:.2ktiri:n:orw?.ll giapliiUxljvsb (sX. V/lvewis break.i:p ky (la; A KM lip ll;i JllLig Sia:L< I'lfxn lilt* i:Uo: valla. lk^ '.nnr* vâl.y. û:i'.y wnsuCkd Iv :l>. winpizinio.l. >vu !l y i:: o.l ;lxcuvr walLs raikvr :l».>n breakc.T. I k.n. ihs 13o MWNTs cm lx L Jl .luirt: tavilv man I'ne MW NT keL-aLl Ikr- f'i^. 1. tt»> lii^n resviai an TCM U.'R-TGM'l ;rrj>v ef a •k.ini-r :'Ai$er v-zll grap.‘«r^ tkw:s af *J-c conpnr/riu oin: :tz..^n cf a DH M'vs.. Tve :.nvwr. inVic.vz %lruLl-rv<. C c w:l! ih:n •• ihe :ipenr? of :nt n>xl.Annum of ;l>- HS-NTA'NTs :s cicely cz,ao;ia*vJ w.ia xlicol u^phi'e^vj; x.rvu’i .mI/v f:i!l::w% '•> '*»' if ;kjftii '.vn*i smie*»icca. Vae-xb-LKe cos seei.onsai:-e from ilc airî'Jv yyyU'i:: alKi:> ('•} H1-TK>Z imifv nl A KS-MXVM *!•••»» iiiu Iim! llii: nin:i vul .% vf ll:c BS -SCV/h- Yi s.un r jrrchnnie^l cml:nf/;fihr jsisiSv Mkioc pl.tre nl ilix ii«j«niali;zl In - the i .o.'cicd in :h? upp:r diccOiT.panmect sirucavc2 w.Lh llimno o-jcci wal: 5;aphei.c wyMr.i" ;'l i'J' T— iki::hi..t* $:l I'•' • if- well< 'Wv^k'S sheers Vhe fricuon fnjee mcoYuromen-s roniv.n Thai ilv il 'iirn :•:: i BS-MXWfs ear ao Mir wi*k conrnni forr.es a< small :he ffir.i.rtt fore.- oeeor r » -vhile. ihe AfM np is \.w.n;-,o.f i

i-.ipuDUCd k? Conic: for N:I1 or.ifvr, r.nri fv:no

llu w,a:< .v^n aC;i|».iilXl '.he IvuCa bc.CiiCc &:id lirgri.ie-.r.r.v l\ii.:iu.rkir. cHOI lvU5 f;i!u IC?v< b.yni rn- RotlT4!riCI'S ion /.u,i'lxra:or Provrair. N;: N*i! ill \%S!! IX!Ul C2fClWl-0lf 4; F.od by -ko tCoxn ^'sci rk f'csie.^mio^ lij fc" S iV^Mh.s N!I..A X‘.«41 ;G»r,i:i N:. and Ef^.r. Koica 21 PJ S kr .h l> I:v.eh4-:l VI. V.-;W..-, dn H-er. Scir.ine 2ZU IN'jtfU II -(»!; >!. ikr i-.e.LSu-en-ifiits j lire Kurexi llcii. ; CCB SVierer liwl.luir (IviSH ms u»:kno*vled,-v:!. C .*. *.v:i< PI 1 Hose.M v/- :n:. \. Z:.1U Symh. Me? 10? i*i 9 ?y'. Wt 686 J.W. Jang et ai / .Solid Slate Communications 135 (2005) 683-686

[4] A. Krishnan, E. Dujardin, T.W. Ebbesen, P.N. Yianilos, M.M. [12] C. Daraio, V.F. Nesterenko, J.F. Aubuchon, S. Jin, Nano Lett. J. Treacy, Phys. Rev. B 58 (1998) 14013. 4 (2004) 1915. [5] S.C. Tsang, Y.K. Chen, P.J.F. Harris, M.L.H. Green, Nature [13] L.M. Viculis, J.J. Mack, R.B. Kaner, Science 299 (2003) 1361. 372 (1994) 159. [14] P.M. Ajayan, J.M. Nugent, R.W. Siegel, B. Wei, Ph. Kohler- [6] J. Liu, A.G. Rinzler, H.J. Dai, J.H. Hafner, R.K. Bradley, P.J. Redlich, Nature 404 (2000) 243. Boul, A. Lu, T. [verson, K. Shelimov, C.B. Huffman, F. [15] Y. Saito, T. Matsumoto, Nature 392 (1998) 237. Rodriguez-Macias, Y.S. Shon, T.R. Lee, D.T. Colbert, R E. [16] W.A. de Heer, P. Poncharal, C. Berger, J. Gezo, Z. Song, J. Smalley, Science 280 (1998) 1253. Bettini, D. Ugarte, Science 307 (2005) 907. [7] Z. Gu, H. Peng, R.H. Hauge, R E. Smally, J.L. Margrave, [17] X. Ma, E G. Wang, Appl. Phys. Lett. 78 (2001) 978. Nano Lett. 2 (2002) 1009. [18] M. Okai, T. Muneyoshi, T. Yaguchi, S. Sasaki, Appl. Phys. [8 ] J.-Y. Park, Y. Yaish, M. Brink, S. Rosenblatt, P. McEuen, Lett. 77 (2000) 3468. Appl. Phys. Lett. 80 (2002) 4446. [19] C.J. Lee, J. Park, Y. Huh, J.Y. Lee, Chem. Phys. Lett. 343 [9] D.-H. Kim, J.-Y. Koo, J.-J. Kim, Phys. Rev. B 68 (2003) (2001) 33. 113406. [20] M.R. Falvo, J. Steele, R.M. Taylor II, R. Surperfine, Phys. [10] S.R. Lusting, E D. Boyes, R.H. French, T.D. Gierke, M.A. Rev. B 62 (2000) R10665. Harmer, P.B. Hietpas, A. Jagota, R.S. McLean, G.P. Mitchell, [21] K. Miura, T. Takagi, S. Kamiya, T. Sahasi, M. Yamauchi, G.B. Onoa, K.D. Sams, Nano Lett. 3 (2003) 1007. Nano Lett. 1 (2001) 161. [11] S.-H. Jeong, O.-J. Lee, K.-H. Lee, S.H. Oh, C.-G. Park, Chem. Mater. 14 (2002) 1859. PHYSICAL REVIEW B 72. 054439 (2005)

Monte Carlo study of the Kosterlitz-Thouless transition in the Heisenberg model with antisymmetric exchange interactions

Kyu Won Lee and Cheol Eui Lee* Department of Physics, Korea University, Seoul 136-701, Korea (Received 10 August 2004; revised manuscript received 6 June 2005; published 25 August 2005)

We have studied the two-dimensional Heisenberg model with antisymmetric exchange interactions by em ploying a Monte Carlo simulation, which confirmed a Kosterlitz.-Thonless transition. The Kosterlitz-Thouless transition temperature and the energy required to create a bound pair of vortices were calculated as a function of the magnitude of the antisymmetric exchange interaction and compared with the results of the supposedly equivalent two-dimensional easy-plane Heisenberg model.

DOI: 10. ll03/PhysRevB.72.054439 PACS number(s): 75. lO.Hk, 75.30.Kz, 75.40.Mg

I. INTRODUCTION a=- vZ (%,+%.+% Molecule-based magnets have been regarded as represen r,a tative quasi-two-dimensional (quasi-2D) Heisenberg where .7=./vT+D2/./2 and \= 1 /V7+D2/./2 X=1 {DIJ=0) magnets ,12 consisting of alternating magnetic and nonmag corresponds to the isotropic Heisenberg model. In the limit netic layers. The anisotropy ratio of the interlayer exchange interaction to the intralayer one is reported to be about Q (/}/./ — r-o), the system leads to the XT model. 10-6-l 3 Because the two-dimensional Heisenberg model The easy-plane anisotropic Heisenberg model in two di does not yield a long-range order at any finite temperature, it mensions provides a prototype for systems exhibiting topo can be expected that the magnetic ordering temperature will logical excitations, such as superfluid films, layered super decrease down to zero with increasing interlayer separation. conductors, layered magnets, lipid layers, and others .12-14 This prediction was confirmed for short chain lengths- ’ but The model undergoes a KT transition, i.e., a vortex unbind failed for long chain lengths .12 An Ising type of ing transition, at a finite temperature for X< 1 and when X anisotropy 3-5 and a dipolar interaction 6'7 were suggested to —► 1 Tkt is predicted to approach zero as 1 /ln( 1 — A).1516 Es be responsible for the finite temperature magnetic ordering. sentially, no finite temperature magnetization can exist, even However, the Dzyaloshinsky-Moryia (DM) interaction, below Tkt. The in-plane susceptibility diverges exponen which gives rise to a spin canting, has not been considered as tially as 7'kt is approached from above and remains infinite an origin of a magnetic ordering. The DM interaction can below fKT. The specific heat does not diverge at Tkt against induce an easy-plane anisotropy, from which a Kosterlitz- usual second-order phase transitions .8 9 In practice, due to a Thouless (KT) transition is expected. 8>) A two-dimensional very slow decay of the spin-spin correlation with distance, a classical Heisenberg antiferromagnet with a DM interaction finite magnetization is observed even in a macroscopic scale was suggested to show a KT transition due to an XT-like and the susceptibility shows a divergent behavior as Tkt is degeneracy of the ground state.10 and a self-consistent har approached from below and above, similar to that of the monic approximation (SCHA) theory was used to calculate conventional ferromagnetic transition .1718 Two types of vor the KT transition temperature. (fKT) for the two-dimensional tices. i.e., in-plane and out-of-plane vortices below and Heisenberg model .11 In this work, we performed a Monte above A—0.7, have been known for the easy-plane aniso Carlo simulation to study the KT transition in the two- tropic Heisenberg model and their static and dynamic prop dimensional Heisenberg model with a DM term. erties were intensively studied.14 The two-dimensional Heisenberg model with a DM inter The KT transition is characterized by a jump in the Felic action is described by 11 ity modulus y from 277it to 0 at 7'KT.'20 y is a measure of the rigidity of an isotropic system under an imposed phase twist *=-jZW,„,+ 2(-ir'o - cl % w, (0 and is directly related to the supcrlluid density in a superfluid film.12 y is defined by the difference in Helmholtz free en where r+a labels the nearest-neighbor sites of >-=(/,/). The ergy obtained by applying periodic and twisted boundary sum runs over all pairs of nearest neighbor spins on the conditions. The difference between the, internal energy ob square lattice, and we have taken 5=1 and .7=1. The factor tained under periodic and anliperiodic boundary conditions (-1 )H/ lakes into account the alternation of the Dzyaloshin- yields the temperature derivative of v.21 Because the finite ski vector from one site to the nearest neighbor. For conve lattice sizes in a computer simulation cause a smoothing of nience, we shall take D to be directed along the r axis, for the jump in v, it is not easy to directly determine 7’KT from which the easy plane will be the v-v plane. The Hamiltonian the jump and several types of finite size analysis were used Lq. (1) can be transformed to the easy plane anisotropic for v.22-24 Heisenberg Hamiltonian by rotating all the spins by an angle In real magnets, the DM interaction is frequently ob of el2 where tan e-DU\!v served and plavs an important role in a magnetic ordering.

1098-012 l/2W5/72i5,/()54434ibi/9,23.00 0544 '4. iT:2UI)5 The \iucricnn Physical Society KYI WON LEE AND ( III OL EUI LEE PHYSICAL. REVIEW B 72. 054439 (2005) whereas the anisotropic exchange interaction in the Hamil product of spins, it may not he the case for the DM interac tonian of Eq. (2) is rather fictitious. The equivalence of the tion with a vector product of spins. Using Eqs. (I) and (3), Hamiltonians and the simpler form of Eq. (2) tempt one to the energy difference £(5new)-£(5„|U) is easily calculated to study the model Hamiltonian of Eq. (2) instead of Eq. (1). in he zero for the first term in Eq. (1) corresponding to the spite of the question regarding the role of the DM interac exchange interaction. In the same way. E'(5ncw)-E’(>5'old ) = tion, in the belief that the physical properties of the two Hamiltonians are the same. Nonetheless, the issue of whether -2£(50ld) is obtained for the second term in Eq. (1) corre the physical properties of the two models are completely the sponding to the DM interaction, indicating that the energy is same or not remains to be resolved. To the best of our knowl not preserved. We have actually tried the ovcrrclaxalion al edge, no systematic Monte Carlo simulation has been carried gorithm in our Monte Carlo calculations. A combination of out, and the KT transition expected from the equivalence has the Metropolis and overrelaxation algorithms greatly reduced never been explicitly confirmed in the model Hamiltonian of the correlation time for the easy-plane Heisenberg Hamil Eq. (1). In this Monte Carlo study, we will show that the tonian Eq. (2). However, the combination did not work well energy required to create a bound pair of vortices is quite for the model Hamiltonian Eq. (1) as is expected for the DM distinct in the two Hamiltonians, while the KT temperature interaction, for the reason of the above-mentioned energy and related critical exponents are quite consistent. Thus, the conservation constraint. As neither the cluster nor overrelax supposedly equivalent Hamiltonians of Eqs. (1) and (2) do ation algorithms can be applied in the presence of a DM not necessarily have to give the same physical properties in interaction, we were left with the only choice of the Me every respect. tropolis algorithm.

IT. MONTE CARLO SIMULATION III. HELICITY MODULUS We performed Monte Carlo calculations in the model The time correlation function of energy was calculated Hamiltonian Eq. (V) using a standard Metropolis algorithm following conventional definitions: with periodic boundary conditions, unless specified other wise. We have used lattices of size L X L with L up to 100. C[(-) = <£(())£(T)> -<£(0)X£M>, (4) All the measurements were carried out by decreasing the temperature from an infinite temperature. 1-5X 101 MC’Ss E(r) = (-7/2L2)E-VTU.S'rwtr) (Monte Carlo steps) were used for thermal equilibration. In order to reduce the correlation time, a combination of Me tropolis and cluster algorithms 25 or a combination of Me + (1 /2L2) E I ' [L r) X LXT)], (5) tropolis and overrelaxation algorithms 26 have been success fully applied to the two-dimensional XT model and the two- dimensional easy-plane Heisenberg model of Eq. (2).,b'27"29 C,., = £2[(£2)-(£)2]/T2, (6) The cluster algorithm cannot be applied to the model Hamil where C\ corresponds to the specific heat. The factor of I /2 tonian of Eq. (1) including the DM interaction. in Eq, (5) was introduced to avoid double counts. An important concept of the Wolff ’s single cluster algo Cz.-(t) was not an exponential function and, if not exact, rithm, known to be very efficient for spin systems with con tinuous symmetry, is the generalization of the spin-flip op was close to a stretched exponential form: eration in the Ising model .25 A random-bond Ising model C(t) = C(0)exp[- (t/t,)" |. (7) without frustration and competing interactions can be con structed from a spin system with continuous symmetry by C[At) was fitted by Eq. (7) in order to acquire a rough esti choosing a random direction, projecting spins on that direc mate of the energy correlation time te. The maximum value tion, and then assigning +1 and -1 to spins of parallel and of r£ is about 20 MCSs regardless of D and is independent of antiparallel projections to it.3,1 The products of these parallel size for L >20. The exponent n had a minimum value of and antiparallel components of spins determine the couplings about 0.5 at I he temperature of maximum r£. In order to between the nearest-neighbor Ising spins. The Ising spins avoid a correlation, measurements were taken every ~I0 with the same direction build a cluster which can be Hipped x - MCSs. As a result, KT - 106 averages for a physical as a whole, leaving no room for interactions with vector quantity were taken. Figure I shows the specific heat for D product as the single cluster algorithm takes an analogy to = 0.5, where the maximum value is independent of the lattice the Ising model. size and the temperature of maximum Cv has nothing to do The ovcrrclaxalion algorithm consists of reflecting the with TK1. which is characteristic of a KT transition. While the analytic form of the helicity modulus is known spin at a given site about which is the sum of the nearest for the XT model and the Heisenberg model.22'2111 it has not neighbor spins.26 25 With l = |l!':X.2' been obtained for the model Hamiltonian Eq. (I) that in

cludes a DM interaction. Thus, the helicity modulus could

only be obtained by calculating the difference dll between While this gives rise to the largest possible step while pre the internal energy under periodic I (/,,) and anliperiodic \ Ua) serving the energy for the exchange interaction with a scalar boundary conditions :21

054439 2 MONTE CARLO STUDY OF THE KOSTERL1TZ-... PHYSICAL REVIEW B 72. 054439 (2005)

1.12-

1.10-

1.08-

K 1.06-

1.04- u 7/0=1.59 * 7/0=1.585 1.02- L=100 o 7/0=1.58 ■ 7/0=1.575

FIG. I. The specific heat Cv measured with D=0.5. Errors are FIG. 3. Finite size scaling plot of y for 0=2. The solid lines are comparable to the symbol sizes. fits to the scaling relation of Eq. (9) and the best fit was given for 7’kt= i.585±().()05. For L< 8 , the scaling relation was not applicable. \ = (ea - Up)hr = dUI tj2, (8 )

IV. MAGNETIC SUSCEPTIBILITY where [1- 1 IT and the internal energy U=(E). The antiperi- The time correlation function of magnetization was calcu odic boundary condition was imposed only along the x axis. lated following conventional definitions: Figure 2 shows the helicity modulus y for D-2, which was obtained by integrating dU. The inset shows dyUiT. which (m(0)/?;(T)) - (/n(0))(/»(r)). (10) shows a divergent behavior with increasing L indicating a discontinuous jump in y. At 7'KT, y is described bv22-23 i(t) = (1/7)2 .SV(t), (11)

2 1 1 \ 7Vrl 1 + (9) „ = L2[(nr) - {m)2]/T, (12) 77 2 In L+C where x,„ corresponds to the magnetic susceptibility. C,„(t) Tkt was obtained by using Eq. (9). Figure 3 shows (he finite was an exponential function above the temperature of maxi size scaling plot of y for D=2 and the solid lines are the fits mum r,„, whereas it was close to a stretched exponential to the scaling relation of Eq. (9), which gives the best fit for form of Eq. (7) below that temperature, with the exponent n 7Vr=1.585±0.0()5. In the same way, 7’KT=0.73±0.005 was decreasing with temperature. The magnetization correlation obtained for D=0.5. time rm was strongly dependent on size due to a critical slowing down with the dynamic critical exponent c — 2.2±0.1. In order to avoid a correlation, measurements were taken every ~5 X rc MCSs. Taking into account the fact that the magnetization is es sentially zero in I he thermodynamic limit, the magnetization > -4- term M~(m) should be eliminated from Eqs. (10) and (12). However, the magnetization is not zero even in the macro scopic samples 17-18 and Ihe critical temperature is hardly af fected by the definitions, as discussed in a previous work .16 The effective critical temperature T, in a lattice of finite size is directly defined as the temperature of maximum x,„ m general, as in the conventional magnetic transition. The pro priety of [he definition can be checked by I he finite size scaling relation intrinsic to the KT transition :1 -‘s ’-

7H7J = 7 VT + (13) 4c( In 1.1 T At T . the spin-correlation length t is prolortional to Ihe lat FIG. 2. The helicity modulus y for /.)= 2 with L- 10. 20. 40, and hee size, i-xL, and the magnetic susceptibility behaves as 60. The straight line corresponds to 27/rr. Inset shows dy/dT. ,Y„, -x with tj- 1 /4,11 Wc tints have

H54439-3 KYU WON LEE AND CHEOL EUI LEE PHYSICAL REVIEW B 72, 054439 (2005)

D=1.2

0.7-

0=0.8

0.5 0.7

cr i.o- 0=0.5

0=0.2

FIG. 4. Finite size scaling plot of Tr(L). The solid lines are fits FIG. 6. Tkj/J as a function of X. The values marked by ■ were to Eq. (13). Errors are comparable to the symbol sizes. obtained from Fig. 4 using Eq. (13). whereas the symbols ° and * were taken from previous works for the two-dimensional easy plane. Heisenberg model (Refs. 16 and 33). Inset: the coefficient c/7 in (14) Eq. (13) obtained from Fig. 4.

The scaling relation of Eq. (14) can also be used to indepen Kosterlitz-Thouless nature of the transition at Tkt. dently check the propriety .16 Figure 4 shows the effective critical temperature TC{L). Figure 6 shows 7KT/.7, obtained from the fits in Fig. 4, as obtained from the temperature of maximum as a function a function of X. Previous works on the two-dimensional of (In LY1. For a KT transition, the effective critical tempera easy-plane Heisenberg model are compared with our results: ture is not described by a power of the system size, but is the symbols ° and * correspond to the results of a Monte described by a logarithm of the size. Tkt can be determined Carlo simulation and a self-consistent harmonic approxima by the finite size sealing formula Eq. (I3).1' IK In fact, the tion theory, respectively. 16'” Our estimates of Tkt/J are effective critical temperatures were well fitted to Eq. (13), as slightly larger, however, the two-dimensional Heisenberg shown in Fig. 4. Figure 5 shows xm{Tr) as a function of L, model with a DM interaction and the two-dimensional easy- which was well fitted to Eq. (14). The exponent r; obtained plane Heisenberg model appear to yield the same Tkt, as from the fit is shown in the inset, r] is expected to be 1 /4 at expected from the equivalence of their Hamiltonian. The in Tkj and our values are very close to 1/4 except near X set of Fig. 6 shows the coefficient c/J in Eq. (13) obtained ~ 1(D~0). For D-2 and 0.5, 7’KT obtained from the fits in from Fig. 4. Its value is about 2.5, except near X— l.9 Fig. 4 were 1.6±0.006 and 0.73±0.005, respectively, which are very close to those obtained from y. These results justify V. VORTEX DENSITY the definitions of Eqs. (1 ())—(12) and evidences the Following a previous work ,'4 each vortex or antivortex was counted with L= 100, a vortex core being composed of * D=2 lour spins on a unit square. When the sum of relative polar □ D=1.2 angles on a unit square is equal to 2ir, the number of vortices A D=0.5 increases by one and the vortex density p is the total number of vortices divided by Lr. The time correlation function of p was calculated following a conventional definition:

C( 7) = (ptOjpl rj> - < p (0 ) )(p( t)) . (15) The lime correlation function of p was not exponential, bul was close to a stretched exponential form of Eq. (7). The measurements were taken even 10-20 <7, MCSs. which reduces the correlation between measuieinems bv more than 90%. Around 7kT. the vortex density follows exp(-2/r/7"), where 2/u is the energy required to create a bound pair of vorticesA ‘ os pjg Ure 7 shows -In p as a function of 1 IT for FIG. 5. xJT, I vs. L The solid lines are tils lo Eq. ( I4l. Inset /) = () 5. In a previous work. ” the data can be divided into shows tire exponent r/. three regions: low-temperature 17'-: TK1), intermediate tem-

05445"-! MONTE CARLO STUDY OF THE KOSTERL1TZ-.. PHYSICAF. REVIEW B 72. 054439 (2005)

-- t

FIG. 7. The vortex density p vs 1/7’ lor 77 = 0.5. The solid and FIG. 8 . The energy required to create a bound pair of vortices the dashed lines are linear fits for 7<7KT and 7KT<7<7M, re spectively. 7V, indicates the temperature of the maximum specific 2/Z = 2/t/ v'l +D2 as a function of X. The square and the circle cor heat. Errors are comparable to the symbol size. respond to the Hamiltonians of Eqs. (1) and (2). respectively. The solid and the open symbols correspond to the regions of 7<7K-r and 7'Kt<: 7< T,w, respectively. Errors are comparable to the sym peralure (TKJ 1: it shows a quite distinct behavior when X—>0. 2/c in the to be in general agreement with those of the two-dimensional easy-plane Heisenberg model of Eq. (2) shows a monotonic easy-plane Heisenberg model, the energy required to create a decrease with increasing X, similarly to the monotonic de bound pair of vortices showed quite distinct behaviors. crease of 7kt. Taking into account the fact that the KT tran sition is a vortex unbinding transition, the energy required to A C KNOWLEDG M EN TS create a bound pair of vortices can be expected to be propor This work was supported by the Korea Science and Engi tional to 7kt. On the other hand, it is quite interesting that neering Foundation (RO1-2005-000-10798-0 and Proton Ac 2/Z in the Fleisenberg model with a DM term of Eq. (1) celerator User Program No. M202AK01002I-04AI101- shows a maximum at \~0.8. While the static critical prop 02110) and by the Korea Research Foundation (Grant No. erties of the two models, such as the KT transition tempera KRF-2004-005-C00060 and Brain Korea 21 Project in ture, are quite consistent as may be expected from the 2005). We thank Dr. H.-J. Woo at the K1GAM for the proton- equivalence of the two model Hamiltonians, the distinct be beam irradiation. The measurements at the Korean Basic Sci haviors of 2/Z. indicating distinct vortex dynamics, demon- ence Institute (KBS1) are acknowledged.

’’Corresponding author Electronic address: rsccK"'korea.ac.hr; Pax 1 V. Yu. Irkhin and A A. Kalanin. Phys Rev B 57, 379 I 1998!. + 82-2-927-329: SY Yu. irkhin. V A Kalanin. and M 1 Katsnelson, Phys. Rev B 1 K. W. Lee, C. H. Lee, C. E. Lee. and J. K. Kang. Pins. Rev. B 60. 1082 11999) 62. 95 (2000). AM. Dnllon, P. Panissod. P Rahu, J. Souletie. V. Ksenolontov, and -M. A. Girtu, C. M. Wynn. W. Eujita, K. Awaga. and A. J. Epstein. P. Gutlich. Phys. Rev B 65. 104404 (2002). Phys. Rev. B 61. 4117 (2000). AS. Ostrovsky. W. Haase. M I billon, and P. Panissod. Phys. Rev 1L. .1. de Jongh, Magnetic Properties oj Layered Transition MelaI B 64. 134418 (2001). Compounds (Kluwer Academic Publishers. Netherlands, 1940). Ah VI. Koslerlitz and 1). .1. Thoultvs. .1. Phys. C 6. 1181 I 1973).

054439-5 KYI WON LEE AND CHEOL EU1 LEE PHYSICAL REVIEW B 72. 054439 (2005)

"J. M. Kosterlitz. J. Phys. C 7, 1046 (1974). (1977). 111 A. R. Vblkel, F. G Mertens. A. R. Bishop, and G. M. Wysin, 21J. E. Van Himbergen, Phys. Rev. B 25. 5977 (1982). Ann. Phys. (N.Y.) 2. 308 (1993). 22H. Weber and P. Minnhagen, Phys. Rev. B 37. R5986 (1988). 11 A. S. T. Pires. Solid Stale Common. 112. 705 (19991. 22 P. Olsson and P. Minnhagen, Phys. Rev. B 43. 3356 (1991). 12M. E. Fisher, M. N Barber, and D. Jasnow, Phys. Rev. A 8 . 11 i 1 24 P. Olsson. Phys. Rev. B 52, 4511 11995). (1973). 25 U. Wolff. Phys. Rev. Lett. 62. 361 (1989). nB. Horovitz. Phys. Rev. B 45, R12632 (1992). :,’M. Creulz, Phys. Rev. D 36. 515 (1987). I4C. S. O'Hern, T. C. Lubensky, and J. Toner. Phys. Rev. Lett. 83. 27J. Apostoiakis. C. F Baillie. and G. C. Fox. Phys. Rev. D 43. 2745 (1999). 2687 (1991). I?C. Kawabata and A. R. Bishop. Solid State. Common. 42. 595 28 R. Gupta, J. DeLapp, G. G. Batmuni. G. C. Fox, C. E. Baillie, and (1982). J. Apostoiakis, Phys. Rev. Lett. 61. 1996 (1988). 16A. Cuccoli. V. Tognetti. and R. Vaia, Phys. Rev. B 52. 10221 29 H. G. Evert/, and D. P. Landau. Phys. Rev. B 54, 12302 (1996). (1995). 2UK. Chen. A. M Perrenberg, and D. P. Landau. Phys. Rev. B 48. I7S. T. Bran we 11 and P. C. W. Holdsworlh. J. Phys.: Condens. 3249 (1993). Matter 5, L53 (1993). 21 N. Sehultka. Phys. Rev. B 55, 41 (1997). lsS. T. Bramwell and P. C. W. Holdsworlh, Phys. Rev. B 49. 8811 22 S. G. Chung. Phys. Rev. B 60, 11761 (1999). (1994). 22 A. S. T. Pires. Phys. Rev. B 54. 6081 (1996). '"G. M. Wvsin. Phys. Rev. B 49, 8780 (1994). 24J. Toboehnik and G. V. Chester, Phys. Rev. B 20. 3761 (1979). 20D. R. Nelson and J. M. Kosteilitz. Phys. Rev. Lett. 39. 1201 25 R. Gupta and C. F Baillie, Phys. Rev. B 45. 2883 (1992).

054439-6 Available online at www.sciencedirect.com SCIENCE ZW) DIRECT- SONd ^ state communications ELSEVIER Solid State Communications 135 (2005) 95-98 www.elsevier.eom/locate/s.sc

Helicity modulus and vortex density in the two-dimensional easy-plane Heisenberg model on a square lattice

Kyu Won Lee, Cheol Eui Lee*, In-mook Kim

Department of Physics and Institute for Nano Science, Korea University, Seoul 136-713, South Korea

Received 16 December 2004; received in revised form 23 March 2005; accepted 24 March 2005 by A. Pinczuk Available online 13 April 2005

Abstract

We have studied the two-dimensional easy-plane Heisenberg model by using a Monte Carlo simulation. The helicity modulus and the vortex density were calculated. The Kosterlitz-Thouless transition temperature obtained from the helicity modulus was consistent with previous results obtained using different methods. The energy required to create a bound pair of vortices shows an anisotropy dependence similar to that of the Kosterlitz-Thouless transition temperature. © 2005 Elsevier Ltd. All rights reserved.

PACS: 75.10.Hk; 75.30.Kz; 75.40.Mg

Keywords: D. Phase transitions

1. Introduction unbinding transition, at a finite temperature for A < 1 and when A-»l the Kosterlitz-Thouless temperature (7Kt) is The two-dimensional easy-plane Heisenberg model is predicted to approach zero as l/ln(l —A) [6,7]. Essentially, described by no finite temperature magnetization can exist, even below Tkt. The in-plane susceptibility diverges exponentially as # = -/ (i) Ekt is approached from above and remains infinite below 7kt. The specific heat does not diverge at 7KT against usual second order phase transitions [4,5J. Two types of vortices, where r+a labels the nearest-neighbor sites of r = (i',;'). The i.e. in-plane and out-of-plane vortices below and above sum runs over all pairs of nearest neighbor spins on the A~0.7, respectively, have been known for the easy-plane square lattice, and we have taken 5=1 and 7=1. A = 1 Heisenberg model and their static and dynamic properties corresponds to the isotropic Heisenberg model. In the limit were intensively studied (8 j. A Monte Carlo simulation was A->0, the system leads to the XY-model. The easy-plane reported for the susceptibility, the correlation length, anisotropic Heisenberg model in two-dimensions provides a internal energy, and specific heal, where 7'K1 was obtained prototype for systems exhibiting topological excitations, from the susceptibility and the correlation length [7], A such as superfluid films, layered superconductors, layered consistent result for Tkt was given by a self-consistent magnets, lipid layers and others [ 1-3]. The model undergoes harmonic approximation [9 1. a Kosterlitz-Thouless (KT) transition [4.5], i.e. a vortex The KT transition is characterized by a jump in the helicity modulus from 2T/r: to 0 at 7K r [ 101. y is a measure * Corresponding author. Tel.: +82 2 3290 3098; fax: -I- 82 2 927 of the rigidity of an isotropic system under an imposed phase 3292. twist and is directly related to the superfluid density in a E-mail address: [email protected] (C.E. Lee). superfluid film |1|. y is defined by the difference in the

Helmholtz free energy obtained by applying periodic and twisted boundary conditions. The difference between the internal energy obtained under periodic and antiperiodic boundary conditions yields the temperature derivative of y I'll]. Because the finite lattice sizes in a computer simulation give rise to a smoothing of the jump in y, it is not easy to directly determine Tkt from the jump and several types of finite size analysis were used for [12-14]. In this work, we performed a Monte Carlo simulation to calculate the helicity modulus and the vortex density. From 0.4 0.5 0.6 0.7 o.i 0.9 1 the helicity modulus, TKt was obtained as a function of A and compared with previous results. Taking into account the fact that the KT transition is a vortex unbinding transition, the energy required to create a bound pair of vortices, which is believed to be directly related to Tkt, was obtained from the vortex density as a function of A and compared with Tkt.

2. Helicity modulus Fig. I, The helicity modulus y for 2 = 0.5 with 7=10, 14. 20, 30, and 40. The straight line corresponds to 2Tin. Inset shows dy/dT. We performed Monte Carlo calculations in the model Hamiltonian Eq. (1) using a standard Metropolis algorithm integration processes of calculating the helicity modulus in with periodic boundary conditions, unless specified other Fig. 2 introduced considerable uncertainties of the order wise. We have used lattices of size LXL with L up to 100. 0.01, to which our temperature resolution was confined. The All the measurements were carried out by decreasing the inset shows dy/dT, displaying a divergent behavior with temperature from an infinite temperature. Usually, 20,000 increasing U, indicative of a discontinuous jump of y in an MCSs (Monte Carlo Steps) were used for thermal infinite lattice. At Tkt, y is described by [12,131 equilibration and at least 50,000 averages for a physical quantity were taken. The time correlation function of energy i i ■ (4) was calculated following conventional definitions. The time 2 InU + C correlation function of energy was not exponential, but if TKt was obtained by using Eq. (4). Fig. 2 shows the finite not exact, was close to a stretched exponential form: size scaling plot of y for A = 0 and the solid lines are the fits C(r) = C(0)exp[-(T/Tj"] (2) After a sufficiently long time, the correlation function approached an exponential form just as is expected. The * T=0.71 maximum value of energy correlation time was about 20 MCSs regardless of A<0.9 and was independent of size. The exponent n had a minimum value of about 0.5 at the temperature of the maximum correlation time. When A> 0.9, the energy correlation time and the exponent n decreased with increasing A. Measurements were taken every ~ 10 X rc MCSs, which reduces the correlation between measurements by more than 90%. The helicity modulus was obtained by calculating the difference dU between the internal energy under periodic (Up) and antiperiodic (Ua) boundary conditions [II]: 1 d(/?y) _ (Ua ~ Up ) dU 2 d/3 tf where /J=lIT and the internal energy U = (£'). The antiperiodic boundary condition was imposed only along the x-axis. Fig. I shows the helicity modulus y for A =0.5, which was obtained by integrating dU. The integration was Fig. 2 Finite size scaling plot of y for 2 = 0. The solid lines are fits carried out using a trapezoidal method. The subtraction and to the scaling relation of Eq. 64). K.W. Lee el al. / Solid Stale Communications 135 (2005) 95-98 97

0.70-

0.65-

0.60-

0.55-

Fig. 3. 7"Kt as a function of X. □ was obtained from the helicity Fig. 4. The correlation times of the vortex density for A = 0.1, 0.8, modulus by using Eq. (4). O and X are cited from previous works 0.9, and 0.98. Inset shows the exponent n. [7,9]. measurements by more than 90%. Fig. 4 shows the to the scaling relation of Eq. (4), which gives the best fit for correlation times of the vortex density and the inset shows Tkt = 0.71 +0.01. In the same manner, 7Kt was obtained as the exponent n. The maximum values of the correlation time a function of A. are nearly independent of A for A >0.4 and increase rapidly Fig. 3 shows 7KT obtained from the helicity modulus by with increasing A for A >0.8. using Eq. (4). Previous works were compared with our Around 7KT, the vortex density follows exp( — 2p.IT), results: O was obtained from a Monte Carlo simulation for where 2p is the energy required to create a bound pair of the susceptibility and the correlation length, and X corresponds to the results of a self consistent harmonic vortices [4,15.16], Fig. 5 shows — In p as a function of 1/7 approximation [7,9]. While our estimates of 7KT are slightly larger than those from the previous works, our results are consistent with them within errors, particularly with the results of a self consistent harmonic approximation for X —► 1.

3. Vortex density

Following a previous work [15], each vortex or anti vortex was counted with L= 100, a vortex core being composed of four spins on a unit square. When the sum of relative polar angles on a unit square is equal to 2tc, the number of vortices increases by one and p corresponds to the total number of vortices divided by 1C. The time correlation 7:. < F<7 function of p was calculated following a conventional T< 7,t definition:

C(t) = (p(0)p(r)) - (p(O)Xp(r)) (5) The time correlation function of p was not exponential, but was close to a stretched exponential form of Eq. (2). Fig. 5. The vortex density p vs. 1/7' for A = 0. The solid and the while it approached an exponential form after a sufficiently dashed lines are linear fits for 7<7KT and 7M<7<7Kr, long time. Measurements were taken every 10-20X t\. respectively. TM indicates the temperature of the maximum MCSs, which reduces the correlation between correlation time of the vortex density. K.W. I.-cc el al. /Solid Slate Communications 1.15 (2005) 95-98 for A = 0. The data have been divided into three regions: low symbols correspond to 7<7KT and 7M<7<7KT, respec temperature (7< 7KT), intermediate temperature (7Kt < 7< tively. 2p decreases gradually with increasing A as 7KTdoes. 7m), and high temperature (7M<7) regions, where 7M The inset shows 2fx as a function of A. Apparently, 2fx is indicates the temperature of the maximum correlation time linearly proportional to 7Kt- The solid lines in the inset are of the vortex density. In a previous work [161, three regions linear fits, with a slope of about 23. The linear fit shows that were discriminated by 7Kt and the temperature of the 2fx vanishes at about 7KT~0.45, which indicates that the maximum specific heat. In other words, — In p was found to linear relation between 2p. and 7KT is an approximation be a linear function of l/7between 7KT and 7M, and its slope valid for a small A. However, the linearity appears to be changed around 7M. Thus, while the time correlation was valid at least for A <0.99. Taking into account the fact that calculated in order to ensure thermal equilibrium and may the KT transition is a vortex unbinding transition, it would depend on the algorithm that is adopted, it is possible that be natural that the energy required to create a bound pair of 7m is really a feature of the model in some way. For A = 0, vortices is proportional to 7KT. 7kt and 7M are 0.71 and 0.76, respectively. Using a linear fir In summary, we have studied the two-dimensional easy- to the data, we obtain 2p = 7.48 ± 0.04 for 7KX < 7< 7M and plane Heisenberg model by using a Monte Carlo simulation. 2^ = 6.57 + 0.03 for 7<7KT. In the XY-model with two- The helicity modulus and the vortex density were obtained component spins S = (Sj,Sv), the KT-prediction is 2ju = 10.2 as a function of anisotropy. The Kosterlitz-Thouless for 7< 7kt and a Monte Carlo simulation gave 2p = 7.55 for transition temperature obtained from the helicity modulus 7<7kt and 2p = 8.82 for 7KT < 7<7M [16]. The 7Kt in the was consistent with previous works. The energy required to XT-model with two-component spins has been known to be create a bound pair of vortices was linearly proportional to ~0.89 [16]. 7kt and 2p in the XT-model with two- the Kosterlitz-Thouless transition temperature. component spins are larger than those in the XT-model with three-component spins (corresponding to Eq. (1) with A = 0). A decrease in 7KT appears to be related to that in 2ju. Acknowledgements Fig. 6 shows 2ju as a function of A. The open and the solid This work was supported by the KISTEP (National Research Laboratory and M102KS010001-02K1901- 01814) and by the Korea Research Foundation (Brain Korea 21 Project in 2004 and Grant No. KRF-2004-005- r< 7 C00060).

References

[1] M.E. Fisher, M.N. Barber, D. Jasnow, Phys. Rev. A 8 (1973) 1111. [2] B. Horovitz, Phys. Rev. B 45 (1992) 12632. [3] C.S. O'Hern, T.C. Lubensky, J. Toner, Phys. Rev. Lett. 83 (1999) 2745. [4] J.M. Kosterlitz, D.J. Thouless, J. Phys. C 6 (7 973) 1181. [5] J.M. Kosterlitz, J. Phys. C 7 (1974) 1046. [6] C. Kawabata, A.R. Bishop, Solid State Commun. 42 (1982) 595. [7] A. Cuccoli, V. Tognetti, Phys. Rev. B 5 (1995) 10221. [8 ] G.M. Wysin. Phys. Rev. B 49 (1994) 8780. [9] A S T Pires, Phys. Rev. B 54 (1996) 6081. [10] D.R. Nelson, J.M. Kosterlitz, Phys. Rev. Lett. 39 (1977) 1201. [1 1] J.E. Van Himbergen, Phys. Rev. B 25 (1982) 5977. [12] H. Weber, P. Minnhagen, Phys. Rev. B 37 (1988) 5986. Fig. 6. The energy required to create a bound pair of vortices 2g as a [13] P. Olsson, P. Minnhagen, Phys Rev. B 43 (1992) 3356. function of A. The open and the solid symbols correspond to the [14] P. Ofsson, Phys. Rev, B 52 (1995) 4256. regions of T<7K1 and 7Kt<7<7m, respectively. Inset: 2/r as a [15] J. Tobochnik, G.V. Chester, Plus. Rev. B 20 (1979) 3761. function of '/KT. which shows a linear relation between 2p and 7Kr- 116] R. Gupta. C.F. Baillie. Phys. Rev. B 45 11992 1 2883. The solid lines in the inset are linear fits. H-rnL ABSTRACT » urxs

A/l l IH> MY

Interwall support in double-walled carbon nanotubes studied by scanning tunneling microscopy Mr Hwan MazK. .ae Won Jar;. 3ed Cteo\ tui -se* !*h ,.V.i'ir. . .Vh v« •.'.••• .;•>. vv/-..' .* :V*.vr. ; V-.,. . Cibo J ■ Lee ...... " j ?«,>••" A- fKtfccw'J i* 2:>:«t: .-.wvp.vJ ! 6 Nv ea.Kr :«;ibli>l.r. vri.iru- : br ;.ry 2!; A.u.nicol y iVsC.vvJ itui:. .t:Vpr en.;e >/<• o iijfc riijiielii:^ ifiir.rw^sy i'*: wOM rb:x i -.) aC vrl v.i nr< :i ihr r.vioi.oe -1 mco>.: u v.u r ~.:i:-62.*u?i5t6i=.

Vdilxh; Hc_v.ltlea (CfS SI lie.vt lxv.1 < :riv^lv cU divJ t:c- hiûre I vticwvx rhe akaiiKÎ y iCSe V-;\1 id:a: kiiijxu- r.n.:v ::l .riiqu-. ::lrji:lan ;n;i J::i?_*v I'hy.wil l.Ht ST.M r ia^v :f a UV/X d.- v: I :> <*. $elivi:ia'.iv J jv am prapL-rlih- scini'sr Ir.u Jittnivry.1 V/I.i.u :.t.Jx> ui Lit i:tt.iu rl ihr Lr;> -V.l'y r! if r.nir ihe nrr.crkat rlilc <’NT:;: .1 nzt- n- <«ith ;lki.-“|v.rV/N'ls; Jiavv ":n;ui MiiL/il b; my ay wOiavu. fvi IIO^V; vj|\ps::im:a:"py. lor n.vi . fy.aUoi :lv ,ph : ri. a . p. a r# cjm bit;:! Lxp.-.imviUl vin.i ixc 11: .1 :hv rail .a1 krv.r.l: ry ni"::.x axx:L:a%,l will: :»ol i nut. .ski r.il. -JV.SS- iv-viivi.i l I It- pivfik v!\.:v C*v: h vi -.-><> 1 |v:t :: LNI r.ibc?..' ' TI.V ylil-vlvr.ll pr; >chk:* CJ IWNTt *x:av k. tlccuoa Jifrvc*:*. t »«.y rirfiacr.cr*/ ur.l rhcoreiicy irulvul.: ".'-ik i! :;|-::v.ri hai H r- _yti -li.v.-.ikt iIk ; I >Vv p41 ^ iv i i.l i - f: c nrr .irH nil* Imi - pmp^nic < iir- pi.u.hii.y ioi ik .irc ^ciiti: • v rue. v.Hv. '.v.vVS. L'al . I lilt vlmiliiv I In. Jcx.r lcA viupî- lit~ Jtp-.aixal J:i Uif vnu.ti axlivvS wait C .zCpOf(CaJ.v Uvx;i'. s:: lima- itanvlirg .v.ivrvy.^ny (6.Mi mvvs:;^- Uixis i. SW\ y ai:J MW V s I jvv :vv\.dlt*J a JriCv. ivlalimi b:.'.v«.i I he:: .:u:r.vint* xL.iciLit uia1 cl: ixpcr. ^ I'M ir.cns-jvc- 3i%.ila i)l XV/N Is ..i:J DV/M, U.T.ini.ulim: i:i W.l: S V/.X i> uul JWN IV Cxjimavt fu; a.lu.1 J ai:ivUi> >».K .iC;^ «i<. vj icl* wcic o.'U.i i?V'/> T> .vn>: t\i îAiivf.v J i<.ia**>1 The sw.\ “ ur.l .1 v:\17 ym/ % V.ViV SVUnA: 11 1 : tU 5# V J -t.,;/.v A‘y l )J. II. ; ;l»-|^V.Vt L \ > V,.'A: Ca-.'j: i:l ci.'i; I aw ml f). .u» ..v. Uv: c -::nlcl *.» : ~ nk- |)vf>tt -.V'L'lii i •. .!> CCeor-.M j ki^n v ui:Oi:».Cf l yr l I ; c«;r-!:i:e HO/C.i a r \e* = co*^ nor.: ir J: >. m ir>> .vjiuut.iU. Vxa:A 1 1 .- S SVVW |.x %i J ./A ? |> 1 .vJ vi ; U; 1 it- v.vi 1 v 1 v>; ^ v.i SI ^ 1 ai.i>vv. vvi .1 -

Lv:;.- r.S!*Ai:n>. S.ik: li:SL jiiiv: .f ! it.... i- .1 / vv:v .v,..i|. l"« I ;**!•: :!•>■• villv ic. • .-.il STv •• T.%'.Vr 1,.^. :!< nievh:.n -.:.l.y p.v.J-. J .:p;;. l ie n^'.v -» r.,c v:«i ..i:ve«! .. < *\:?\ i.y . SI ': In .-oin 1 .• :n»*l.n. tuiy. «t .1- w ilh .1 C.I 'VjiiI r .;vI:* Vv> *i. r in |l.”i:» * . In «•: • ; t :i*.*ti ,"*% . «li: .M'fi 11: .10 •• f x I .- III.1 z:. 1: Ian: k hi *: e'i .r 1 lli. iivlv.;. •" i oil *»:•.•> . ui i.i Ik —.'!:H x *-o .11. i \ :• • «t ill III. •e.tIt* , .1 .' nJ A:: V >.<■ a.J|-c,:;U ^l.« 1;:.... I n»uv. wi !•»: :!:!• • •> ) X Ivin. *• i 1 :• all »iv . lit. p in •ui. 0. • ■ l .> . v . .Ii v.li.u

ux.. Ills:..v>::;r.v^ . Cl % r.ti t*\w *• f -V tiyi.xr. P ,y#..y n.z' &r::jp: bi.1d.1U U.oeer "I.) CM u..*J he HOl<* auxlit.k. .i: viiei re nvn rv.io :bi.: vffee:. W«:b-rc50ln»»or images 'êie nken wi't Ihc Y‘.ir ciny roo n;m!> |vrf* lb\ *5 +c OjT -x. T <7 W IK rtrr filing | ivâa ihc y M !i|: ;,\1 U.VJs w.H: J ir.nxoi •c-.-si. dcns.iy uf ■ 6-fiavFUi-y |l fc. i..;* ceeuMtic sir.KCure.- '* 7he :ia:ui!LiK k!W£C Vzil.i | lie r.iiii.oliri;; biu$ lOI-vfOt ll‘C li > on* :!• C NT VI ne ;f:X4G viibr^.c/ 1' A01 dirh*n-v r.H icl r; «;ipK b'- t'.vm do f V"t aor .he .tOPC r.irveh-.-.i' ^ricuVy S:ripvr llte ii:uxl C M lyu:v nr a "dO'.'C <=u;;fr.ui^ ie. :he .< r fia:>e A iU h: wlc *> = der.xec a< ;hc .= lLW.ivi. jKi^. I ill] II:.' V0 i>ic»ae:.1 i):e cf a CMl if. ô'.- cizILy ir.divau-J by ihc ia:i) c f lie poq:c.idiûlni n >J ho;*.- a. D" (nm) /Au.iai Icnciha. •:.-/!•>!. whc:c r ic rho CNT indium ai d n is the liinre.l'n^ h-i\vr Kxa.iri.e, I IU. '.m *: tv -nUii.; n ?ki ••>!. p. rvjrimhr *»0i:: •• fy 'h. ?."."pl:*. C ' ^v.,z*. I; ; «4 .'VzN15 . *c VJjh .t lix i**m pW? •* n»n .h: .i Ivpio ’ li.!ir»Kliri* ;cî> lviv.ru- 1 i)k STM lip a:ic Inc C N'I ::: Vh l rzlii .s*z' 1I1 ;-•> * •nun v*» !>-' 'v. (•'».*5 nil >/• mi U.*1 ;ir i ai i:l .1.; iulv iaJivv is V* V ii i:. U C !nu ne :S d;iU-au ill 1 |.v «nl.' 1 Ui:.<‘:.„ile c.*tl..*.* khlui x«* j /Hid: TV/Ml', laws. ' «C by * 60%.' VrVoi! liiC :.p '3 s?V,.-n ) OnOi-l^ ’(• Own'll ncu.ly .X>* . "Ih. IWN!’, v l" . ft % I i*:;iAUf 2i.X*i/. U. l>. ll lb: .' C* il:v a*'.,Ai .1 a»..'.lor or" i K dS\*1 ’. hr i i:i r c-ôrx n I'.pir ..•'tv X ** v d>ib::l in: c; 12.Xml. !>»•"*• ^1 Tei.qh.r m fr:pi -h:i runiriic:.; dzl..it:u. ::-. nl* .l v CT4T.-. T> is it V. in gevr.!l ml pjfsy Im nrUiri lie rxa ;-ia*r {\ va iCA n MU . -.-c C^K’ I -.Y ho A ""!.r :n:nt -hr l’.li :ad l. «.it o: i:K C \ I a.I.Li dcatAiiioi' or. :h.e :ob>i.<.ic ii if nccc^.y v :..V;id.'r ihc -ir,e c 0 n ci- f:h' ir. '"Ciitiic u «T-' •' e*-":. 7ho ra-.ioK o' •«'*? z:r Avoai 15 nol V a/W (f S.tr • Yf % -o |V ::r.i /i are nl«n .x-hia' d nnmr .nr thr ’*.«:spl nl .he i'WTx b> rliin?. " V r.i i.;s ' nr .'?/ « :;n L'o.r rnfvtcr.l .if da by fill r.:i rv ho.k h* lh.y in«s la:mr ia*; I'simsi! yiily li> Mic <*ST iri lleîvi. ' and va y »liO <,iee of lie ON" ielorn.% an "fjirr.ft *..• i.e* ?*r.:.l .-r ha a : IIi :a.h*;li:iU r. Ayjtir, !. •> hee.ar .i: Juîvi. - ,1 ll‘>vS ax'ic v ;lui rhe CNTf KC <,e:ornf(i jy hr . r o.-.r V/a;.lx I ho- CM .nv lae I IO?(‘. uihxlra'.e %: lie tp.no Uii.; II:*. for:e b.;> eer he iuVo m>d rho -10^0 s.ib^rato luaixiia^ ^:.p IvlwtCa llie l.p Oi.J CNI* i* kepi C0is:aoi l iî Z 2 sho-.r; *hc ., f.ld h vr.I.lCJ ell ir.jJ f\nn IS:: wlaiv :hv y iNi a ; .^1 i Koîvh e *irJ :i cenvoi.r n.êrr 7ho rV.i.i^' poeedi «*v tej ml-cvr Ii $ -rei iW ib% c vrhie Angle 0 v.'i'l "hen rvprftnai hninc- decs 101 iK'C^sr I rC-'/ly xv Ih the vnlar In vz-.v :d'l ve tries $ain nuily Hk :l w\ ij’ias:i«f Lie SlM U.i VieSely iC.ICCCS iI'C x i ia rizfTftY- or'defiirr.ii'd CNT Th*r. .1 vci d:a: Uio e-jiv. u-ai.of . ie Cn f tvhilc :.io rip iy i: oc node $ V *> lI'V vr%$:-seeiiO'V I deferra.rinr indi.:..t:l iy hr v. a dar A.vd:: aysuiaev la: “ icpreYen .alive yl i|;e vz ae-lc s wpx elb mCC :::.via*:l!«xiv b.e.vev;. .lie :. ii.ei-.46v> ii by 11 i Jfy n:»ir.L' iv c(;.i li::«% o***he S. M »i ^ evrcsiponc- I>iapeI- l e J: iixelvi v. CMS. bog..ks. :l e Sv.*NJ< my .a v. zrz -.N h^k : h.i v c ar. be m-nr<.MKX .„«S I: .ve 'T^ :,Sv 1 vi I rhe f)Yvh I: lalvU I .lie yevi: i > g jfM "I**, vv •'- • esi iVce . r rl:( i ll if.:'i Iiniai ik: ;y%VNT-> .ine Ornvi. L*v.< a: ihv M.::.Mnik- r V to »h. 1 i < hi.Yd f ,fT •«. vram.r. rhr a i: f.iixm «r i u:::i aa.iKin'.»ill: . io SWNl- |\% i'. :i %i> re.l.vc y:n ~ ::|A ;*:xi:; a..l «lu itij- Ire * e.miiiaj ir Lie i:V":i rv:-.i : Ol.vi.'u: *y it«y.v.s rhg ihe aC j iic*.t of ihr. Tier > *• c:i* 6 au* ii".. CN I* I:ax u i vili.*.:. : au x Jiflintid »y i re e. -C :if y<: I i'.*?s in I ir V 11S« • e riY. :or I ron: f e :0: .e i.y f.'.v % "< vid; i h* oe i .silver, fi e>v ;• »s n lekm-Tinv. . ifvj.Hi !i: i.i<, v.-: i ::l:v de ce. > i,o . I" h.- ;* r- .1' •• spur fi.i'iaa 2i " >::•••< i X:italic : i: I. ,‘i. :.* .a d .V v.lje> aa ; eiv’.vs: ii* I :lb . err. ve i.c : : vv.i:x:r *:on lii. iy :!T iv j*-:nz :!.*.* • i:ij:le lUuiivn o* / e> ' J : a hi; .:efiim*.!l .-TV in re v-hc-r.:. !>: * .: i- •: / ■< I';." l:: I :: vka I ii at. v.v ivf.fi/: ».:J. ' .W •'*' •:*J.ie i C. .% is v if"nr:* I n iv* I I i/ )i \l. : ia .ho v i lira i'iv i. Iuh i j _a ii nll-eav.i). i .|»r • - •• j : ,p ; ,b- A.v. !> v: •/.**:. r:l ...idl e d < .id ;l d::*»*.f*"V.Iio .1 to ii*.:: x. • S' nr" ai v cm*, b? Si xiier. v/n we e* nr\ o he a'x :;li .o :»vp: are ■ '••,*.•. i i : :l v ■. il _.v i : i I •:: : a . -: M i. I nr i -kc 0MvdnK.il .1 -r ed sVi IT: . vine v I ,a" ! ::r::. i i ; .:!:/• ..".* . i i.: • l.tvor z. vi i *. ; rro !•• ;. W.'bC :x * i^l . •/ lei . tar. llu i •>! i ue :o x: . >;• it.: . i rii.il;* 'i .• ....: .!• 10a :v; c 3 ; :...iiei vif<: • .e. .: -Id: V : (Cfvtir I III ': . vi'l 0. VIIVI Ml . •- l l I V A' :|;|« ::d--.0 y : ie •... .kr . !:• ', *..% Dov,nVJUilfc '* *. Vi mi ' ' . ... 023110-3 Park et ai Appl. Phys Lett, 86, 023110 (2005)

In summary, we have used high-resolution STM to elu hR. Saito, G. Dresselhaus, and M. S. Dresselhaus, J. Appl. Phys. 73, 494 cidate the three-dimensional structure of individual SWNTs (1993). M.-C. Charlier and .i.-P. Michenaud, Phys. Rev. Lett. 70, 1858 (1993). and DWNTs. The actual diameters and heights of CNTs were *R. Saito, R. Matsuo, T. Kimura, G. Dresselhaus, and M. S. Dresselhaus, estimated from a line profile. The cross sectional deforma Chem. Phys. Led. 348, 187 (2001). tion induced by the van der Waals force between the CNTs ’m. Koeiak, K. Suenaga, K. Hirahara. V. Saito, T. Nakahira, and S. lijima, and substrate was investigated for the DWNTs and the Phys. Rev. Lett. 89. 155501 (2002). SWNTs with different diameters. Much smaller deformation "T. VVildoer. L. Venema, A. Rinzler, R. Smalley, and C. Dekker, Nature in the DWNTs than in the SWNTs with same diameters can (London) 391, 59 (1998). mT. Odom, J.-L. Huang. P. Kim, and C. Lieber, Nature (London) 391, 62 be taken to be evidence of the interwall support. Thus, it was (1998). found that addition of the inner walls will give rise to greater A Rochefort. P. Avouns, F. l.esage, and D. R. Salahub, Phys. Rev. B 60, rigidity of the multi-walled CNTs. 13824 (1999). L'J. W. Janssen. S. G. l.ernay, L. P. Kouwenhoven, and C. Dekker, Phys. This work was supported by the KISTEP (National Re Rev. B 65. 115423 (2002). search Laboratory and Grant No. MI02KS010001-02KI901- IJP. Li. Lammert, P. Zhang, and V. H, Crespi, Phys. Rev. Lett. 84, 2453 01814) and the Korea Research Foundation (Brain Korea 21 (2000). I5L. P. Biro, S. Lazarescu, Ph. Lumbin, P. A. 1'hiry, A. Fonseca, .1. B. Nagy, Project in 2004 and Grant No. KRF-2004-005-C00060). and A A Lucas, Phys. Rev. B 56, 12490 (1997). C'.J.L. was supported by the Center for Nanotubes and Nano- "1. P. Biro. .1. Gyulai, Ph. Lumbin. J. B. Nagy, S. Lazarescu, G. I. Mark, A. structured Composites at Sungkyunkwan University. Fonseca. P. R. Surjan, Zs, Szekeres, P A. TIliry, and A. A. Lucas, Carbon 36. 689 (1998) 'S. lijima. Nature (London) 354. 56 (1991). 'T.-X. Zha, R, Czenv, D. L. Carroll, Ph. Kohler-Redlich, B.-Q. Wei, A. ~S. Bandow. M. Takizawa, K. Hirahara, M. Yudasaka, and S. lijima. Chem. Loiseau. and S. Roth, Phys. Rev. B 61, 4884 (2000). Phys. Lett. 337, 48 (2001). I!"G. I. Mark. L. P. Biro, J. Gyulai, P. A. Thiry, A. A. Lucas, and Ph. Lambin. "'j. Wei, B. Jiang, X. Zhang, H. Zhu, and D. Wu, ('hem. Phys. Lett. 376. Phys. Rev, B 62. 2797 (2000). 753 (2003). '"j. TerslT and D. R. Flamann, Phys. Rev. Lett. 50, 1998 (1983). ft. M. Zuo, I. Vartauyauts, M. Gao. R. Zhang, and L. A. Nngahava, Science "h.-X. Zha, D. I.., Carrol], R. Czerw, A. Loiseau, H. Pascavd, W. Clauss, and 300, 1419 (2003). " S. Roth, Phys. Rev. B 63. 165432 (2001). AM Abe, H. Kataura, H. Kira, T. Kodama, S. Suzuki, Y. Achiba. K. Kalo, 2iL. C. Venema, V. Meuntcr, Ph. Lambin. and C, Dekker. Phys. Rev. B 61, M. Takata, A, Fujivvara, K. Matsuda, and Y. Maniwa. Phys. Rev. F3 68 , 2991 (2000). 041405IR) (2003). 2:T. Ferlel, R. E. Walkup, and P. Avouris, Phys. Rev. B 58, 13870 (1998). Available online at www.sciencedirect.com SCIENCE DIRECT" solid state communications ELSEVIER Solid State Communications 133 (2005) 83-88 w w w. el se v i er. com/i oc ate/s sc

Distinct critical fluctuations and molecular motions manifest in a model biomembrane

Kyu Won Lee3, Cheol Eui Lee3'*, J.Y. Choib, Joon Kimc

‘Department of Physics and Institute for Nano Science, Korea University, Anatn-dong, Sungbuk-ku, Seoul 1367 Id, South Korea bDepartment of Computer Science and Institute for Nano Science, Korea University, Seoul 136713, South Korea cSchool of Life Sciences and Biotechnology, Korea University, Seoul 136-701, South Korea

Received 3 July 2004; accepted 11 October 2004 by A. Pinczuk Available online 21 October 2004

Abstract

Lattice dynamics in bis-(n-Ci6H37NH3);>SnCl6, where the hydrocarbon part is analogous to lipid membrane, was investigated by means of 200 MHz 'H nuclear magnetic resonance. As a result, critical fluctuations and molecular dynamics associated with the phase transitions, an order-disorder and a conformational phase transition, were distinguished in a wide temperature range. The dynamical origin of the critical fluctuations, observed in the long-chain compounds but not in the short-chain compounds, by the laboratory frame spin-lattice relaxation measurements, is revealed and discussed in this work. © 2004 Elsevier Ltd. All rights reserved.

PACS: 64.70.kb; 76.60. -k

Keywords: D. Phase transition; E. Nuclear magnetic resonances

1. Introduction (1/3, 2/3, z) and (2/3, 1/3, z), with the alkyl chains alternately pointing upwards and downwards 13], The bis-n-alkylammonium hexachlorostannates (n - In our previous 'H NMR studies of the CnSn systems, C„H2„+ iNH3)2SnCl6 (CnSn for short) are layered com two typical successive phase transitions were observed [4- pounds of alternating organic and inorganic layers. The 7]: (a) order -disorder transition of the rigid alkyl chains SnCl|~ octahedra do not form a 2D macroanion but exist accompanied by a uniaxial reorientation about their long separately [1-3], The NH3 group of alkylammonium ion axes along with the flipping of the polar groups, such as links the three closest octahedra through equivalent NH3, between the potential wells, and (b) conformational hydrogen bonds of the N-H- • Cl type, forming an inorganic transition leading to a partial melting of the alkyl chain part. layer. The distance between the ammonium groups or In the low temperature (LT) phase below the order-disorder between the tin atoms in CnSn is great enough (7.3-7.5 A, transition temperature, the alkyl chains are rigid and adopt depending on the chain length) to accommodate a mono- an all-trans conformation. In the intermediate (IT) phase layer of interdigitated alkyl chains. The alkylammonium between the order-disorder and the conformational tran groups are statically disordered around the three fold-axes at sition temperatures, the uniaxial rotation of the rigid alkyl chain about the chain axis as well as the chain-end trans- * Corresponding author. Tel.: +82 2 3290 3098; fax: +82 2 927 gauche isomerization take place. In the high temperature 3292. (HT) phase above the conformational transition tempera E-mail addresses: rsccl @korea.ac.kr (C.E. Lee). ture, the average chain axis is destroyed and the chains [email protected] (1 Kim). undergo liquid-like isotropic motions. The three phases and

0038-1098/S - see front matter © 2004 Elsevier Ltd. All rights reserved, doi: 10.1016/j. s sc. 2004.10.012 84 K.W. Lee el al. / Solid Stale Communications 133 (2005) 83-88 the two phase transitions are analogous to those in hydrated for which the shorter time constant only reflects a critical lipid membranes. fluctuation near the order-disorder transition temperature Hydrated lipid membranes, layered systems of alternat 14], The spin-lattice relaxation is a single-exponential form ing lipid and water layers, show diverse phases depending at 45 MHz, where the time constant also shows a divergent on the degree of hydration as well as some other parameters anomaly near the order-disorder transition temperature [5], such as temperature [8,9]: lamellar crystalline phase Lc, While the spin-lattice relaxation time of C12Sn at 200 MHz lamellar gel phase Lp, and lamellar liquid crystalline phase does not reflect a critical fluctuation [7], the rotating frame La. In the Lc phase, the hydrocarbon chains are well ordered spin-lattice relaxation time at 55 kHz shows a divergent and essentially all-trans. In the Lp phase, the uniaxial anomaly near the order-disorder transition temperature rotation of rigid hydrocarbon chain about the chain axis is 126]. Thus, it appears that the transitional behaviors characteristic and the chain-end defect can appear. In the La reflected in *H NMR spin-lattice relaxation time are phase, the hydrocarbon chains are highly disordered with associated with the observing frequency (Larmor fre energetic conformational defects [10,11], Regarding the quency). In other words, the dynamics near the order- hydrocarbon chain, the typical phases in hydrated lipid disorder transition temperature appears to be the origin of membranes are very similar to those in CnSn, although there the different transitional behaviors. can exist several sub-phases in the lamellar gel phase and a As observed in ClOSn and C12Sn [6,7], the dynamics rippled gel phase Pp between Lp and La phases. For low contributing to the !H NMR spin-lattice relaxation consists enough hydration, all the water molecules are tightly bound of some molecular motions: three-fold reorientations of CH3 to the head group, undergoing a uniaxial rotation without a and NH3, a chain-end trans-gauche isomerization, and an translational diffusion, and the Pp phase is absent [ 12|. Thus, unknown defect motion. In the LT phase, molecular motions the CnSn system is quite analogous to a low-hydration lipid of CH3 and NH3 contribute significantly to the spin-lattice membrane. relaxation in the low and the high temperature region, In contrast to the case of bis-rc-alkylammonium tetra- respectively. The chain-end trans-gauche isomerization is chlorometallates (rc-C„H2„+,NH3)2MC14 with M = Cd, Cu, characteristic of the IT phase and dictates the spin-lattice Mn and Zn (CnM) [ 13-17], little information is available on relaxation in the IT phase. The molecular motion of the the phase transitions in CnSn 14-7,18-20], The first unknown defect was also observed to contribute to the spin- systematic studies on the layer structure intercalation lattice relaxation in the temperature range between those compounds including ClOCd were made by Blinc and dominated by CH3 and NH3 motions. Near the order- coworkers using magnetic resonance. Measurements of the disorder transition temperature, the molecular motions of order parameters by 35C1 and 14N NQR and *H and ,3C NH3 and the chain-end trans-gauche isomerization are NMR made possible a consistent understanding of phase important because they can obscure the critical contribution transitions and dynamics of the NH3 and the hydrocarbons to the spin-lattice relaxation. using a Landau theory similar to that in the liquid crystal In this work, the chain length dependence of the [13,14], In a 2H NMR study of C4Cd [17], the evolution of transitional behavior was investigated by employing ‘H orientational and conformational orders of the hydrocarbon NMR for C16Sn in comparison to other CnSn systems. The chain was shown to be similar to those in lipid membranes, simple dynamical structure and the chain length dependence although the order parameter tensor was not uniaxial and the of molecular motions will be shown to be the dynamical value in the high temperature phase was two times greater origin. than that in the La phase. Critical fluctuations have been observed in a few studies near the order-disorder transition temperature of a lipid layer intercalated between the 2. Experiment inorganic layers [4.5,21], In hydrated lipid membranes, (pseudo)critical swelling has been intensively studied near The C16Sn sample used in this work was synthesized with the main transition temperature, mostly near the La <-> Pp much care to avoid impurities by the chemical reaction: transition temperature [22-24], and much less frequently, 2(n-Cl6H33NH3Cl) + SnCl4 • 5HzO -> (n-C16H33NH3)2SnCl6 + near the La<~* Pp transition temperature [251. 5H20. After filtering and two recrystallizations, white While ClOSn and C18Sn show the same transition sugar-like crystals were finally obtained and then vacuum- sequence, the transitional behaviors reflected in 'H NMR dried and kept in a dry condition for further works. The spin-lattice relaxation time at 200 MHz are considerably stoichiometry and the structure were checked by elemental different. In ClOSn as in well as C12Sn, the critical analysis and X-ray diffraction (XRD). Differential scanning fluctuations were not reflected in the spin-lattice relaxation calorimetry (DSC) carried out between 123 and 453 K [6.71, but were observed in Cl 8 Sn near the order-disorder shows two reversible phase transitions. The low and the high transition temperature [4,5 ]. In C18Sn, the spin-lattice temperature transitions were accompanied by enthalpy relaxation pattern near the order-disorder transition tem changes of about 0.5 and 10 keal/mol corresponding to the perature depends on the Larmor frequency. At 200 MHz. the single gauche formation (a chain-end trans-gauche iso spin-lattice relaxation pattern is a double-exponential form. merization) and the chain melting, respectively [9], The line K.W. late el at. / Solid Slate Communications 133 (2005) 83-88 85 shape and the spin-lattice relaxation time measurements Fig. 2 shows the FWHM (full width at half maximum) were made using 200 MHz *H NMR in the temperature linewidth as measured directly from Fig. 1 as a function of range 150-400 K. The spin-lattice relaxation time was temperature, in which two phase transitions are apparent measured by the conventional inversion recovery method, each at about Tcl — 329 K and rc2 = 355 K. The linewidth and the line shapes were obtained by Fourier-transforming shows a continuous and a discontinuous change at 7cl and the FID (free-induction decay) signals. 7c2, respectively. The continuous decrease at Tc] is accompanied by the diminution of the dipolar splitting as shown in the inset of Fig. 2. Through the transition at Tc2 the linewidth shows an abrupt drop and an extreme motional 3. Results and discussion narrowing. The extreme narrowing of the linewidth above Tq 2 indicates a partial chain melting due to the activated Fig. 1 shows the evolution of the NMR line shape in chain defects such as GG conformer (two successive C16Sn with increasing temperature, which is similar to that gauche-bonds). in ClOSn and C12Sn [6,7]. At low temperatures it consists The spin-lattice relaxation of C16Sn showed a single of a relatively narrow component and a broad one. The exponential pattern in the low temperature region. However, broad line can be mostly attributed to the rigid part of the above about 270 K it was nonexponential and was well fitted long alkylammonium chains, corresponding to the methyl into a double-exponential form with two distinct time ene groups, as the chains are supposed to be well ordered at constants, T[L and Tis, corresponding to the longer time low temperatures. The narrow line is attributed to the mobile constant and the shorter one, respectively. Fig. 3 shows the part of the alkylammonium chains, presumably correspond spin-lattice relaxation rates as a function of temperature. As ing to the methyl and ammonium groups, whose fast reorientation averages out the dipolar field, resulting in the observed in C18Sn [4], Tls is believed to reflect the critical sharp line. The low temperature line shape was well fitted fluctuation around the phase transition. The variation of the into a dipolar powder spectrum (Rake doublet) and a fraction of Tls, shown in the inset of Fig. 3, supports the Lorentzian line [21 ]. The Lorentzian line corresponds to the supposition, as the fraction increases approaching 7cl. reference frequency and the dipolar powder spectrum shows The two 7] components do not directly correspond to the two symmetrically separated shoulders about the central two superimposed peaks in Fig. 1. In fact, while Fig. I Lorentzian line. The separation, which is ascribed to a shows the two separate line components at low tempera dipolar splitting of the rigid hydrocarbon chain, gradually tures, the spin-lattice relaxation shows a single-exponential decreased with increasing temperature. form with a single T, component. Furthermore, while the intensity fractions of the two line components remain almost unchanged with varying temperature, below the TC1 Tc2

100-

Temperature (K)

-80 x -60 a -40 --a

■0 Q

-200 -100 0 ' ioo ' 200 Temperature (K) Frequency (kHz) Fig. 2. The FWHM linewidth as a function of temperature. Inset: The dipolar splitting obtained from the line shapes as a function of Fig. 1. Line shape evolution as a function of temperature. temperature. 86 A". W. Lgg gf a/. /SoW #3—##

Tci T, flops of the NHj polar head (about the three fold axis) and the gear-like collective reorientation of the alkyl chain around its long axis [4,21,26.27]. Since the chain reorienta tion is believed to occur in the frequency range typical of the 0.6 dipolar interactions, i.e. in the kHz range [21], the high- frequency critical fluctuation observed in the 200 MHz NMR spin-lattice relaxation is believed to be the collective flip-flops of the NH3 polar head. Because the NH3 polar 250 300 350 head is known to spend its time between two favorable sites Temperature (K) in the disordered phase [3,28,29], we have employed the kinetic Ising model in order to check the conjecture, satisfactorily describing the experimental results [4,27]. As in C18Sn, the kinetic Ising model was introduced for TTS' [4]: □ T rrs' = a ,M‘",-j-t(i -|/|2/3), 7<7; (1)

7,1' = A2|rr'-A-\ 7> 7C (2) Teimperative (K) in the limit iOT{q) = 0.63, y= 1.25, tions of the Ti components show a marked change around d= 3, and [7 = 0.33, A = 1.23 + 0.05, A] =0.23 +0.05, and A2 = 0.28+ 0.02 were obtained from the fit. A dynamic 7cl as displayed in the inset of Fig. 3. Thus, it is evident that scaling prediction for A = is 1.26 with z = 2 [30,31 ], and a the two T\ components do not have a one-to-one vz high temperature series analysis suggested that J = 1.32 + correspondence to the two superimposed peaks in the line 0.03 [32], Indeed, the value of A = 1.23+0.05 obtained is shape. quite close to these values. As observed in Cl0H2iNH3Cl A double-exponential type of the spin-lattice relaxation (C10C1) [21], which also has an interdigitated chain pattern is interesting for a strongly dipolar-coupled proton configuration, the three-dimensionality of the collective system, where spin diffusion is expected to give a single critical dynamics indicates that the interlayer interactions exponential pattern of the spin-lattice relaxation. Thus, the are appreciable in the order-disorder transition of the proton system of CnSn may be supposed to consist of two hydrocarbon chains. distinct spatially well-separated regions. In other words, Our power-law analysis was done over a wide tempera spins in a portion of the system undergoing critical ture range, about A/= 0.06, appropriate for a mean-field fluctuation are spatially separated from those in the rest of description. Nevertheless, Eqs. (1) and (2) with the mean- the sample so that the spins cannot communicate through field values of the critical exponents, result in 7^' vanishing spin diffusion near 7C,. In fact, the increasing fraction of T[s as the temperature approaches 7C], in contrast to the approaching 7cl is consistent with an increase in the critical experimental data. Including the results for C10C1 and region. If a molecular motion dominates the spin-lattice Cl8 Sn, the three-dimensional values of A obtained from the relaxation even in the critical region as well as in the normal power-law analysis are scattered between 0.9 and 1.6 [4,27], region, the spin-lattice relaxation is a single-exponential which may result from the wide temperature range used for type. If a critical fluctuation dominates the spin-lattice our power-law analysis. Nonetheless, it should be taken into relaxation in the critical region, the spin-lattice relaxation account that our aim is not an accurate estimate of the curve becomes a double-exponential type, whose short- and critical exponent but is a check of the Ising-type of critical long-time decays reflect the critical fluctuation and the fluctuation. The critical fluctuation reflected in the 200 MHz molecular motions, respectively. In other words, if the NMR spin-lattice relaxation is believed to be the three- contribution to the relaxation rate from the molecular dimensional Ising-type, which can be assigned to the motion is greater in the region of the critical fluctuation, the collective flip-flop motion of the NH3 polar group. observed relaxation rate for both of the populations will be The longer time constant component of the spin- similar and a single-exponential decay will be observed. lattice relaxation below 7cl was well fitted by the In previous works, we have proposed the possible intramolecular dipole-dipole interactions modulated by candidates for the critical fluctuation as the collective flip- various types of molecular motions following examples K.W. Lee el al. / Solid Slate Communications 133 (2005) 83-88 87 in similar systems [33]; contributions from CH3 and NH3 are very small in the temperature range investigated, and a main contribution to a ,. 4 Td r the spin-lattice relaxation comes from an unknown defect, 1 + (WTC,)2 + (2wtc,)2_ (3) / J whose second moment and activation energy are 1.2( + 0.2) G2 and 14( + 3) kJ/mol, respectively. The unknown defect Tc, = = ,2...... n). (4) motion should be fast enough to dominate the spin-lattice relaxation at 200 MHz and have an amplitude too small to where y is the proton gyromagnetic ratio, M2 the second affect the dipolar splitting. A possible candidate for this moment, w the Larmor frequency, and E is the unknown defect motion is a small-amplitude torsional activation energy. Three different types of the molecular motion [17], While the activation energy of the NH3 motions (n = 3) were introduced, and the results of the motion, 80 kj/mol, may be unreliable due to its large error, fitting according to Eqs. (3) and (4) are shown in Figs. the activation energy of the NH3 motion can be taken to be 3 and 4 as a solid line. very large, indicating that the NH3 motion contributes little Fig. 4 shows the measured (symbols) and the fitted to the spin-lattice relaxation around Tcl. In practice, the (lines) spin-lattice relaxation rates corresponding to the large error simply originates from the insignificant contri longer time constant, TlL. Data below 320 K were used for bution to the spin-lattice relaxation. As discussed in our the fit. In our previous work for C12Sn [7], the activation previous work, the activation energy of the NH3 motion energy of NH3 was shown to be strongly dependent on the appears to be strongly dependent on the chain length [7], chain length. In order to study the chain length dependence Now, we are able to discuss how a critical fluctuation can of the CH3 and NH3 molecular motions, the second dominate the spin-lattice relaxation in C16Sn but not in moments of CH3 (2.5 G2) and NH3 (3.1 G2) previously ClOSn and C12Sn. In ClOSn and C12Sn, the contribution obtained from ClOSn were used, and an unknown defect from the NH3 motion dominates the spin-lattice relaxation motion was introduced with free fitting parameters [6], The near Tcll and apparently the critical fluctuation is not results are shown in Fig. 4, where the dotted, dashed, and reflected in the spin-lattice relaxation. In Cl6Sn and C18Sn, dot-dashed lines describe the contributions from the on the other hand, due to the high activation energy of the molecular motions. The activation energy of the CH3 NH3 motion, the critical fluctuation can dominate the spin- group is 10, 11, and 7( + 3) kj/mol in ClOSn, C12Sn, and lattice relaxation. In fact, as shown in the inset of Fig. 3, the C16Sn, respectively, which values are not much different. critical fluctuation component accounts for most of the On the other hand, the activation energy of the NH3 group in spin-lattice relaxation around the phase transition tempera C16Sn is 80( +60) kJ/mol, which is much larger than 40 kJ/ ture TC|. Of course, this situation may change at a different mol in ClOSn and 55 kJ/mol in C12Sn. As shown in Fig. 4, Larmor frequency. While CnSn shows a typical transition sequence for all n> 10, the transitional behaviors, reflected Tc, TC2 in the spin-lattice relaxation, are quite distinct, which is attributed to a dynamical origin. In effect, the relatively simple dynamic structure of the CnSn systems, consisting of CH3 and NH3 molecular motions and some defect motions, provides an opportunity for the study of critical dynamics in model membrane systems. In summary, two successive phase transitions in C16Sn were investigated by means of *H NMR. An order-disorder transition and a conformational transition were character ized by the line shape evolution and the spin-lattice relaxation. Around the order-disorder transition tempera ture, the spin-lattice relaxation was well separated into the molecular motion and critical fluctuation contributions. The distinct transitional behaviors, depending on the chain length, were found to have a dynamical origin. Our work demonstrates that the simple dynamical structure of CnSn systems can provide us with a valuable opportunity for the study of the critical dynamics in model membrane systems.

Temperalure (K)

Fig. 4. The longer time constant component of the spin-lattice Acknowledgements relaxation fitted by Eqs. (3) and (4) (solid line). The dotted, dashed, and dot-dashed lines correspond to the contributions from the NFL. This work was supported by the KISTEP (National defect, and CFL molecular motions, respectively. Research Laboratory and M1 02KS010001-02K190 1 - 88 K. W. Lee el at. / Solid Slate Communications 133 (2003) 83-88

01814) and by the Korea Research foundation (Brain Korea [16] S. Jurga, K. Jurga, E.C. Reynhardt, P. Katowski, Z. 21 Project in 2004 and Grant No. C00060). Measurements at Naturforsch. 48 (1993) 563. the Korea Basic Science Institute (KBSI) are acknowledged. [17] H. Zhang, N.P. Kulshrestha, S.E. Woehler, R.J. Wittebort, J. Chem. Phys. 105 (1996) 2891. [18] J. Kroupa, A. Fuith, K.J. Shenk, H. Warhanek, M. Ceramak, Ferroelectrics 159 (1994) 109. References [19] M. Ceramak, F. Fuith, P. Vanek, J. Silha, J. Maikova, Phys. Status Solidi B 182 (1994) 289. [1] O K. Knop, W.J. Westerhaus, Can. J. Chem. 58 (1980) 270. [20] H. Elleuch, M. Kamoun, A. Daoud, J.M. Reau, J. Senegas, [2] K. Kitahama, H. Kiriyama, Y. Baba, Bull. Chem. Soc. Jpn. 52 Phys. Status Solidi B 214 (1999) 141. (1979) 324. [21] K.W. Lee, D.K. Oh, C.E. Lee, J.K. Kang, C.H. Lee, J. Kim, [3] M.H.B. Ghozlen, A. Daoud, T. Molk, H. Poulet, M. Le J. Chem. Phys. 117 (2002) 8004. Postllec, N. Toupry, J. Raman Spec (rose. 16 (1985) 219. [22] R. Zhang, W. Sun, S.T. Nagle, R.L. Headrick, R.M. Suter, [4] K.W. Lee, C.H. Lee, C.E. Lee, J.K. Kang, Phys. Rev. B 54 J. F. Nagle, Phys. Rev. Lett. 74 (1995) 2832. (1996) 8989. [23] J.L. Lemmich, K. Mortensen, J.H. Ipsen, T. Hpger, R. Bauer, [5] K.W. Lee, C.H. Lee, C.E. Lee, J.K. Kang, J. Chem. Phys. 104 0.0 Mourisen, Phys. Rev. Lett. 75 (1995) 3958. (1996) 6964. [24] P C. Mason, B.D. Gaulin, R.M. Epand, J. Katsaras, Phys. Rev. [6] K.W. Lee, M.W. Park, C. Rhee, C.E. Lee, J.K. Kang, E 61 (2000) 5634. K.W. Kim, K.S. Lee, J. Chem. Phys. 108 (1998) 3019. [25] P C. Mason, J.F. Nagle, R.M. Epand, J. Katsaras, Phys. Rev. E [7] K.W. Lee, C.E. Lee, J. Kim, J.K. Kang, Solid State Commun. 63 (2001) 030902. 124 (2002) 185. [26] K.W. Lee, D.K. Oh, C.E. Lee, J. Phys. Soc. Jpn. 72 (2003) [8] M.J. Janiak, D M. Small, G.G. Shipley, J. Biol. Chem. 254 2398. (1979) 6068. [27] K.W. Lee, C.H. Lee, C.E. Lee, J.K. Kang, Phys. Rev. B 53 [9] R. Koynova, M. Caffrey, Biochim. Biophys. Acta 1376 (1998) (1996) 13993. 91. [28] R. Blinc, B. Zeks, R. Kind, Phys. Rev. B 17 (1978) 3409. [10] P. Meier, E. Ohmes, G. Kothe, J. Chem. Phys. 85 (1986) 3598. [29] J. Seliger. V. Zargar, R. Blinc, R. Kind, H. Arend, G. Chapuis, [11] R.M. Venable, B.R. Brooks, R.W. Pastor, J. Chem. Phys. 112 K. J. Schenk, F. Milia, Z. Phys. B 69 (1987) 379. (2000) 4822. [30] B.l. Halperin, P C. Hohenberg, Phys. Rev. 117 (1969) 952. [12] S. Konig, E. Sackmann, D, Richter, R. Zorn, C. Carlile, [31] B.L Halperin, P.C. Hohenberg, Phys. Rev. B 10 (1974) 139. T.M. Bayed, J. Chem. Phys. 100 (1994) 3307. [32] Z. Racz, M.F. Collins, Phys. Rev. B 13 (1976) 3074. [13] R. Blinc, M L Brugar, V. Rutar, B. Zeks, R. Kind, H. Arend, [33] (a) C.P. Slichter, Principles of Magnetic Resonance, Spir- G. Chapuis, Phys. Rev. Lett. 43 (1979) 1679. inger, Berlin, 1990; [14] R. Kind, S. Plesko, H. Arend, R. Blinc, B. Zeks, J. Seliger, (b) A. Abragam, Principles of Nuclear Magnetism, Oxford B. Lozar, J. Slak, A. Levstic, C. Filipic, V, Zagar, G. Lahajnar, University Press, 1983; F. Milia, G. Chapuis, J. Chem. Phys. 71 (1979) 2118. (c) M.A. Lambert, A. Rose, R.W. Smith, Advances in [15] J. Fenrych, E.C. Reynhavdt, S. Jurga, K. Jurga, Mol. Phys. 78 Semiconductor Science, Pergamon, London, 1959(p. 49, (1993) 1117. for a book reference). 1118 ./. Phys. Chem. B 2005, 109. 1 1 18-1 124

Layered Copper Hydroxide /z-Alkylsulfonatc Salts: Synthesis, Characterization, and Magnetic Behaviors in Relation to the Basal Spacing

Seong-Hun Park* and Client Eui Lee* Institute for Nano Science and Department of Physics, Korea University. Seoul 136-701, Korea Received: July 13. 2004: In Final Form: October 11, 2004

A series of hybrid inorganic —organic copper(ll) hydroxy zi-alkylsulfonate with a triangular lattice, Cu2(OH)3- (C„H2„ i |SOj) (;? = 6, 8 , 10), are prepared by anion exchange, starting from copper hydroxy nitrate Cu 2(OH) j- NOj. These compounds show a layered structure as determined by X-ray diffraction, with interlayer distances of 14.3 — 34.8 A in alternation with inlerdigitaled hi layer packing. Magnetic properties have been investigated by means of dc and ac measurements. All the compounds show similar metamagnet behaviors, with a Neel temperature of about 11 K. A subtle difference in the ac magnetic susceptibility among the compounds is understood by the existence of hydrogen bonding between the sulfonate headgroup and the hydroxide anion. A detailed molecular structure of the alkyl chains incorporated to the inorganic copper hydroxide layer is also discussed from the FT1R data.

1. Introduction and sulfate) on magnetic properties, in particular, the magnetic In recent years, inorganic —organic (I/O) materials' have been properties of materials involving coppeifl I) ions have been a subject of great attention because of the wide variations in known to be very sensitive to any structural modification. For the structures and molecular interactions and because of their example, the n-alkyl carboxylate derivatives7 with bilayer extraordinary ability to combine synergistically the properties structures showed anomalous behaviors going from antiferro- unique to purely organic or inorganic materials. Sustained magnetic (AFM) to ferromagnetic (FM) states with varying layer interest in these materials can be attributed, in general, to a vast spacing; in contrast, the analogous /i-alkyl sulfate derivatives'5 array of possibilities they offer in tailoring material properties with monolayer structures were all antiferromagnets. As a result, simply by the correct selection of functionalized organics and it is expected that large variations may be observed in the variously coordinating inorganic precursors .2 In particular, the magnetic behavior of the exchanged compounds. field of inorganic —organic materials dominates the area of As regards the alkylsulfonalc intercalated compounds, the molecular magnets .3 examples are scarce in the literature. M2(01 IbtCTjTTsSCDTTO The majority of these layered I/O materials are prepared by (M = Cu, Ni) and Vv id I •■!(' .11 • S< i.i -'i 1 < i are the only selective intercalation of organic moieties into a layered, examples known so far.1' The copper compounds exhibit a short- inorganic host. Examples of this type include classes of materials range antiferromagnelic interaction, whereas the nickel com obtained by intercalation of ionic/polymeric molecules through pounds arc ferromagnets and the cobalt compounds are ferri- specific electrostatic or ionic interactions into layered host magnets. On the other hand, in our previous work, long-range lattices of clays and silicates, double hydroxides, and metal magnetic order, with both antiferromagnelic and ferromagnetic phosphates and phosphonatesT In particular, there is consider interactions in copper-based alkylsulfonalc anhydrales, has been able interest in anion-exchangeable layered compounds, which observed .10 To discuss the magnetic behavior of such com appear well-adapted to favor covalent links between the organic pounds in relation to their dimensionality, a complete series of and inorganic constituents, because the anions are bonded to alkylsulfonalc salts with different basal spacings is necessary. the metal ions and form the framework of the crystal, as in the In this extended work, we intend to report a detailed study case of the Botallackile-type layered compounds, M2(OIT)jX' for the synthesis of large molecular species based on sulfonate zHnO (M" = Co, Ni, and Cu; X = inorganic or organic anion ).5 ligands and characterization of the chemical and magnetic These systems possess positively charged magnetic sheets properties of the series of layered copper(Il) hydroxy salts, Cu2- formed by layered triangular networks of metal ions that arc (OH):,(C„IT„i iSOf) with n — 6, 8 , and 10 (denoted hereafter interleaved by an anion X”. A remarkable feature o(These series as Cut.AS-//) In particular, it is necessary to understand the is that the intercalation of large organic species enables us to diverse magnetic properties of the copper-based hydroxide modulate the interlayer spacing over very large distances and magnets (TiTOlIhX-clFO with different anion functions in hence to tune the magnetic properties depending on the interlay er relation to the basal spacing and alkyl chain packing. The chain separation/' Therefore, these systems appear appropriate for the packing and conformation of the //-alkyl chains are discussed design of TO hybrid materials with outstanding magnetic from a combined 1 R spectroscopy and X-ray diffraction study. properties. Many efforts have been devoted in recent years to studying 2. Experimental Section the role of the intercalated anions (carboxylate. dicarboxylate. 2.1. S\ nthesis. A series of the copper(II) hydroxide-based * Corresponding author, h-mail rscc!n/l,orca..iv.ki . ’ Present address Nano Material Team. Korea basic Science Institute system. Cut \S-». were synthesized bv an anionic exchange (KBSfial Daejeon. reaction starting from the layered copper! II) hydroxy nitrate

10.1021 ip04o902u VVC: $30.25 f: 2005 American Chemical Society Published on Web I/TO 2004 Layered Copper Hydroxide ri-Alkylsulfonate Salts J. Phys. Chem. B, Vol. 109, No. 3, 2005 1119

Cu 2(0HHN03 (Cu -NO j). The parent compound was obtained as in a previous report .11 The anion exchange reaction used was very simple: the parent material (0.2 g) and the corresponding sodium /i-alkanesulfonate (0,1 mol) were dispersed in distilled water. The mixture was stirred in an air-free round-bottomed flask for 48 h at room temperature. The shiny blue powder was a then filtered, washed w ith distilled water and ethanol, and dried ■S in a vacuum at 35 °C. 5 2.2. Characterization. Elemental analysis for carbon, nitro * S gen, hydrogen, and sulfur was performed at the Seoul Branch of Korea Basic Science Institute (KBSI). The powder X-ray diffraction data were collected on a MAC Science diffractometer (MXP3A-HF) operating at 40 kV and 30 mA in the Bragg — Brentano BZ20 mode (Cu Kcxl, 1.5424 A). The X-ray measure ments were done on naturally oriented samples on glass sample holders. 10 20 30 40 50 60 70 80 The FT-1R spectra were obtained using a FT1R Bomem 20 /degree Michelson spectrometer. Each sample was cast on a KBr pellet Figure 1. Powder X-ray diffraction patterns of (a) Cu2(OH)3NO, and measured in the transmission mode from 500 to 4000 cm"' (parent compound) and (b—d) Cu2(OH)3(C„H2„+1SO,), where n = 6, with a resolution of 4 cm™1. The magnetic measurements on 8 , 10, respectively, showing the shift of 00/ diffraction lines. the powdered samples enclosed in a medical cap were carried out using a Quantum Design MRMS-7 SQUID magnetometer. TABLE 1: Refined Cell Parameters and Basal Spacings for the CuLAS-/! The temperature dependence of the static susceptibility was examined in the range 5 — 300 K in a magnetic field of 5000 basal Oe. The hysteresis curves were obtained at 5 and 20 K in applied sample a (A) 6(A) c(A) (3 (deg) spacing (A) fields up to 7 T, after cooling with no applied field through the Cu-NO, 5.598 6.085 6.930 94.75 6.96 magnetic transitions. The ac susceptibility measurements were CuLAS-6 5.588 6.074 17.813 106.03 17.1 CuLAS-8 21.614 20.1 performed in a 1 -Oe field in the frequency range 10—10 000 5.591 6.065 111.53 CuLAS-10 5.599 6.050 25.268 114.04 23.3 Hz, in the temperature range 5—30 K (PPMS, Quantum Design Co.). X-ray diffraction data can be analyzed on the basis of a monoclinic structure, so that the resultant cell parameters can 3. Results and Discussion be compared with those of the parent material. The cell 3.1. Syntheses. While synthesis of the whole CuLAS-u (// parameters were refined with sufficient precision using the — 2—18) series was attempted by means of anion exchanges, “TREOR 90 ” indexing software and by profile (Pawley) which was unsuccessful for CuLAS-2—4 and CuLAS-12— 1 8 refinement in Materials Studio v2.2 platform. As seen from because of decomposition (/; = 2—4) and incompleteness of Table 1, the cell parameters a and b for the intercalated the ion-exchange (/; = 12—18). The latter may be due to the compounds are very close to those reported for the monoclinic weak coordinating power of the sulfonate anion to the metal unit cell of the parent material, although the c parameter depends ion. As a result, only the CuLAS-u (/? = 6, 8 , and 10) on the length of the alkyl chains. compounds were obtained as polycrystals, whereas the Cu LAS-4 It is well-known that the alkyl chain packing can be and CuLAS-12 were obtained as mixed phases. Chemical understood through an analysis of the variation in the funda analysis confirmed the composition of the CuLAS-// materials, mental layer spacing (dom) with the alkyl chain length. As was indicating the total exchange of nitrate anions. [Anal. Calcd established in a previous work, the basal spacing is related to (found) for C.6Cu2Hl(iS06 (CuLAS-6): C, 20.99 (21.03); H, 4.70 the carbon chain length (//) through the relationship d(A) — do (4.86); N, 0.00 (0.07); S, 9.34 (8.20). Anal. Calcd (found) for + //(1.27/; cos 0), where ;/ = 1 or 2, depending on the chain CsCuiHzoSO* (CuLAS-8 ): C, 25.87 (25.84); H, 5.43 (5.38); packing and 0 is the till angle of the chains.12 In particular, the N, 0.00 (0.00); S, 8.63 (8.52). Anal. Calcd (found) for distance d 0 involves the size of the bridging group, the van der CIo Cu 2H2.1S06 (CuLAS-IO): C, 30.07 (29.92); II, 6.056 (6.033) VVaals distance between cither facing methyl groups or methyl N, 0.00 (0.00): S, 8.027 (S.124)J. For all compounds, it is noted groups and hydroxide layers, and the thickness of the inorganic that no lattice water is contained. layer. 3.2. X-ray Powder Diffraction. As shown in Figure l, the For the u-alkylsulfonatc derivatives, as shown in Figure 2, CuLAS-// compounds exhibit a lamellar structure as is evident the variation of the basal spacing increases linearly with the from the powder XRD patterns exhibiting, in the low 20 range, aliphatic chain length //, according to the relationship d(A) = intense (007) relied ions, up to at least the third harmonic, 8.39 -F 1.48//. This indicates that, if the molecular area of the corresponding to the slacking periodicity of the hvdroxidc-based chains is constant, regardless of the // value, the //-alkyl chains layers. First of all. before extracting the crystallographic structure are organized in a partially inlcrdigitated bilayer packing with from the powder XRD patterns, it is necessary to confirm the a lilt angle 0 — 36'. in very good agreement with previous completeness of the anion exchange reaction in the CuLAS-// findings for the Cu(il) parent compounds .1 ’ Till of the aliphatic series compounds, which can be checked simply via the relative chains or partial interdigitation of the chains results in a value changes in peak positions with respect to the starting material. between 1.27 and 2.52 A/carbon atom. Using the same argu

Cud OH),NO, (dm = 6.96 A). In all cases, since no reflections ments, the intercept of the basal spacing vs the carbon chain from the starting compound arc observed, wc can conclude that length plot, dn. can be assigned to the thickness of the inorganic the total exchange reaction has been successfully realized. In layer (-• 5.0 A) plus the surfactant hcadgroup (2.65 A); 2.65 x addition, thanks to the high crystallinity of the samples, the 2 + 3.0 — 8.3 A. This simple calculation agrees with the 1120 J. PAys. CAe///. g. fW. /09, #o. 3, 200 J Park and Lee

~ 25-

.= 20-

“ 10-

number of carbons Figure 2. Dependence of the basal spacing on the carbon number in the copper(II) hydroxy n-alkylsulfonates. The vertical lines show the calculated molecular heights of sulfonates. Linear variations are observed. The values for CuLAS-4 and CuLAS-12 were obtained from structure analyses of the mixed phase samples obtained. experimental value, 8.39 A. From the resultant XRD data analyses, we are capable of extracting the lamellar structure for the CuLAS-z/ (n = 6, 8 , and 10) as is displayed in Figure 3, consisting of alternating inorganic copper hydroxide layer and organic layers of aligned alkyl chains with partially inlerdigitated bilayer packing. 3.3. Infrared Study. IR spectroscopy has been known to be an effective technique capable of checking the completeness of the anion exchange reaction as well as understanding the molecular structure of the organic anions in the interlayer spaces.14 For instance, from IR spectroscopy of alkyl chain assemblies such as zz-alkylsulfonates, it is not difficult to extract some information on the metal—anion coordination, on the chain conformation, and on the packing of the alkyl chains even if the complete crystal structure is lacking. Indeed. IR spectroscopy is able to provide a firm basis of extension of the vibration analysis to related classes of chain assemblies including the organic —inorganic materials of concern here. Figure 3. Model description of the proposed structure of the CuLAS-rz Fundamental Mode. The whole patterns of the characteristic (zz = 6, 8 , and 10). This structure is a modification of the botallackite IR spectra obtained arc shown in Figure 4. For comparison, (Cu 2(OH)3C1) type, where Cl ion is replaced with alkylsulfonate anion, the IR spectrum of the parent material is also shown. The most consisting of alternating inorganic copper hydroxide layer and organic layers of aligned alkyl chains with partially inlerdigitated bilayer distinct feature for the CuLAS-zz and the parent material. Cu- packing. NOs, is noticed in two different wavenumber regions, i.c., 2900-3500 and 1000-1500 cm L that correspond to the frequency regions in detail. In fact, the high-frequency region hydroxy groups and to the stretching vibrations of anions, (2700 -3100 cm"1) reveals the C 11 stretching modes of the respectively. In other words, the stretching vibration of the methyl and the methylene groups while the low-frequency hydroxy groups is shifted to a lower frequency, which may be region of 600 1800 cm"1 displays the stretching modes of the attributed to the weakness of O IL and the weakness anchor sulfonate group as well as the scissoring, rocking, wagging, and absence of a broad feature in the hydrogen-bonding region twisting modes of the methylene groups (Figure 5). (3550—3200 cm"1) and the HOI 1 bending mode (1630 -1600 The characteristic band-signatures (Figure 5) of overlapping cm"1). These features are indicative of no lattice water, in peaks observed in the high-frequency region of the IR spectra accordance with the elemental analysis. On the other hand, the arc straightforwardly assigned lo the C 11 stretching modes of disappearance of vibrations due to the N-O of the nitrate group, the polymethylene [ -(Cl L).,—] sequence and end-methyl identified around 1423 cm"1, and the appearance of the S O (-CIL) groups according to the prow ious assignments of the band from the sulfonate group in the intercalated materials are long-chain //-alkanes.1-' More specifically, the two intense bands observed. In effect, the IR spcclra qualitatively confirm the total at 2853 -2851 and 2934 ....2919 cm 1 are assigned, respectively. exchange of the nitrate by the sulfonate as well as the absence lo the symmetric (iv(Cl L)a/') and the antisymmetric (TJS(CIL'),t7") of lattice water. stretching vibrations of the methylene groups. The peaks Molecular Structure. To understand the molecular conforma observed at 2872 and 2958 cm 1 are assigned to the symmetric tion or packing characteristics of the alkyl chains as well as the ( i',(CH; j.r' ) and the antisymmetric t n ALTL),''") stretching metal..anion coordination mode In the interlayer spaces, it is vibrations of the methyl groups, respectively. In the high- necessary to check lire C 11 vibration at high- and low - frequency region, two shoulder peaks are additionally identified Layered Copper Hydroxide n-Alkylsulfonale Salts J. p/tyj. C/ten,. B. Co/. /OP, /Vo. j, 2005 1121

The low-frequency region (650-1800 cm"1) provides ad ditional structural information regarding CuLAS-u. As men tioned previously, peaks appearing in this region are associated with the stretching vibration of the sulfonate group as well as with the scissoring, rocking, wagging, and twisting modes of the methylene group. The three samples of CuLAS-u exhibit Ihe characteristics of the fundamental and the split iq(S -O) stretching modes in the range of 1000 — 1250 cm"1. Two stretching vibrations of SO.;, i.e., izJSO;" ) and rs(S03 " ), are observed at 1195-1198 and 1035 cm"1, respectively, whereas in free ligands the two bands appear at 1175 and 1050 cm"1, respectively. Furthermore, the difference Av (=vas — vs) obtained here is ca. 160 cm"1, which is greater than that of the ligand (An — 125 cm"1), as expected for the unidentate SO; coordination, and is very close to that reported by Kurmoo ct at, for the isotypic compound Ni2(OH)r 2500 2000 (C,iH2sS6,)-H2OA Wavenumbers (cm' The appearance of a single narrow peak at 1473 or 1467 cm "1 Figure 4. Infrared spectra for (a) Cu2(0H)3N03 (parent compound) has been attributed to triclinic or hexagonal subcell packing, and (b—d) Cu2(OH)3(C„H2„+1S03), where n = 6, 8 , 10, respectively. respectively. The appearance of a well-resolved doublet with two distinct components is known to occur either as a result of the intermolecular vibrational coupling due to a crystal-field splitting in the orthorhombic or monoclinic packing or as a result of the coexistence of triclinic and hexagonal packing in the material.17 In addition, the peak is known to be broad when the alkvl chains assume a disordered conformation. On these grounds, the fact that a well-resolved doublet band is observed at 1467 ctn 1 suggests a hexagonal packing with disordered conformation of all-trans chains. The 1R and X-ray diffraction studies show that the structural features of the CuLAS-n arise mainly from the amphiphilic nature of the intercalated alkylsulfonate anions in two ways: (i) the hydrophobic alkyl chain enables the layers to be spaced up to 23.3 A apart, being tightly packed with hexagonal subcells between the layers, and (ii) the hydrophilic sulfonate headgroups are directly bonded to copper(ll) ions, leading to partial Wavenumbers (cm') interdigitated bilayer packing between the inorganic layers. Figure 5. Infrared spectra of Cu2(OH)3(C«H2n+1S03) where n = 6 (a), 3.4. Magnetic Properties. The magnetic properties of materi 8 (b), and 10 (c), respectively, in the high-frequency (2700-3100 cm"1) als involving copper(ll) ions are known to be very sensitive to region, showing overlapping contributions from CH2 and CH3 stretching any structural modification that may be introduced by sulfate modes (left), and in the lower frequency (500—1700 cm"1) region, or carboxylate bridging species. A remarkable feature of those showing the stretching vibration of the sulfonate headgroup as well as systems is that the magnetic properties and the stoichiometries, the scissoring, rocking, wagging, and twisting modes of the methylene group. determined by the lattice water contents, are strongly correlated. It has been known that both /j-alkyl sulfate and carboxylate at 2860 and 2965 cm1. The former peak can be attributed to derivatives with long chains have two different structural the Fermi resonance absorption due to the c/+ mode, whereas varieties. In particular, for the carboxylate derivatives, the the latter peak has to be attributed to the r mode. It is well- hydrous a-phase is AFM, but the anhydrous //-phase is FM. known that for the two antisymmetric stretching vibrations of Thus, great variations should be observed in the magnetic a CH3 group to become degenerate and appear as a single broad behaviors of the exchanged compounds depending on the water peak, the CHj group must be at least in a C3 symmetry. 1'' In contents as well as the alkyl chain packing. the present ease, the symmetry is apparently further lifted due Our preceding work showed that the newly synthesized to the interpenetration of the alkyl chains. The two antisym copper-based alkylsulfonate anhydrous compounds possess a metric vibrations will then no longer be equivalent, splitting long-range order .111 in contrast lo the hydrated derivatives into two peaks. exhibiting no sign of long-range order. To see the effect of It has been well-established that the t/ and ct modes are molecular size of Ihe intercalated anion on the magnetic behavior strong indicators of the chain conformation. The d' and d" of the. inorganic layer, we have carried out magnetic measure modes usually lie in the narrow ranges of 2846 —2850 and ments on the new series of alkylsulfonate derivatives with no 2915-2918 cm-1, respectively, for all-trons extended chains lallice w ater. and in the distinctly different ranges of 2854 -2856 and 2924 .. The magnetic susceptibility data were recorded for all 2928 cm"1 for disordered chains characterized by a significant Oul.AS-u from 5 K to room temperature with = 0.5 T presence of gauche conlbnners. On this basis, the observed peak The temperature dependences of % and the yT product for frequencies of the d 1' and d~ modes suggest that the alkvl chains CuLAS-u are plotted in Figures 6 — 8. All the compounds exhibit are in an all-lrans conformational state with little or no a similar magnetic behavior. At high temperatures, the yT values significant gauche population. correspond to two CutII) (d". .S' — 1 21 per mole lea. 11.8 emu 1122 J. f/zyj. C/zgfM. B, Ko/. /OR /Vo. 3, 2003 Park and Lee

0.030 TABLE 2: Main Magnetic Results basal sample spacing (A) 7WK° e/K6 C/cm3 mol 1 Kh CuLAS-6 17.1 11.5 8.54 0.668 CuLAS-8 20.1 10.2 3.04 0.833 CuLAS-10 23.3 11.0 22.1 0.689 d The Neel temperature, 7n , was determined from the sharp peak in the ac magnetic susceptibility. h The © values were obtained from the high-temperature fit of the dc magnetic susceptibility, = {T - ©)/ C.

, n = 10 0.000 ------,------,------,------,------,------,------,------,------T------T------,------t------T------O.o T=2DK 0 50 100 150 200 250 300 n=8. T=5K Temperature/ K Figure 6. The temperature dependence of % and the %T product for n=10.T=5K CuLAS-6. The solid line indicates a high-temperature fit to the 2D T=20K Heisenberg model. 1500-

1000-

500-

10000 20000 30000 4W00 50000 60000 70000 60000 H/Oe Figure 9. Magnetic field dependence of the magnetization of Cuz- (0H)3(C,(H2^iS03) at 5 K (below) and 20 K (above 7n ), for n = 6, 8 , and 10. Inset: the derivative of the field dependence, clearly showing a discontinuity behavior.

0 50 100 150 200 250 300 Temperature IK expansions for the spin 1/2 2D Heisenberg triangular lattice,19 Figure 7. The temperature dependence of % and the product for a very good agreement between theory and experimental CuLAS-8 . The solid line indicates a high-temperature fit to the 2D findings was obtained with Jik = 2.725. 0.38, and 10.8 K for u Heisenberg model. = 6, 8 , and 10, respectively. A summary of the main magnetic results is given in Table 2. As shown in Figure 9. the field dependences of the magne tization below rN (at 5 K) also show a pronounced sigmoidal shape with an inflection point at a critical field of about 2.5 T Si and with an increase in the slope al higher fields, which is O attributed to a spin-flop transition. This is consistent with the "E ! coexistence of the ferromagnetic and antiferromagnetic interac J tions. which is characteristic of a metamagnet. The observed behaviors arc similar to those of the transition metal hydroxides M(OH)2, with M = Fe, Co, and Ni, which are well-characterized as melamagnets .29 Further information on the three-dimensional ordering is

Temperature /K provided by the temperature dependence of the ac magnetic susceptibility, in a magnetic field //»,= 1 Oe and at the ac Figure 8 . The temperature dependence of % and the %T product for frequency of 10 kHz (Figure 10). The in-phase y.iC of the ac CuLAS-lO. The solid line indicates a high-temperature fit to the 2D Heisenberg model. susceptibility for all the samples increase wilh decreasing temperature, with a rather broad maximum at around 13 K., kmol-1 ). The values of yj for CuLAS-i; ;U high temperatures which is expected for a two-dimensional antil'erromagnet. No increase with decreasing temperature, indicative of a ferromag frequency-dependent bchav lor was observed. While all the netic interaction. However, the marked decreases in //’ al low samples exhibit a broad maximum at around 13 K, apparently temperatures arc indicative of an interlayer anliferromagneiie distinct behaviors are noticed between the CuLAS-6. -10 and coupling. 1,s The magnetic data above 1 50 K can be tilted well CuLA$-8 materials. Specifically, while a pronounced sharp peak (o I he Curie..Weiss law with C ~ 0.668 cm'1 mol 1 K and 0 is observed for Cu!,AS-6 and CuLAS-10, it appears that the = 8.54 K (;i = 6). C — 0.835 cirri mol "1 K and 0 — 3.04 K. (n sharp peak is much suppressed for the CuLAS-8 , leaving only — 8 ), and C — 0.689 enri mol"1 I< and 0 — 22.1 K (n — 1 0). a sign of I lie anomalous behavior. The sharp peaks in both y'.K A lit of the magnetic susceptibility was made in the paramag and are taken to be indicative of a FM order, expected for netic regime to determine the in-plane exchange interaction a metamagnet built of spin-canted antiferromagnetic layers. The between the copper(II) ions. Using the high-temperature series Neel temperature. 7i,. of the CuLAS-n {u — 6. S. and Hi) Layered Copper Hydroxide n-Alkylsulfonate Salts y. PAyj. CAem. g. Co/. /09, No. j, 2005 1123

0.05 1

0.04 -

0.03 -

0.02 -

0.01 -

Temperature /K Temperature /K Figure 10. The temperature dependence of the ac susceptibility of Cu 2(0HMC„H2„+iS03). material was determined to be 11.5, 10.2, and 11.0 K, existence of hydrogen bonding may block a 3D long-range order respectively, from the sharp peaks in the ac magnetic suscep between the layers, resulting in the much suppressed sharp peaks tibilities. indicative of the 3D order as is observed 111 our ac magnetic It is worth noting that long-range magnetic behavior is susceptibility measurements. observed with the interlayer distance up to 23.3 A. in contrast to the hydrated coppcr(ll) alkylsulfonate derivatives exhibiting 4. Conclusions no sign of a long-range order. While all the compounds exhibit A novel series of layered copper hydroxide-based sulfonate similar magnetic behaviors, as shown in Table 2, it is worthwhile compounds, CuLAS-n (n = 6, 8 , 10), has been prepared by to discuss the origin of the distinct magnetic behaviors of the exchange reaction starting from the corresponding layered CuLAS-8 versus the CuLAS-6, 10 materials; The CuLAS-8 copper hydroxy nitrate. X-ray powder diffraction investigations material shows a higher Curie constant and a lower Weiss show that these materials derive from the brucite structure by temperature as well as the suppressed low-temperature sharp replacing nitrate groups by sulfonates and that alkyl chains are peak in the ac magnetic susceptibility. A small difference in interleaved with partially interdigitated bilaycr packing between the alkyl chain lengths in the compounds may give rise to slight the copper hydroxide layers. According to the FTIR study, the in-plane structural modifications and thus to different exchange alkylsulfonates are completely exchanged and bonded to the interactions. In fact, a careful inspection of the 1R spectra (Figure metal via a unidentate binding mode. The magnetic properties 4) reveals subtle differences between the compounds in the have been investigated and explained by taking into account characteristic frequency regions, i.e., hydrogen bonding (3200 - the main structural features evidenced by XRD and 1R studies. 3550 cm'1) and sulfonate group regions (1000— 1250 cm-1), These anhydrous materials exhibit a long-range magnetic order, indicating the existence of hydrogen bonding between an oxygen while the hydrated derivatives show no sign of a long-range in the SO3 hcadgroup and the OH in the copper hydroxide layer. order. The CuLAS-n series show a similar metamagnet behavior Drillon and Panissod 21 have proposed a model to explain the at lower temperatures, with 7'N ~ I 1 K, corresponding to the magnetic behavior of layered compounds in relation to the basal presence of ferromagnetic subnetworks weakly coupled by spacing. In this model, for the in-plane ferromagnetic interac anliferromagnelic coupling. The different behaviors arc under tions, the in-planc spin moments align within the correlation stood by the existence of hydrogen bonding, giving rise to slight domains whose sizes increase upon cooling, thus leading to giant in-plane structural modifications and thus to exchange inlcrac- magnetic moments in each layer. The interaction between these iions. moments depends on the interlayer distance. For small distances, superexchange interaction occurs via hydrogen bonds and may- Acknowledgment. This work was supported by the K1STEP lead to anti ferromagnetic or mctamagnetic order. In turn, for I National Research Laboratory and M102KS010001-02K1901 - large distances, the through-space dipole - dipole interaction 01314) and by the Korea Research Foundation (Brain Korea between the layers becomes predominant and favors long-range Project in 2004 and Grant No. KRF-2004-005-00060). Measure ferromagnetic order. In the present system, although the ments at the Seoul Branch of the Korea Basic Science of magnetic layers are well-separated with huge distances. I In Institute (KI3SI) arc gratefully acknowledged. 1124 V. P/aw CW;. g. Kr;/. /W. Ak;. Park and Lee

References and Notes Fujita. Wg Awaga. K.: Yokoyama, T. Inorg. Chem. 1997, 36, 196. (8 ) (a) Laget, V.; Rouba, S.: Hornick, C.; Drillon, M, J. Magn. Magn. (1) (a) Levy. F. /w/erccy/aW Amrrgf/ A/omnVz/.t: D. Reidel: Dordrecht, Mater. 1996, 154, L7. (b) Okazaki. M.; Toriyama. K.; Tomura, S.; K.odama, The Netherlands, 1979. (b) Wittingham, M. S.; Jacobson. R. A. /wfivcwW/Y/; Tg Watanabe. E. /norg. CAem. 2000, 49, 2855. Chemistry: Academic: New York, 1982. (c) Giannelis, B. P. Adv. Mater. (9) Kurmoo. Mg Day, P.: Derorv, A.: Estoures, Poinsot. Rg Stead. 1996, 8, 26. (d) Fergusson, G. Sg Kleinfeld. B. R, Adv. Mater. 1995. 7. M. J.; Kepert, C. J. V. W. 1999. /<5, 452. 414. (e) Alberti, G.; Marmottini. F.: Murciamascaros, S.; Vivani, R. Angen-. (10) Park, S.-Hg Lee, C. H.: Lee, C. Eg Ri. H.-C.: Shim. S. Y. AWcr. Chern., hit. Ed. Engl. 1994. 44, (5. (f) Alberti. Gg Casciola, M.; Constantino, /(f.T. ^///. 2002, 47. 1773. U.; Vivani, R. Adv. Mater. 1996, 8. 291. (g) Novak, R. M. A dr. Mater. (11) Meyn, Mg Beneke, Kg Lagnly, G. /wo;#. CAevn. 1993, 42, 1209. 1993, 5, 422 (12) Kitaigorodskii, A. I. CAcm/m/ Consult- (2) (a) Desiraju, G. fAc De.rigH o/"Jo/n/s: ants Bureau: New York, 1955. Elsevier: New York, 1989. (b) Stein, A.: Keller, S. Wg Mallouk, T. B. Science 1993, 259, 1558. and references therein. (13) Rabu. Pg Drillon. Mg Awaga, Kg Fujita, Wg Sckine, T. In Magnetism: Molecules to Materials II: Miller, J. S., Drillon. M., Eds.; Wilev (3) (a) Day, P. J. Client. Soc., Dalton Trans. 1997, 701. (b) Mitzi. D. VCH. Weinheim, 2001,357. B. Prog. Inorg. Chem. 1999, 48, 1. (c) Coronado, E.; Galan-Mascaros, 1- R. ; Gomez-Garcia, C. J.; Lankin, V. Nature 2000, 408, 447. (d) Clear Held. (14) (a) Nakamolo, K. u/'/nonyun/c u;/4 A. Prog. Inorg. Chem. 1998, 47. 371. (c) Tolbert, S. Hg Sieger, P.; Stucky, Coordination Compounds, 5th ed.: John Wiley & Sons: New York. 1997. G. D.; An bin, S. M. J.; Wu, C.-Cg Hendrickson, D. N. J. Am. Chem. Soc. (b) Colthup, N. B. //Rroc/frcf/un m 3nl 1997, 119, 8652. (f) Carling. S. G.: Mathoniere. C.; Day, P.: Malik, K. M. ed,: Academic Press; New York, 1990. A.; Coles. S. J.; Hurslhouse. M. B. V. C/?gm. ,%/c.. Du/m;? TTu/;.?. 1996. (15) (a) Wallach, D. F. H.: Vemia, S. Pg Fookson, J. /J/ucA/m. 1839. .4cm 1979. 559. 153. (b) Snyder, R. Gg Schachlschnelder. J. H. 5/;ec/mc/;//n. (4) O’Hare, D. New J. Chem. 1994, 18, 989, and selected references Acta 1963, 19, 85. (c) "Fastimi, Mg Shimaanouchi. T.: Watanabe. Ag Goto, therein. R. Spectrochirn. Acta 1964, 20, 629. (d) Nuzzo, R. Gg Kerenic, E. Mg (5) (a) Miyata. S.; Kuniura. T. Chem. Lett. 1973, 843, (b) Yamanaka. Dubois. L. IT. ,/. Chem. Phys. 1990. 93, 767. S. ; Sato, T.; Hattori, M. Chem. Lett. 1989, 1869. (c) Yamanaka, S.: Sato. (16) (a) Snyder, R. Gg Strauss, H. Lg Elligcr, C. A. 4. P/ryj. C/mw. T. ; Seki, Kg Hattori, M. Solid State Ionics 1992. 53~56. 527. (d) Rabu, 1982, 86, 5145. (b) Snyder. R. Gg Maroneelli, Mg Strauss, H. Lg Hallmark, Pg Angelov, S.; Legoil, P.; Belaiche, Mg Drill on. M. Inorg. Chem. 1993, V. M. J. Phys. Chem. 1986, 90, 5623. (c) MacPhail, R. A.: Snyder, R. G.: 32, 2463. (e) Rouba, S.; Rabu, P.; Ressouche, E.; Regnault, L.-P.; Dr ill on, Strauss, H. L J. Chem. Phys. 1982, 77, 1118. M. J. Magn. Magn. Mater. 1996, 163, 365. (17) (a) Borja, Mg Dutla, P. K. 4. Mry.s. CWr 1992, 96. 5434. (b) (6) (a) Rabu, P.; Drillon, M. Adv. Eng. Mater. 2003, 5, 189. (b) Rabu. Aimirantc, Cg Minotii, Gg Zerbi, G. 4. Phys. Chem. 1986, 90, 852. P.; Drillon, M.; Hornick, C. Analysis 2000, 28. 103. (c) Laget, V.; Homick. (18) Carlin. R. \... Magnetochemistry: Springer-Verlag: Berlin, 1986: C. ; Rabu, Pg Drillon. Mg Ziesscl, R. CooroT /fee. 1998, /7/*-/6Y/. p!22. 1533. (d) Fujita, W,; Awaga. K.: Yokoyama, T. Appl. Clay Sci. 1999, 15. (19) Elstncr. Ng Singh. R. R. P.: Young, A. P Phys. Rev. Lett. 1993, 281. 7/, 1629. (7) (a) Rabu, P.: Rouba, S.; Laget, V., Hornick, C.; Drillon. M. J. Chem. (20) (a) Takada, Tg Ban do. Yg Kiyama, Mg Miyamoto, H. 4. Phys. Soc.. Chem. Commim. 1996, 1107. (b) Laget, V.; Drillon. M.; Hornick, C.; Soc. Jpn. 1966, 21, 2726. (b) Takada. Tg Bando, Yg Kiyama. M.: Miyamoto, Trabu. P.; Romero, F.; Turek. Pg ZiesseL R. J. Alloys Compels. 1997, 262— H. 4. ,5"oc. 4/?n. 1966, 2/. 2745. (e) Miyamoto. Hg Shinjo, Tg Bando, 263, 423. (c) Laget, V.; Hornick, C.; Rabu. P.; Drillon, M. J. Mater. Client. Yg Takada, T. 4. PAy.s. Am. 4/;/?. 1967, 25. 1421. 1999, 9, 169. (d) Fujita, W\: Awaga, K. Inorg. Chem. 1996, 35, 1915. (b) (21) Drillon, Mg Panissod, P. 4. Magn. Magn. Mater. 1998, 188, 93. Available online at www.sciencedirect.com

SCIENCE DIRECT”

ELSEVIER Thin Solid Films 497 (2006) 185 188 ww\v. elsevier.com/locatc/lsf Short communication Effects of oxygen-plasma treatment on lanthanum-substituted bismuth titanate ferroelectric thin films

E.R. Park3, Cheol Eui Lee3'*, S.J. Yeom b

1 Department of Physics and Institute far Nano Science. Korea University. Seoul 136-713. South Korea 11 Memory R&D Division. Hynix Semiconductor. Inc , Kyoungki-do 7T'T'■'6/ South Korea Received 25 March 2005; received in revised form 1 August 2005; accepted 19 September 2005 Available online 14 October 2005

Abstract

Oxygen-plasma treatment has been done on Bi s jjLoo S5Ti30|2 (BLT) films for lower crystallization temperatures and improved ferroelectric properties. The films, prepared by metal organic decomposition method on Pt and Ir02 electrodes sputtered on ABOySiOVSi substrates, were pre annealed at the relatively low temperatures of 550 and 600 °C, followed by room temperature microwave oxygen-plasma treatment. The oxvgen- plasma-treated BLT films showed very good ferroelectric characteristics after a final furnace annealing at 600 °C. © 2005 Elsevier B.V. All rights reserved.

PACS: 77.80. —e; 81.40.-z; 85.50.Gk; 52.77.-j Keywords': Ferroelectric properties; Plasma processing and deposition

1. Introduction possess ferroelectric properties as well as processing tempera tures intermediate between PZT and SBT. and is expected to Among the various ferroelectric thin films developed for solve the aforementioned problems to some extent. Much work application to the nonvolatile ferroelectric random access has been done to achieve BLT films of optimum quality using memories (FRAMs) [1,2], Pb(Zr,Ti)Oj (PZT) and SrBi2 Ta2 various techniques. To our knowledge, BLT films thus reported Ot, (SBT) thin films have been extensively studied for their needed ~ 650 °C or higher annealing temperatures for good ferroelectric properties [3], PZT thin films have large crystallization in order to obtain fair ferroelectric properties remanent polarization and lower fabricating temperatures [7-11]. In order to realize high-density device fabrication whereas SBT thin films have an excellent fatigue endurance compatible with the present information and communication against repeated switching of polarization even with Pi technologies, it is highly desirable to cany out the fabrication electrodes. While the PZT and the SELF thin films are leading process at the lowest possible temperatures. Thus, lowering the candidates for FRAM applications, both of them possess processing temperatures is currently one of the critical issues. problems to be solved for the FRAM applications and Rastogi el al. reported highly crystalline SBT films obtained environmental safeties, such as the degradation of the at a low temperature of 700 °C by introducing oxygen-plasma polarization state of the capacitor with Pt electrodes and the discharge during pulsed laser ablative deposition [12]. Besides, Pb toxicity for PZT, as well as the relatively lower remanent polarization, higher processing temperatures, and low Curie Table I temperature for SBT. Annealing and plasma-treatment conditions for the BLT films La-substituted bismuth titanate (Bi4La4. rTi3Ol:. BLT) thin Sample no. BIT film La;-annea ling Ox ygen plasma Final annealing films, emerging as a new ferroelectric material for FRAM 1 BUI Pt 5U min al 6U0 V 50 min 30 min at 600 "C applications by Park el al. 's systematic approaches [T 6], 2 BIT Tr 50 min ai 6(10 T' 60 mm 30 min at 600 T 2 BLT Ir(); 90 min at 550 <; 60 min 30 min at 600 T BLT lifw 90 min at 60(1 T 50 min 30 min at 600 "C * Corresponding ouihor. Id.: = H2 2 3200 200)4: lax. -R2 2 02"? 3202. 4 BLT. IrO. 90 min al 600 -x: 00 min 50 min at 600 "C /wnm/ rscck/koreanckr (C.E. Lee). 5

Jang et al. introduced oxygen-plasma treatment on PZT films 2. Experimental details for improved electrical properties [13]. According to their X- ray photoelectron spectroscopy (XPS) study, oxygen-plasma The BLT thin (100 nm) films were spin-coated on Pt (200 treatment promotes the transformation of metallic Pb to PZT Pb mil) or lr02 (200 nnr) bottom electrodes which were sputtered and thereby improves the stoichiometry in the bulk as well as on ABOj-buffered Si substrates (ALOVSiCT/Si), using metal - near the surface. This work provided an idea of the possibility organic decomposition solutions whose compositions were of its application to thermal annealing processes. In this paper, optimized to Bi;, 25Lao.K 5TijOi2 for the largest polarization we report on the application of oxygen-plasma treatment for [14]. The solutions were dissolved in 2-ethylhexane acid thermal annealing of BLT thin films at temperatures as low as (C7H15COOH) and octane (CSHi2), which was used as a 550 and 600 °C. solvent of metal-hexanoate chain agent for further stability of

Polarization (pC/cnr )

Fig. 1. P~E hysteresis loops for line BIT films on Pi and lu'K bo Uom eleeliodes after '.lie final annealing process I'm sample nos. (a) l.ih)2. tel 3. lit) 4, and (e) 5 as indicated in Table I. i, R Park nl al ' Thin Solid Films -407 (2006i ITT - 1ST 187 the precursor. After baking at 260 °C for 4 min in air, the BLT (a) films were pre-annealed in oxygen atmosphere in a conven tional furnaee annealing (FA) chamber at the temperatures of 550 or 600 °C. for 30 or 60 min. An oxygen-plasma treatment process was subsequently executed by exposing the BLT films to a 200-Pa oxygen-plasma ambient in a 2.45 GHz microwave plasma reactor (Plasma-finish GmbH Inc.) at room tempera ture, with a microwave power of 60 W for optimized results. For the measurement of electrical properties, Pt/BLT/Pt and Pt/BLT/IrO, capacitor structures were fabricated by sputter- deposition of Pt electrodes with an area of 5.2 x 10 4 cm2 on BLT films after the plasma treatment process. Then, a final FA process was done on BLT films in oxygen atmosphere at 600 °C for 30 min in order to obtain the ferroelectric hyslereses. The conditions of the thermal annealing processes as well as the oxygen-plasma treatment of BLT films reported here are summarized in Table 1. The crystallinity of the BLT films was analyzed by the X- ray diffraction (XRD) patterns after each process. The X-ray diffraction data were collected on a MAC Science diffractom eter (MXP3A-HF) operating at 40 kV and 30 mA in the Bragg -Brentano 0-20 mode (CuKoc, 2-0.15418 nm). The polarization-electric field (P-E) and fatigue characteristics yiasma tri-armenl were taken by a RT66A tester, and the leakage current-electric field characteristics was taken by a F1P4155A semiconductor parameter analyzer.

2l) (deg'TVs) 3. Results and discussion Fig. 2. Typical XRD patterns for BLT/Pt and BLT/IrO, films each step of the processing, (a. b) Films 1 and 3, respectively. A sharp peak near -4If arises Fig. 1 shows the P-E hysteresis loops of the Pt/BLT/Pt and from the Pt top electrode. *XRD patterns with plasma treatment for 60 min, Pt/BLT/lrCL capacitor samples in Table I after the final FA instead of 70 min. process. Measurement of the P-E hysteresis loops right after the deposition of the top electrodes gave no indication of are well correlated with the fonnation of the perovskite phase ferroelectricity. However, after the final annealing process, the in the XRD patterns. While the oxygen-plasma treatment is capacitors showed very good ferroelectric behaviors, as shown shown to certainly assist the crystallization of the BLT films, its in Fig. 1, for all the plasma-treated samples in Table 1. In effect on the change in the crystallinity is less prominent than particular, a Pt/BLT/IrCL sample, which was pre-annealed for that of the final annealing. Thus, it may seem that the 90 min at 600 °C and oxygen-plasma-treated for 90 min crystallization is mostly dictated by the final FA process. followed by FA for 30 min at 600 °C (sample no. 5), exhibits a However, the BLT films that were prepared under the same quite large remanent polarization despite a leaky hysteresis. condition except for an oxygen-plasma treatment process did Here, it is quite encouraging that the ferroelectric character not show distinctive changes after the final FA process, istics of a BLT/lrCL sample (sample no. 3), in spite of pre indicating that the oxygen-plasma treatment has played an annealing at a temperature as low as 550 °C, was induced by indispensable role in the crystallization of the BLT films during plasma treatment. It is worthwhile to note the best values the thermal annealing. We have also carried out the XPS obtained here for the remanent polarization (2Pr) and the measurements in order to investigate the roles of oxygen- coercive field (£/.), 9.5 pC/cm 2 and 114 kV/cm for the Pt/'BLT/' plasma treatment in the crystallization process, which revealed Pt, and 8.0 |iC/cnr and 106 kV/cm for Pt/BLT/lrCL, that the oxygen-plasma treatment process induces changes iir respectively. In particular, the P, values arc comparable to the chemical states of constituent atoms, which is believed to the best previous results [14,15], ll is also to be noted that no facilitate the crystallization [13], The XPS spectra of Bi 4/1 O ferroelectricity was observed for any annealing condition in lx, and La 3d suggest that the La atom substitutes for the Table I without a plasma treatment. position of Bi at the perovskite layers, and that oxygen-plasma Fig. 2 shows the typical XRD patterns of the Bl.T/Pt and the treatment plays a much more ‘oxidative' role than the BLT/'lrOn films after finishing each step of the processing. It is conventional furnace annealing under oxygen atmospheres. seen that the crystallization is hardly completed solely bv the The fatigue characteristics was also studied. The P- E FA processes. The crystallinity is shown to be improved with curves were measured before and alter being subjected to each step of the processing. All the samples turned out to show 5 v 10s reacl/write cycles, where the cycling frequency and primarily a e-axis orientation, anil the ferroelectric properties electric field were kept at 500 kl Iz and 500 kV/cm. respectively. 188 & R Funt' c; / 7%m A'/"/ FfW V#" /R5 /RR

The BLT capacitors exhibited good fatigue resistance up to 10* [2j R. Dal. IK Let'. O. Aucicllo, At. Kmgon, Appl, Phys Lett, 67 (I 995) cycles within 10% loss oftheir initial polarizations. The leakage 572; C.H Lee. K..W. t.ee, C.L Lee, Cure, Appl Pliys. 3 (2003) 477. current density was similar for the Pt/BLT/Pl and Pt/BLT/IrO: [3 | I. Li. V.D Zhu, 5.B. Desu. C.H. Peng. M, Nagata, Appl. Pliys Lett 68 capacitors, with a magnitude of order of 10"5 A/cm2 at 300 kV/ (1096)616 cm. The current density of the sample no. 5 in Table 1 could not |4J 13 H Park, S..I. Myun. S.D. Bu. T.VV. Noli. Appl Pliys. Lett. 74 (1999) be measured because of its highly leaky characteristics. 1907. According to one of the arguments on the fatigue characteristics (5J B S. Kang, B.H Park. S.D. Bu. S.H. Kang, T W Noh, Appl. Pliys Lett. 75 (1999) 2644. in the fenoelectric thin films, it originates from the increase in [6] B.H. Park, B.S. Kang. S.D. Bu, T.W. Noh, W. Jo, Nature (Lond.) 401 the leakage current density [16-18]. (1999)682. In summary, oxygen-plasma treatment was introduced in the 17] D. Wu, A. Li, T. Zhu, Z. Liu, N. Ming. .1. Appl. Pliys. 88 (2000) 5941. processing of BLT/Pt and BLT/IrO2 ferroelectric thin films in [8] Y. Hou. X.-H. Xu, H. Wang, M. Wang, S.-X. Shang, Appl. Phys. Lett. 78 order to reduce the annealing temperatures and to improve the (2001) 1733. [9J C.-L. Sun. S.-Y. Chen. S.-B. Chen. A. Chin. Appl. Phys. Lett. SO (2002) electrical characteristics of the BLT films. The oxygen-plasma 1984. treatment gave rise to prominent ferroelectric P-E hystereses [I0J .1. Zhtii. H. Chen. Appl. Phys. Lett, 82 (2002) 442. even for the BLT films that were annealed at the low [1 11 D Bao, r.-W. Chiu. N. Wakiya, K. Shinozaki. N. Mizutani. .1. Appl. Phys. temperatures of 550 and 600 °C. 93 (2003) 497. [1 2] A C. Rastogi, S. Tirumala, S B. Desu, Appl. Phys, Lett. 74 (1999) 3492. [13] H.K. Jang, S.K. Lee, C.L, Lee. S..I. Noh, W.l. Lee, Appl. Phys. Lett. 76 Acknowledgement (2000) 882. [14] W.S. Yang, N.K. Kim. S.J. Yeom. S.V. Kweon, J.S. Roll. Jpn .1. Appl. This work was supported by the Korea Science and Phys. 41 (2002) 727 Engineering foundation (Proton Accelerator User Program 115 j W.S Yang. N K. Kim. S.J. Yeom. S.Y. Kweon, L.S. Choi. J.S. Roll, Jpn. .1, (no. M202AK010021-04A1.101 -02110 and no. RO1-2005- Appl. Phys. 40 (2001) 5569. [16] H.M. Duiker. P.D. Beale, J.F. Scott, C.A, Paz de Araujo. B.M. Melnick, 000-10798-0) and by the Korea Research Foundation (Grant J.D. Cuclnaro. L.D. McMillan, .1. Appl. Phys. 68 (1990) 5783. No. KRF-2004-005-C00060 and Brain Korea 21 Project in [17] T. Miliara, H. Watanabe, C.A. Paz de Araujo, Jpn. .1. Appl. Pliys. 33 2005). The measurements at the Korean Basic Science Institute (1994) 3996; are acknowledged. S.H. Kim. C.H. Lee, K.W. Lee. C.L. Lee. W. Kang, K.S. Hong. Cure. Appl Phys. 4 (2004) 452. [15] T. Miliara, H Watanabe. Jpn. .1. Appl. Phys. 34 (1995) 5674; References C.H. Lee. K.M. Lee, C.L. Lee. W. Kang. Cure. Appl. Phvs. 3 (2003) 359.

[1J C.A. Paz de Araujo, .I D. Cuchlaro, L.D. McMillan, M.C. Scott, IF. Scott, Nature (Lond.) 347 (1995) 627; C.E. Lee. N.S. Dalai. R. Fu, Cure Appl. Phys. 3 (2003) 335. Available online at www.sciencedirect.com INCE^DI Current Applied Physics ELSEVIER Current Applied Physics 6 (2006) 182-184 An official journal of the KM www.elsevier.com/locate/cap www.kps.or.kr

Study of the proton-beam irradiation effects on T1H2P04

Se Hun Kim a, Kyu Won Lee a, Jae Won Jang a, Cheol Eui Lee a’*, K-S. Lee b, S.J. Noh c

“ Department of Physics, Korea University, Seoul 136-713, Republic of Korea b School of Computer-Aided Science, Inje University, Gimhae 621-749, Republic of Korea c Department of Applied Physics, Dankook University, Seoul 140-714, Republic of Korea Received 28 January 2005 Available online 15 August 2005

Abstract

The H+-ion treatment effect on T1H2P04, a KH2P04 (KDP)-type ferroelectric:, was studied by nuclear magnetic resonance (NMR) and AC dielectric constant measurements. A sample of TIH2P04 was irradiated by 1-MeV H+ ion beams to a dose of 1015 ions/cm 2. The change in the NMR relaxation behaviors after the irradiation was attributed that in the hydrogen-bond geom etry, presumably affecting the order-disorder proton dynamics. A prominent decrease in the dielectric constant was also observed after the irradiation. The macroscopic and microscopic changes due to the irradiation are discussed in the light of the proton dynamics. © 2005 Elsevier B.V. All rights reserved.

PACS: 64.70.Kb; 76.60.-k; 77.90.+k

Keywords: Proton beam irradiation; T1H2P04; Ferroelectric:; NMR; Dielectric constant

1. Introduction ric and form zigzag chains along the oaxis. Hydrogens of these bonds are at special positions at a center of TDP undergoes two major phase transitions: antifer- inversion and undergo an order-disorder phase transi roelectric phase transition at Tc = 230 K and ferroelastic tion through the phase transition temperature Tc. The phase transition at T'c = 357 K [1-4], The room temper longest bond, 0.25 nm, is asymmetric along the 6-axis ature phase is paraelectric and ferroelastic, whereas the and the protons are at a general position both above low-temperature phase is known to be antiferroelectric and below Tc [2]. and the high-temperature phase to be paraelectric and The tunneling model including some variants, as a paraelastic. TDP has a monoclinic primitive cell at traditional model for the ferroelectric phase transition room temperature with a = 14.308 A, b = 4.518 A, c = in the hydrogen-bonded ferroelectrics, may be severely 6.516 A, and /? = 91.76° [5.6]. TDP has three different challenged by several experimental and theoretical crystallographic hydrogen bonds as determined by works [7-11 ]. A promising candidate is the theory of X-ray and neutron diffraction, and the crystal structure the so-called geometric isotopic effect, where the shift of TDP illustrating three inequivalent H sites can be in the transition temperature with deuteration is attrib found in the literature [2,6]. The two shorter bonds, uted not to the decrease in tunneling but to the changes 0.243 nm and 0.245 nm, respectively, arc ccntrosymmet- in the hydrogen-bond geometry [10,11]. There are two characteristic lengths in a O-H ■ O hydrogen bond, one being the oxygen separation R (hydrogen-bond Corresponding author. length] and the other being the separation 5 between E-mail address: rscel'Tkotea.ac.kr (C.E. Lee). the two possible hydrogen sites. Several works were

1567-1739/$ - see front matter

~~^ .~~

~~i: ■ w~~

~~Fig. I. Optical micrographs showing (a) surface morphologies of the virgin TDP and (b) proton beam irradiated TDP. 184 S.H. Kim et at. / Current Applied Physics 6 (2006) 182-184~~

an order-disorder transition at a higher temperature than 7j., in agreement with the geometrical isotopic effect theory. The macroscopic changes in the proton-irradiated sample were also revealed by the dielectric constant mea surements. Fig. 3 shows the temperature dependence of the dielectric constant measured at 255 Hz. The pro ton-beam irradiation is observed to decrease the dielec tric constant over the entire temperature, with a characteristic drop at the ferroelastic phase temperature. In summary, we have studied the effect of the hydro gen ion irradiation on a hydrogen-bonded ferroelectrics, T1H2P04, before and after the proton irradiation. The D Before # After 'H rotating frame nuclear magnetic resonance showed that the proton motions are severely hampered after the irradiation. The hydrogen off-centering on the elon gated hydrogen bond may be the origin of the local in 1000/ T (10 3 K'1 ) crease in the antiferroelectric transition temperature, which is consistent with the geometric isotope effect Fig. 2. Temperature dependence of the rotating-frame spin-lattice theory. relaxation rates before (blank) and after (solid) the irradiation. The data were obtained by the stretched-exponential fit below and by the double-exponential fit above T'c. Inset: the exponent n of the stretched-exponential fit. Acknowledgements

This work was supported by the KISTEP (National t r Research Laboratory and Proton Accelerator User Pro □ Before gram No. M202AK010021-04A1101-02110) and by the • After Korea Research Foundation (Grant No. KRF-2004- 005-C00060 and Brain Korea 21 Project in 2004). Mea 1000- surements at the Korean Basic Science Institute (KBSI) are acknowledged.

~~References~~

[1] R. Blinc, B. Zeks, Ferroelectrics 72 (1987) 193. [2] J. Seliger, V. Zagar, R. Blinc, V.H. Schmidt, J. Chem. Phys. 88 (1988) 3260. [3] K. Kanazawa, M. Komukae, T. Osaka, Y. Makita, M. Aral, T. Vagi, J. Phys. Soc. Jpn. 60 (1991) 188. [4] N. Yasuda, S. Fujimoto, T. Asano, Phys. Lett. A 76 (1980) 174. [5] Y. Oddon, A. Tranquard, G. Pepe, Acta Crystallogr. Sect. B 35 Temperature (K) (1979) 542. [6] R J. Nelmes, R.N.P. Choudhary, Solid State Commun. 38 (1981) Fig. 3. Temperature dependence of the dielectric constant in TDP 321. [7] M. Ichikawa, K. Motida. N. Yamada, Phys. Rev. B 36 (1987) R874. after the irradiation was obtained to be Ea = 0.41 cV [8] M.I. McMahon. R.J. Nelmes, W.F. Kuhst, R. Dorwarth. R.O. and 0.59 eV, respectively. After the irradiation, the acti Piltz, Z. Tun. Nature (London) 348 (1990) 317. vation energy required for the proton motion shows a [9] I.V. Stasyuk. R.R. Levitskii. A.P. Moina. Phys. Rev. B 59 (1999) 8530. marked increase, which is consistent with the prolonged 110] S. Tanaka, Phys. Rev. B 42 (1990) 10488, low temperature plateau in 7^'. The paramagnetic (and [11] S. Koval. J. KohanolT. R.L. Migoni, E. Tosatti, Phys. Rev. Lett. presumably charged) defects or the hydrogen off-center 89 (2002) 187602. ing on the elongated hydrogen bonds induced by the [12] C.H. Lee. K.W. Lee, C.E. Lee, K.S. Lee, Phys. Rev. B 55 (1997) irradiation may severely hinder the hydrogen motions 11088; C.E. Lee, C.H. Lee, J.H. Kim, K.S. Lee, Phys. Rev. Lett. 75 or disordering. The regions with elongated hydrogen (1995) 3309. bonds due to the irradiation, with a higher activation [13] S.D. Sctzler, K.T. Stevens, L.E. Halliburton, M. Yan. N.P. energy required for a hydrogen motion, may undergo Zaitseva. J.J. DeYoreo. Phys. Rev. B 57 (1998) 2643. Available online at www.sciencedirect.com Current SCIENCE DIRECT* Applied Physics ELSEVIER Current Applied Physics 6 (2006) 141-144 An official journal of the KPS www.elsevier.com/locate/cap www.kps.or.kr

~~Lateral force microscopy of bamboo-shaped multiwalled carbon nanotubes~~

~~Jae Won Jang a, Cheol Eui Lee a,\ Tae Jae Lee b, Seung Chul Lyu b, Cheol Jin Lee b~~

~~a Department of Physics, Korea University, Seoul 136-701, Republic of Korea b Department of Nanotechnology, Hanyang University, Seoul 133-791, Republic of Korea~~

~~Received 28 January 2005; received in revised form 29 April 2005 Available online 15 August 2005~~

~~Abstract~~

Morphologies of bamboo-shaped multiwalled carbon nanotubes (BS-MWNTs) grown on Fe catalyst deposited onto SiCb/Ti substrates by thermal chemical vapor deposition were investigated by means of atomic force microscopy (AFM) and lateral force microscopy (LFM). For the structural speciality, the bamboo structures were more distinctly observed by LFM. © 2005 Elsevier B.V. All rights reserved.

~~PACS. 68.37.Ps; 81.05.Tp; 81.07.De; 81.40.Pq~~

~~Keywords: Carbon nanotubes; Bamboo-shaped; AFM; LFM~~

1. Introduction materials. For example, adhesion of particles at the dis continuous sheet region can be useful for gas sensor Since the discovery [t], carbon nanotubes (CNTs) and storage applications, and improved mechanical have been attracting great interest because of their stability is expected from the bamboo structure. Fur attractive properties and potential for practical use [2- thermore, the discontinuous graphitic sheet regions 4], Recently, much attention has been paid to a variety may be employed as gate electrodes for sub-pm size of special types of CNTs such as bamboo-shaped [5.6], 1-D transistors. octopus [7], and fish-bone [8 ] structures, different from The growth mechanism of BS-MWNTs has been the conventional CNTs. The bamboo-shaped multi studied by many researchers [14]. For the study of the walled carbon nanotubes (BS-MWNTs) were reported growth mechanism and other properties, investigation in various cases, such as arc-discharge evaporation of of the structural properties should be preceded. Trans graphite [6,9], catalytic pyrolysis of hydrocarbons at mission electron microscopy (TEM) has been typically high temperature [10], thermal chemical vapor deposi used in direct investigation of the structure of CNTs tion on catalytic film deposited substrates [11]. and [15-17], While TEM only gives information on the nitrogen-containing CNTs (CN V nanotubes) [12.13], intrinsic CNT structures, with scanning probe micros The structural uniqueness of the BS-MWNTs can be copy (SPM) one is able to not only observe the struc exploited for various applications as functional nano- tures of the CNTs [18.19] but also measure the electrical properties of the CNTs [20,21 ] and manipulate the CNTs [22.23], Recently, the modified atomic force microscopy (AFM) techniques of SPM, such as scanned Corresponding author. gate microscopy (SGM) [24- 26], electrostatic force E-mail address: vscelid. korca.ac.kr (C.K. Lee). microscopy (EFM), [26] and conductive AFM [21,27]

f itlon :y very. }p~\.v,\r cf-s-: ti-u/C Sfr.Iica:cn i:i a0C:iuriC fo: in Loin; have Ixki: wmi^i I oitl. In A 5,PM sludiex of fSS M'.VN lx /.no l.ixn A\:it ilrbrz-o:. v:i i Maid i yv.

.1. KcmjIU aiul dbCWimi 2. bA|ir/iiiJKiilal :'i;\ 1 displays lhe AFM and LFM uiif.g::- for i"ic In triis wv:k w v.jdicv. :.ie bamboo ?u utvun-x oT BS- xr.mv BS-MWN'J': •arnf.x^ ll ix xlio%v:i t.nv Il.a I .KM MVVfx |y h> M M :-nrl ’.?ilar:ii icnv. ir::myîpy 11 I 'M) nrayc provide* a riryh bvtxr ilr.lihlcd view of lhe mor- The BS-MXVKTs were on Fo r;i:x.^\ depoîto. pholof.v of U o BS-NfW'NTs rhsn no AFM image docs, OHIO S;0:/1: Au::x:!&CCA dl 950 V by tlv.:nTU. diviT.itJ in F!%. .iv.i 2nd »;b;. no :cruj>j/imc.i! bouidii.:; aiv vvifx>r tkpnki inn. Th- AYM ar.il I I'M :r it. wur.-merit iruJiaulfttJ by ihu jrrrr^ 1 4, where;# '.:ie tide rvûci w-cic carried cut vs.n^ y. Park ScicrAitir lnsrtiuncpAv miirkov * 5V c;*n x idoniiricd lo :e ur. urv.Vuvl due to (S*58l: Anl:>Pr»:*« C.I1' opentud al ix>:nn lmipLT.il.nc i;: ':io pyi miiilt.l Af-M tip r.diny i.p oi O.fWn ll:c sample. an^fri: wndtuns vvio$> a pyramid;*I Lpi'I'M Vtu:r^l

~~.V. 5UU AJl~~

£ * i ir; •*. I- V .* « ihi I r V :rryt SS-MV- N < ^»n/ K<* V hy i.xrm;il <.VF> ic .• f.* iiii.-^r"ivi:'* c:* iv: in.'ie-u*.-:. List : :V TCM n..:c ». • va< -• :vri:$v. I*.; :v"rM.n:v i e:v/ujM ./livv^ Ik i:%o - v- mv rti */a * T* si!’

R& < ;u» 7I\M znw « .h: BS Vs. ik.i Ili^h rfscl1. .i?n TCM :c :h?di«r. Vjv.out xh.iriA a lir o .i t:: —ill i'tiM mm. K' Ll'Il 'A l/.:: ti~ vr,A[ ^ . vail'*:. »* indiv* .. U bvurrC7Y> i, 2. Slid S. Wull? jUbecc-tiin .-y-? gri^hiiK Mgii n jr: i..i:k.' 4*:.l H' i. •::o-ui lix inni* arr »Ik- :!i-»•lir nil* *k.vi rvijini* ixjyivMy Thi. xrrnui initiialK d:5frtr.t wi'i Ihivknv.'.ti- mJ lk.-> -JilTrrcr: :vra:i edition; :t.-e\-!tx tv ;k< L"M irctajrcsof the '".cp" of lx? nf lavipl.oiogies of ccmpnrtxcrjts cl the bamboo atmem:?. in «vhjvh the BS-MWNTs u;i:ipirlrn;:n( Jixlar.;x: -.s 3t>-J 00 :i:ri. I'liC l cM i;i age in the hifr:c cf I-ic. !'ci confirms that the AIM irvcced AckifcowJcdgmetits prov,:;^ ,i rr;:’ im;:^ of the* . M2:>2AK;:'HIii:M>IAI Ivl H/.I IUJ and hy C:ie changes cf the c.p arising I;on (he drary.iig fcncitio.is Kcrea Rebirth rounchiiion fBrvin Kuresi 2 Prvjev; and sIoiikn or rhe suifac;: of the sampi: ,2£] Theicf/ic. i.i 2004 and Gram Vo Kft F-2fC/-00?-CCO0^'. different f.urlav? VOr.diti. dis continuous .‘irnphilii* shi*r,Y rc.i'iufx xvnr <*bi^v::il * uvt Kvfcri'iitvs chc v'ompanr.cir.s of the BS-MWM = (29,:< ’|

liv. *>(;,! v< IV VI .iri.y* I.f I he I^.MWMx which l. S. I:. -a Vai iiv ?M .;i?>l) *,*- show* that the diameter at (he corr.pnri:ucol ts narrower ’2; A C. vn. X.M. X?n« T A. DekkeCi.-.l. CII .Xifc.15 L* 5. I him thy I ?i: (he stem Uv v uw^r obierviiiiw of ihr :i:**h. Exlr.nih-. M.; H.'xn. k4aU v 35< •;I$?') )??. t: C. R.H :n, k < k-4 Ir.:*. I .!<. M< I in. Nil..; W< icsoiurion TFzM image of the wall of the CNT% • IW..V-S ;l :y; ic can be le nd i.XcU -.hr wwl n the ovr.rT uvnx l j o>v. r.>..w h i>i rOLghiux; id (he liS VWNTv t'i^. ?(r) .show; ;r m:h?. si v %i':f It.-kor. ;.'||^VF '». 1'I I.B A4VMV C v.lif ^k.iihv:..; I K ^.h'. . V./ /.4,< av.( . rrv.nv picture o' (be dettan box :n Vi;/. 7iul. ./here VY IT/ Poiwva. D.N. Vwcoru Io* V:<>vii>k: sc L'vv*. NA $a>ov. Cho Pkvr. Cr.r. ^l- si .lie *A* ur.il 10 niv.>.> Iks :«.W. >• (!>/$> ?$?. v; BS-MWNT? :c be ciisnuc-.lv oovnvor! by LFM. ;i 1.: : •.**., h i*.»k 1 I’m a r x": i>i.y5 ui i;m:vxi Ki:i I"i xivnrr. ry. y::invr.:: f mho iricrc ^opv was >'• I: X. C. Cc .ox. S. I . rj Ch;m. »v. s plOyeu i l Older to ob'on : mw |'li.Vuz:2-, hml'ro Ui MOi>vt0 3o> dr-ixV .;pr.l. H:..> iMi. VXieX-.VTy *u Vv L S X v. W. L 11. ?, Zh,ic. V. Z'vinz.

[16] D. Goldberg, Y. Bando, L. Bourgeois, K. Kurashima, T. Sato, [23] M R. Falvo, G.J. Clary, R.M. Taylor, V. Chi Jr., F.P. Brooks, S. Carbon 38 (2000) 2017. Washburn, et ah. Nature 389 (1997) 582. [17] S.Q. Feng, D P. Yu, G. Hu, X.F. Zhang, Z. Zhang. J. Phys. [24] S.J. Tans, C. Dekker, Nature 404 (2000) 834. Chem. Solids 58 (1997) 1887. [25] M. Bock rath, W. Liang, D. Bozovic, J.H. Hanker, C.M. Lieber, [18] U. Hubler, P. Jess, H P. Lang, H.J. Guntherodt, J.P. Salvetat, L. M. Tinkham, et ah, Science 291 (2001) 283. Forro, Carbon 36 (1998) 697. [26] A. Bachtold, M.S. Fuhrer, S. Plyasunov. M. Forero, E.H. [19] A. Hassanien, M. Tokumoto, P. Umek, D. Mihailovic, A. Mrzel, Anderson, A. Zettl, et ah, Phys. Rev. Lett. 84 (2000) 6082. Appl. Phys. Lett. 78 (2001) 808, [27] P.J. Pablo, C. Gomez-Navarro, M.T. Martinez, A.M. Benito, [20] J.W.G. Wildoer, L.C. Venema, A.G. Rinzler, R E. Smalley, C. W.K. Maser, J. Colchero, et ah, Appl. Phys. Lett. 80 (2002) 1462. Dekker, Nature 391 (1998) 59. [28] G. Meyer, N.M. Amer, Appl. Phys. Lett. 57 (1990) 2089. [21] H. Dai, E.W. Wong, C M. Lieber, Science 272 (1996) 523. [29] C.J. Lee, J. Park, Appl. Phys. Lett. 77 (2000) 3397. [22] C. Thelander, L. Samuelson, Nanotechnology 13 (2002) 108. [30] C.J. Lee, J. Park, Carbon 39 (2001) 1891. PHYSICAL REVIEW B 73, 012505 (2006)

~~Decoupled critical dynamics of the TMTSF donor molecules in (TMTSF)2A' organic superconductors~~

~~S. H. Kim, K. W. Lee. and Cheol Eui Lee* Department oj Physics and Institute for Nano Science. Korea University. Seoul 136-713. Korea~~

~~W. Kang Department of Physics. Ewha Womans University. Seoul 120-750. Korea~~

~~K. S. Hong Korea Basic Science Institute. Daejeon J05-333. Korea (Received 1 August 2005; revised manuscript received II October 2005; published 11 January 2006)~~

We have studied a high-temperature phase transition in the organic superconductors (TMTSF.)2PFf) and (TMTSF)2C104 (where TMTSF indicates tetramethyltetraselenafulvalene) by means of 'H laboratory-frame and rotating-framc nuclear magnetic resonance (NMR) relaxation measurements. The 'H NMR spin-lattice (T,) and spin-spin relaxation time (T2). representing dynamics of the TMTSF donor molecules, manifested a divergence associated with the structural phase transition at 160 K. As no anomalies were observed in the l9 P NMR 7, measurements representing dynamics of the PF6 union, the TMTSF donor molecules and the anions are shown to be well decoupled regarding the critical fluctuations accompanying the structural phase transition.

~~DOl: 10.1103/PbysRevB.73.01 2505 PACS number(s): 74.70.Kn, 75.30.Fv, 76.60.-k, 78.30.Jw~~

1. INTRODUCTION with an interlayer shift. An unexpected high-temperature anomaly near 160 K, an abrupt increase in the temperature The Bechgaard salts (TMTSF):X, where TMTSF is tet- coefficient of resistance, has also been reported in the ramclhyltetraselenafulvalene. show various ground states.1 (TMTTF)2A/F6 compounds (M = P. As.Sb).12 The role of the One of them, superconductivity, has been found in anion motion as well as the molecular dynamics of the (TMTSF)2Y with monovalent inorganic anions X such as TMTSF donor molecules need to be understood in associa PF6, ReO.,, or C104.2 The anions are placed in centrosym- tion with it.'2 We have been able to observe NMR and elec metrical cavities confined by the methyl groups of the tron paramagnetic resonance as well as electrical conductiv TMTSF molecules forming dimerized pairs. While the ity anomalies in (TMTSF)2C104, at around 160 K.1314 In (TMTSF)jY family members are quasi-onc-dimensional this w ork, the 'H NMR rotating-framc as well as laboratory- metals at high temperatures, their electrical conduction is frame nuclear spin relaxation measurements were employed known lo be two or three dimensional at low temperatures.14 in order to probe the molecular dynamics associated with the The anion X, sitting on sites possessing an inversion symme TMTSF donor molecules in (TMTSF)2C104 and try, is believed to undergo rotational motions at high tem (TMTSF),PFCv It is the purpose of this work to elucidate the peratures. The anions interact weakly with surrounding do nature of (he structural phase transition at 160 K in nor molecules ,5 and weak hydrogen bonds between the (TMTSFhY with different anions by means of (he 'H NMR anions and TMTSF donor molecules are present .6 Through nuclear spin relaxation measurements. competing electronic instabilities, the anion sublattice may play an important role in the rich phase diagram of the Bech gaard salts.7 II. EXPERIMENT NMR is a powerful tool for studying the lattice dynamics and microscopic environments in solids, and has been em Electrochemically grown samples of (TMTSFHPF,, and ployed to a large extent in order to investigate the spin dy (TMTSF)2CI04 were investigated by using 200 Mllz and namics in (TMTSFhX (A'=PU6,C1C)4). In particular, in a de 45 MHz NMR spectrometers as a function of temperature. tailed study of the relaxation data, reorientation of I he methyl The Tl NMR spin-lattice relaxation time (T",) was measured groups has been discussed considering different environ by the inversion recovery method, and the spin-spin relax ments. Besides, it was thus found that the 'ft NMR spin- ation time (7Y) by the solid echo decay pulse sequence. lattice relaxation is dictated by lire hyperline coupling of the charge carriers at high temperatures above 200 K. whereas III. RESULT AND DISCUSSION reorientation motion of the methyl groups is responsible be low ii as indicated by the frequency dependence of the re The magnetization recovery followed a single-exponential laxation rate.8 "411 form. Figure I shows the temperature dependence of the 'll A structural phase transition in (TMTSFbPF,, lias been NMR spin-lattice relaxation rate 17", '). A divergent behavior reported in (lie vicinity of 160 K by x-ray measurements,4 in the rotating-framc as well as in the laboratory-frame spin- which revealed a change in the f> angle by «=!’■', associated lattice relaxation measurements is manifest in Fig. I at

~~1095-012 1/2006/73IIW12505131/523 00 012505 I syyuop The American Musical Socieiv BRIEF REPORTS PHYSICAL. REVIEW B 73, 012505 (2006)~~

~~O PF, Z " 55.3 kHz~~

~~a pfr r:1200 MHz~~

~~0 20 40 60 80~~

~~Temperature (K) Temperature (K)~~

FIG. 1. Temperature dependence of tire laboratory-frame and FIG. 2. Temperature dependence of the spin-spin relaxation rate rotating-frame 'H NMR spin-lattice relaxation rates in in (TMTSF) TF),. Inset: The solid echo decay pattern at 178 K, (TMTSF)2PF6 and (TMTSF)2C104. The dotted line is a guide to the titled by a Gaussian form. eye. ported by the l9 F NMR measurements of (TMTSF)2PF6.1718 In contrast to the case of the 'll NMR nuclear spin relaxation around 160 K, which is attributed to a structural transition. I j measurements, no anomaly was found in the temperature de The methyl (CHj groups in the TMTSF donor molecule, pendence of the l9 F NMR spin-lattice relaxation rate probing being embedded in a rigid lattice, undergo a hindered the reorientational motions of the PF6 anion molecules .1718 rotation .15,16 In fact, a relatively strong frequency depen While NMR measurements have been made in dence as shown in Fig. 1 indicates that the anomaly arises (TMTSF)2CI04 and (TMTSF)2PF6 by various workers, 8' "1 from molecular dynamics rather than being of an electronic close investigation has not been done by other workers fo origin. As the reorientational motion of the methyl groups is cused on the structural phase transition, in the temperature expected to make a significant contribution to the !H NMR range around 160 K. We have reported NMR anomalies in nuclear spin relaxation, the critical fluctuation manifested by (TMTSF)2C104 and in (TMTSF),PF6. the divergence in the spin-lattice relaxation at 160 K can be In summary, the organic superconductors (TMTSF)2PF6 ascribed to the structural phase transition arising from the and (TMTSF)2C104 were studied by means of the 'H NMR tilting of the TMTSF donor molecules as described above. nuclear spin relaxation measurements. As a result, a struc As only relatively small peaks are observed at 160 K. the tural phase transition associated with the TMTSF donor mol structural phase transition is most likely to be of a weak first ecules was identified al 160 K, in the two systems with the order, which may give rise to a divergence in the relaxation same donor molecules but with different anions, whereas no rate. anomalies associated with the anions have been observed. Figure 2 shows the temperature dependence of the spin- Thus, our work shows that the structural phase transition at spin relaxation rate (7T1) in (TMTSF)2PF6. As shown in the 160 K for the TMTSF donor molecules in those organic su inset, the solid echo decay was well fitted by a Gaussian perconductors takes place well decoupled from the anions. form at all temperatures, except around 160 K. The spin-spin relaxation anomaly at 160 K can also be attributed to the ACKNOWLEDGMENTS change in the reorientational motions of the methyl groups in This work was supported by the Korea Science and Engi the TMTSF donor molecules, arising from the structural neering Foundation (Proton Accelerator User Program) phase transition. (Grants No. M202AK010021-04A1101-02110 and No. ROI- While it may seem very strange to find exactly the same 2005-000-10798-0) and by the Korea Research Foundation 7',. for the PF6 and C104 salts, this in fact strongly suggests (Grant No. KRF-2004-005-C00060 and Brain Korea 21 that the critical dynamics of the TMTSF donors is decoupled Project). The measurements at the Korean Basic Science In quite well from the anions. This point can be further sup stitute are acknowledged.

♦Corresponding author. Electronic address: rscel#korea.ac.ki N. Thorup. Solid State Conimun. 33. 1119 119801. 1 V. J. Emery, R. Bruin.-una. and S. Barisic. Rhys. Rev. Lett. 48. D. Jerome and 11. J, Schulz. Adv. Phys. 31. 299 (1982): D. Jer 1039 (1982). ome. P. Auhan-Senzier. L. Balicas. K. Behnia. W. Kang, P. VVzi- 2K. Becheaiuxl. C. S. Jacobson. K. Mortensen, J. H. Pederson, and etek. C. Berthier. P. Caretta, M. Horvalic. P. Segransa. L. Hu

~~012505-2 BRIEF REPORTS PHYSICAL REVIEW B 73, 012505 (2000)~~

bert, and C Bourbonnais, Synth. Met. 711. 719 (1995). 1:1. C. Scott. E. M. Engler. W. G. Clark. C. Muroyama. K. Bech- 4 A. Schwartz, M. Dressel, G. Grimer, V. Vescoli. L. Degiorgi. and gaaul. and H. J. Pedersen, in Pwceedings o f the International T. Giamarchi, Pliys. Rev. B 58, 1261 (1998); J. Moser. M. Conference on latte-Dimensional Conductors. Boulder, 198 1 Gabay. P. Auban-Senzier. D. Jerome. K. Bechgaard. and .1. M. [Mol. Cryst. Liq. Cryst. 79. 61 (1982)]. Fabre, Em. Plays. J. B 1, 39 (1998). |:C. Coulon, S. S. P Parkin, and R. Laversanne. Phys. Rev. B 31. 5B. Gallois, J. Gaultier, C. Hauvv, and T. D. Lamcharfi. Ada Crvs- 3583 (1985). tallogr.. Sect. B: Struct, Set. 42. 564 (1986). '■'C. H. Lee. K. M. Lee. C. E. Lee, and W. Kang. Curt. Appl. Phys. 6J. P. Pouget and S. Ravy, J. Phys. 1 6, 1501 (1996). 1C. Bourbonnais, P Wzietek. F. Creuzet, D. Jerome, P Batail. and 3, 359 (2003). K. Bechgaard, Phys. Rev. Lett. 62. 1532 (1989); K. Sengupta 14C. E. Lee, C. H. Lee, W. Kang, and O. H. Chung, Solid State and N. Dupuis. Phys. Rev. B 65, 035108 (2002); C. Coulon, P. Common. 109. 69 (1999). Delhaes, S. Flandrois, R. Lagnier. E. Bonjour, J. M. Fabre. and !1C. Choi and M. M. Pintar, J. Chern. Phys. 109. 5542 (1998). K. Bechgaard, J. Phys. (Paris) 43, 1059 (1982). I6J, C. Scott, H. J. Pedersen, and K. Bechgaard, Phys. Rev. B 24. *P. C. Stein, A. Moradpour, and D. Jerome. J. Phys. (Pans). Lett. 475 (1981). 46, 1.241 (1985). 17V. J. Mcbrierty, D. C. Douglass, and F Wudl. Solid State 9P Wzietek, F. Creuzet, C. Bourbonnais, D. Jerome. K. Bechgaard. Common. 43, 679 (1982). and P Batail, J. Phys. I 3, 171 (1993). '\S. H. Kim, C. H. Lee, K. W. Lee, C. E. Lee, W, Kang, and K. S. I0T. Takahashi. D. Jerome, and K. Bechgaard. J. Phys. (Paris) 46. Hong, Curr. Appl. Phys. 4, 452 (2004). 945 (1984). Journal of the Korean Physical Society, Vol. 47, No. 2, August 2005, pp. 337~338

~~Magnetic Moment in Proton-Irradiated Graphite~~

~~Kyu Won Lee, Young-Ho Lee, In-Mook Kim and Cheol Eui Lee* Department of Physics and Institute for Nano Science, Korea University, Seoul 136-713~~

~~(Received 13 June 2005)~~

In an attempt to understand the origin of the magnetic moment in proton-irradiated graphite, we have studied the proton irradiation effect in highly oriented pyrolytic graphite (HOPG). A low- dose irradiation induced Curie paramagnetism in HOPG, and the magnetic moment per irradiated proton was estimated to be ~2.4^b.

~~PACS numbers: 78.70.-g, 75.50.-y Keywords: Graphite, Ferromagnetism, Magnetic moment, Proton irradiation~~

I. INTRODUCTION Curie paramagnetism in HOPG. Thus, the magnetic mo ment per irradiated proton could be determined by using Recently, ferromagnetism has been reported to be the Curie law. induced by proton irradiation in highly oriented py rolytic graphite (HOPG) [1-4]. Among a great deal of irradiation studies [5-7], the irradiation-induced fer II. EXPERIMENT romagnetism in graphite is of particular interest in view of applications as well as science. The proton- The HOPG (ZYA grade) samples with dimensions irradiation studies of HOPG revealed that ferromag of 2.5 x 2.5 x 0.4 min3, purchased from the Struc netism in carbon-based materials, with a Curie temper ture Probe Inc,., were irradiated with a 2.25-MeV proton ature well above room temperature, is closely related to beam with a spot radius 0.5 cm to a dose of 1.875 x the hydrogens adsorbed, opening a possibility for micro- 1017 ions/cm 2 at room temperature. The magnetization patterned ferromagnets of light elements. Helium-ion ir was measured employing a superconducting quantum in radiation produced a very weak magnetic signal, in con terference device (SQUID) magnetometer (Quantum De trast to the case of proton irradiation, indicating the sign MPMS series). Before the irradiation, only the G decisive role of the hydrogens [4], which was supported (graphitic) band was observed in the Raman spectrum by some theoretical works [8,9]. In particular, hydro at room temperature whereas after the irradiation, the genated graphite was theoretically predicted to display D (disorder) band appeared in addition to the G band, spontaneous magnetization arising from different num with an areal fraction D/G ~ 1/5. bers of mono- and double-hydrogenated carbon atoms [8 ] . Hydrogen adsorption on the irradiation-induced car bon vacancy was theoretically predicted to give rise to III. RESULTS AND DISCUSSION a magnetic moment double that of the naked vacancy [9] , Ferromagnetism was also observed in other forms Figure 1 shows the magnetization of the irradiated of carbon-based materials, such as amorphous carbon HOPG, measured with a magnetic field of 1 T perpen and C60 polymers [10-12], In a hydrogen-free carbon dicular to the c-axis. Three types of contributions to nanofoam with hyperbolically curved graphitic sheets, the magnetization were measured with the magnetic field ferromagnetic-like behaviors were observed [13], which perpendicular to the c-axis: the Curie paramagnetic sus indicates that different forms of carbon-based materials ceptibility ( \c ) of the isolated magnetic moment induced other than HOPG may have different origins of the fer by the proton irradiation, the Pauli paramagnetic sus romagnetism [13-15]. ceptibility {\p) due to the conduction electrons, and the While it is evident that hydrogens play a decisive role diamagnetic susceptibility (\d ) due to the valence elec in the ferromagnetism in graphite, the origin of the mag trons. \C L inversely proportional to the temperature, netic moment remains unclear. In this work, HOPG was \p is proportional to the temperature [16], and \d is in irradiated with a low dose of protons, which was expected dependent of the temperature. Thus, the magnetization to induce many isolated magnetic moments, leading to of the irradiated HOPG can be approximated as

~~"Corresponding Author ; rscc)@korca.nc.kr 4 AT+G/T. (1) -33 -338- Journal of the Korean Physical Society, Vol. 47, No. 2, August 2005~~

~~Woo at the Korea Institute of Geoscience and Mineral Resources for the proton-beam irradiation. The mea surements at the Korean Basic Science Institute are acknowledged.~~

~~REFERENCES~~

[1] P. Esquinazi, D. Spemann, R. Hohne, A. Setzer, K.-H. Han and T. Butz, Phys. Rev. Lett. 91, 227201 (2003); C. E. Lee, N. S. Dalai and R. Fu, Curr. Appl. Phys. 3, 355 (2003). [2] P. Esquinazi, A. Setzer, R. Hohne, C. Semmehack, Y. Kopelevich, D. Spemann, T. Butz, B. Kohlstrunk and M. Losche, Phys. Rev. B 66, 024429 (2002); C. H. Lee, K. W. Lee and C. E. Lee, Curr. Appl. Phys. 3, 477 (2003). [3] D. Spemann, P. Esquinazi, R. Hohne, A. Setzer, M. Dia- conu, H. Schmidt and T. Butz, Nucl. Instr. Meth. B, in press (2005); C. H. Lee, K. M. Lee, C. E. Lee and W. Fig. 1. Magnetization of the irradiated HOPG, measured Kang, Curr. Appl. Phys. 3, 359 (2003). with a magnetic field of 1 T perpendicular to the c-axis. The [4] K. H. Han, D. Spemann, P. Esquinazi, R. Hohne, V. Riede solid line is a fit to Eq. (1). and T. Butz, Adv. Mater. 15, 1719 (2003); S. H. Kim, C. H. Lee, K. W. Lee, C. E. Lee, W. Kang and K. S. Hong, Curr. Appl. Phys. 4, 452 (2004). [5] F. Stobiecki, B. Szymariski, M. Urbaniak, T. Lucinski, J. The solid line in Figure 1 is a fit to Eq. (1) Dubowik, M. Kopcewicz, J. Jagielski and Y. P. Lee, J. with xo = —1.6 x 1CT2 emu/gT, A = 1.8 x 10"5 Korean Phys. Soc. 45, 1 (2004). emu/gTK, and C = 1.23 x 10-2 emuK/gT. From the [6] H. S. Kim, Y. Kim, S. J. Noli and M. H. Kim, J. Korean Curie constant C, the magnetic moment per irradiated Phys. Soc. 45, 820 (2004). proton was estimated to be //. ~ 2.4/rg. According to a [7] J. Kim, W. Hong, H. J. Woo and C. H. Eum, J. Korean previous theoretical work [9], hydrogen adsorbed directly Phys. Soc. 43, 582 (2003). into the carbon vacancy gives a magnetic moment of 2.3 [8 ] K. Kusakabe and M. Maruyama, Phys. Rev. B 67, 092406 Hb, in good agreement with our estimate, the hydrogen (2003). configuration being taken to be metastable in the same [9] P. 0. Lehtinen, A. S. Foster, Y. Ma, A. V. Krasheninnikov and R. Nieminen, Phys. Rev. Lett. 93, 187202 (2004). work. [10] K. Murata, II. Ushijima, H. Ueda and I\. Kawaguchi, J. In summary, we have studied the proton-irradiation ef Chem. Soc. Chern. Common. 1265 (1991). fect in highly oriented pyrolytic graphite. As a result, the [11] K. Murata, H. Ushijima, II. Ueda and K. Kawaguchi, J. magnetic moment per irradiated proton was estimated to Chem. Soc. Chem. Commun. 567 (1992). be ~ 2.4 Hb- [12] T. L. Makarova, B. Sundqvist, R. Hohne, P. Esquinazi, Y. Kopelevich, P. Scharff, V. A. Davydov, L. S. Kashe- varova and A. V. Rakhmanina, Nature 413, 716 (2001). ACKNOWLEDGMENTS [13] A. V. Rode, E. G. Garnaly, A. G. Christy, J. G. Fitz Gerald, S. T. Hyde, R. G. Elliman, B. Luther-Davies, A. I. Veinger, J. Androulakis and J. Giapintzakis, Phys. Rev. This work was supported by the Korea Science B 70, 054407 (2004). and Engineering foundation (Proton Accelerator User [14] A. A. Ovchinnikov and V. N. Spector, Synth. Met. 27, Program No. M202AK010021-04A1101-02110 and No. 615 (1988). RO1-2005-000-10798-0) and by the Korea Research [15] K. Wakabayashi, M. Fujita, H. Ajiki and M. Sigrist, Foundation (Grant No. KRF-2004-005-C00060 and Phys. Rev. B 59, 8271 (1999). Brain Korea 21 Project in 2005). We thank Dr. H-.J. [16] G. Wagoner, Phys. Rev. 118, 647 (1960). Journal of the Korean Physical Society, Vol. 47, No. 6, December 2005, pp. 1084~I086 Brief Reports

~~Rotating-Frame Nuclear Magnetic Relaxation in a Long-Chain Model~~

~~Biomembrane (Ci8H37NH3)2SnCI6~~

~~Kyu Won Lee and Cheol Eui Lee* Department of Physics and Institute for Nano Science, Korea University, Seoul 136-713~~

~~J. Y. Choi Department Computer Science and Institute for Nano Science, Korea University, Seoul 136-713~~

Joon Kim*

~~School of Life Sciences and Biotechnology, Korea University, Seoul 136-713~~

~~(Received 23 August 2005)~~

The lattice dynamics in bis^n-CigHsyNHa^SnCle, where the hydrocarbon part is analogous to a lipid membrane, was investigated by means of 55-kHz *H rotating-frame nuclear magnetic resonance. The critical fluctuation near the order-disorder transition of the rigid alkyl chains is attributed to the high-frequency dynamics for long-chain compounds, in contrast to the low-frequency dynamics for short-chain compounds. Besides, the non-exponential spin-lattice relaxation is ascribed to chain segmental motions, giving rise to a distribution of correlation times and freezing below 230 K.

~~PACS numbers: 64.70.kb, 76.60.-k Keywords: NMR, Lipid membrane, Low-frequency dynamics~~

I. INTRODUCTION chain about the chain axis, as well as a chain-end gauche, takes place. In the high-temperature (HT) phase above the conformational transition temperature, the average The bis-n-alkylammonium hexachlorostannates (n- chain axis is destroyed, and the chains undergo liquid-like CmH2m+iNH3)2SnCl6 (CmSn for short) are layered isotropic motions. The three phases and the two phase compounds of alternating organic and inorganic layers. transitions are analogous to those in hydrated lipid mem The SnClg 2- octahedra do not form a 2D macroanion, branes [11-17]. but exist separately [1-3]. The NH3 group of the alky- In C18Sn, the order-disorder and the conformational 1 ammonium ion links the three closest octahedra through transitions were observed at 347 K and 371 K, respec equivalent hydrogen bonds of a N-H- • Cl type, forming tively [4, 5], A slow chain segmental motion, as well an inorganic layer. The distance between the ammo as molecular motions, such as NH3 and CH3 rotations, nium groups or between the tin atoms in CroSn is great was detected in the Larmor frequency dependence of the enough (7.3 ~ 7.5 A, depending on the chain length) for laboratory-frame spin-lattice relaxation time [5], The *H the interdigitated alkyl chains to form a lipid monolayer. NMR second-moment measurements indicated freezing In our previous *H NMR studies of the CmSn systems, of the slow chain segmental motion below 230 K. In this two typical successive phase transitions were observed work, the 'H NMR rotating-frame spin-lattice relaxation [4-10]: a) an order-disorder transition of the rigid alkyl time (Tip) was measured to study the low-frequency dy chains accompanied bv a uniaxial reorientation about namics in C18Sn. their long axes along with a flipping of polar groups, such as NH.3, and b) a conformational transition leading to a partial melting of the alkyl chain part. In the low- temperature (LT) phase below the order-disorder transi II. EXPERIMENT tion temperature, the alkyl chains are rigid and adopt an all-trans conformation. In the intermediate (IT) phase The ClSSn sample used in this work was synthesized, between the order-disorder and the conformational tran with much care to avoid impurities by using the chemi sition temperatures, a uniaxial rotation of the rigid alkyl cal reaction 2(n-CigH 3TNH3Cl) + SnCl.r SHjO —> (n- C,gH 3-NH3)2SnClQ + SIDO. After filtering and two re- * Corresponding Author : rsccl'Qkoro&.ac.kr crystallizations, white sugar-like crystals were finally ob t E-mail: joonkimQkorca.ar.kr tained, then vacuum-dried, and kept in a dry condition -1081 Rotating-Prame Nuclear Magnetic Relaxation- ■ ■ - Kyu Won Lee et al. -1085-

for further works. The stoichiometry and the structure sharp divergence [7,8]. In contrast to the case of m < 14, were checked by elemental analysis and X-ray diffraction the critical fluctuation near the order-disorder transition (XRD). Differential scanning calorimetry (DSC) carried was reflected in the laboratory-frame spin-lattice relax out between 123 K and 453 K shows two reversible ation in C16Sn and C18Sn; however, as shown in Fig. 1, phase transitions. The rotating-frame spin-lattice relax it is not so prominent in the rotating frame spin-lattice ation time was measured at the frequency of the rotating relaxation [4,5,9], Thus, it appears that low-frequency frame, 55 kHz, by using a BRUKER MSL 200 spectrom dynamics for m < 14 and high-frequency dynamics for eter. m > 14, respectively, are responsible for the critical fluc tuation near the order-disorder transition. In Fig. 1, besides the anomalies at the two phase III. RESULTS AND DISCUSSION transitions, the divergent behavior in T^1 in the vicinity of 230 K, below which temperature freezing of the slow chain segmental motion was observed, is interesting [5], The rotating-frame spin-lattice relaxation shows The divergent behavior, presumably reflecting a critical a stretched-exponential pattern, M(t) = M(0) exp fluctuation, can be attributed to a collective slow chain [—(t/Tlp)n \, below the conformational transition tem segmental motion. In the low-temperature phase be perature, and a single-exponential form above it. The low the order-disorder transition temperature, the alkyl spin-lattice relaxation pattern in C18Sn near the order- chains are rigid and adopt an all-trans conformation as disorder transition, temperature depends on the Lar- a whole. The stretched-exponential type of spin-lattice mor frequency. At 200 MHz, the spin-lattice relax relaxation indicates that chain segments with distinct ation pattern is a double-exponential form, for which correlation times are spatially separated well enough to the shorter time constant only reflects a critical fluctu disturb the nuclear spin diffusion among them. An addi ation near the order-disorder transition temperature [4], tional anomalous behavior of T^1, a small discontinuity The spin-lattice relaxation follows a single-exponential around 265 K, as well as the full low-frequency dynamics form at 45 MHz, showing a divergent anomaly near the of the complicated chain motions, are not understood at order-disorder transition temperature [5]. The stretched- present. exponential type of decay usually indicates a distribution Fig. 2 shows the temperature dependence of the ex of the correlation times or a distribution of the spin- ponent n for the stretched exponential type of rotating- lattice relaxation times. Slow, complicated chain seg frame spin-lattice relaxation. Below the conformational mental motions may be responsible for the stretched- transition at 371 K, n is much smaller than 1, with n = 1 exponential type of decay. corresponding to a single-exponential decay with a single The critical fluctuation near the order-disorder transi spin-lattice relaxation time and, thus, to a single corre tion was not reflected in the laboratory-frame spin-lattice lation time, n, with a minimum value of 0.45 around relaxation time of CmSn with m < 14, but was reflected 275 K, can be taken as a measure of the correlation time in their rotating-frame spin-lattice relaxation time as a

~~35- d o 309 % i<9 25~~

~~15- d3~~

~~□ □ □ 10-~~

~~5 150 200 250 300 350 400~~

~~Temperature (K.) Temperature (X)~~

Fig. 1. Rotating-frame spin-lattice relaxation rate (Tlp ) Fig. 2. Exponent n for the rotating-frame spin-lattice re as a function of temperature. laxation as a function of temperature. -1086- Journal of the Korean Physical Society, Vol. 47, No. 6, December 2005 distribution for mobile chain segments. Below about 230 2] K. Kitaharna, H. Kiriyama and Y. Baba, Bull. Chem. Soc. K, n is nearly constant with a value of 0.6 whereas it Jpn. 52, 324 (1979). shows an increase with temperature above about 275 K. 3] M. H. B. Ghozlen, A. Daoud, T. Molk, H. Poulet, M. Le Finally, n jumps discontinuously to 1 at the conforma Postllec and N. Toupry, ,J. Raman Spec. 16, 219 (1985). tional transition, where all the chains undergo liquid-like [4] K. W. Lee, C. II. Lee, C. E. Lee and J. K. Kang, Phys. isotropic motions. Rev. B 54, 8989 (1996); C. E. Lee, N. S. Dalai and R. Fu, Curr. Appl. Phys. 3, 355 (2003). In summary, we have studied the low-frequency [5] K. W. Lee, C. H. Lee, C. E. Lee and J. K. Kang, J. Chem. dynamics in (n-CmH2m+1NH3)2SnCl6 by using Phys. 104, 6964 (1996); S. H. Kim, C. H. Lee, K. W. Lee, rotating-frame nuclear magnetic relaxation time mea C. E. Lee, W. Kang and K. S. Hong, Curr. Appl. Phys. surements. The low-frequency dynamics for rn < 14 and 4, 452 (2004). the high-frequency dynamics for m > 14 appear to be 16] K. W. Lee, M. W. Park, C. Rhee, C. E. Lee, J. K. Kang, responsible for the critical fluctuation near the order- K. W. Kim and K. S. Lee, J. Chem. Phys. 108, 3019 disorder transition. The slow chain segmental motions (1998); C. H. Lee, K. M. Lee, C. E. Lee and W. Kang, in C18Sn are believed to weaken the critical fluctuation Curr. Appl. Phys. 3, 359 (2003). near the order-disorder transition in the rotating-frame 7] K. W. Lee, C. E. Lee, J. Kim and J. K. Kang, Solid State spin-lattice relaxation. A critical behavior, which was at Commun. 124, 185 (2002). [8 ] K. W. Lee, D. K. Oh and C. E. Lee, J. Phys. Soc. Jpn. tributable to the freezing of mobile chain segments, was 72, 2398 (2003). observed below 230 K. [9] K. W. Lee, C. E. Lee, J. Y. Choi and J. Kim, Solid State Commun. 133, 83 (2005). [10] K. W. Lee, C. E. Lee, J. Y. Choi and J. Kim, J. Korean ACKNOWLEDGMENTS Phys. Soc. 46, 245 (2005). [11] M. J. Janiak, D. M. Small and G. G. Shipley, J. Biol. This work was supported by the Korea Science and En Chem. 254, 6068 (1979). gineering foundation (Proton Accelerator User Program [12] R. Koynova and M. Caffrey, Biochim. Biophys. Acta 1376, 91 (1998). (No. M202AK010021-04A1101-02110 and No. ROl- [13] P. Meier, E. Ohmes and G. Kothe, J. Chem. Phys. 85, 2005-000-10798-0)) and by the Korea Research Foun 3598 (1986). dation (Grant No. KRF-2004-005-C00060/D00057 and [14] R. M. Venable, B. R. Brooks and R. W. Pastor, J. Chem. Brain Korea 21 Project in 2005). The measurements at Phys. 112, 4822 (2000). the Korean Basic Science Institute are acknowledged. [15] S. Konig, E. Sackmann, D. Richter, R. Zorn, C. Carlile and T. M. Bayerl, J. Chem. Phys. 100, 3307 (1994). [16] J. Sung, K. Park and D. Kim, J. Korean Phys. Soc. 44, REFERENCES 1394 (2004). [17] Y. H. Chang, C. H. Park and K. Matsuishi, J. Korean Phys. Soc. 44, 889 (2004). [1] O. K. Knop and W. J. Westerhaus, Can. J. Chem. 58, 270 (1980); C. H. Lee, K. W. Lee and C. E. Lee, Curr. Appl. Phys. 3, 477 (2003). Journal of the Korean Physical Society, Vol. 47, No. 4, October 2005, pp. 650~654

~~Critical Dynamics in SnCh 2H20~~

~~Kyu Won Lee and Cheol Eui Lee* Department of Physics and Institute for Nano Science, Korea University, Seoul 136-713~~

~~(Received 4 August 2005)~~

Our 1H NMR spin-spin relaxation time (Tsb) measurement in SnCL 2H20 (SCD) shows a be havior similar to that of the electrical conductivity, which indicates that they are determined by the same proton dynamics. A critical decrease in the correlation time was observed around the critical temperature and was attributed to the long-range order of soliton motion.

~~PACS numbers: 64.70.Kb, 76.60.-k Keywords: Stannous chloride dihydrate, Soliton motion, Critical dynamics, Nuclear magnetic relaxation~~

I. INTRODUCTION lar, the equilibrium statistical mechanics of two model Hamiltonians supporting soliton solutions, the c/>4 model Recently, proton motions in a hydrogen bonded net and the sine-Gordon model, have been discussed in great work have attracted much attention. In ice, biological detail [12,13], systems, and many hydrogen bonded ferroelectric mate Stannous chloride dihydrate (SnClg 2HgO, or SCD) rials, cooperative and collective proton motions are re undergoes a peculiar isostructural and order-disorder- sponsible for static and dynamic properties, such as con type phase transition at about 218 K [14], The two- ductivity, dielectric relaxation, and optical spectra [1- dimensional nature of the phase transition has been ob 3]. The soliton model has been suggested to account for served from heat capacity measurements, which show the protonic conductivity [4,5]. Translation of protons a symmetric divergence [15] and from neutron diffrac along hydrogen bonds in a double well potential creates tion and X-ray studies, which show that it undergoes an ionic defect, and the rotation of the water molecule no structural change through the phase transition [14, creates a Bjerrum defect. Some workers have suggested 16,17], These uncommon behaviors have drawn particu a permanent flow of solitons, giving rise to a nontran lar attention to the study of the static and the dynamic sient proton current in a hydrogen-bonded network [6- properties of the phase transition in SCD, which is com 8 ]. Nevertheless, while the soliton model is about the posed of alternating layers of SnClg and H20 [18], Some only plausible ionic conduction mechanism in hydrogen- ambiguities are found in the critical dynamics reflected bonded systems, it is difficult to identify a soliton contri in the *H NMR spin-lattice relaxation time [19,20] as de bution from among others because various contributions scribed by the pseudospin Ising model. Besides, the crit can induce similar noncritical behaviors. However, near ical behaviors of the spin-spin relaxation time [21] and the critical temperature, a soliton contribution, such as a the conductivity [22,23] obviously cannot be explained central peak, may manifest its presence in a drastic man by that model. ner [9], While the electrical conductivity measurements SCD can be considered as a quasi-one-dimensional represent macroscopic proton transport, measurements ionic conductor. One well-known property of hydrogen- of the *H NMR (nuclear magnetic resonance) spin-spin bonded protonic conductors is the spontaneous forma relaxation time by using solid echo decay can rep tion of ionic and Bjerrum type defects. It has been re resent two-spin correlations associated with microscopic cently shown that when the dynamics of protons in a proton dynamics and can be used to throw some light hydrogen-bonded quasi-one-dimensional network is de onto the origins of protonic conductivities. scribed in terms of a diatomic lattice model with a doubly A great deal of interest has been generated in the soli periodic on-site potential, the discrete system is reduced tary wave solutions of nonlinear field theory. Solitons. to a continuum double sine-Gordon equation for the pro which have come to refer to any solitary waves that prop tonic part plus an easily solvable differential equation lor agate without changing shape or velocity, have been in the heavy part [13]. In this case, its two-component kink strumental in discussing a variety of phenomena in con solutions were shown to correspond to ionic and Bjer densed matter physics, including superionic conduction rum defects. Hydrogen-bonded networks, such as those and structural phase transitions [10, 11]. In particu in SCD, are usually modelled by the association with each proton of a doubly degenerate proton position per unit cell. Propagation of the ionic defect lias been con 'Corresponding Author : rscd akoi ca.ar.kr -(All- Critical Dynamics in SnCH 2HgO - Kyu Won Lee and Oheol Eui Lee -651- nected to a collective soliton dynamics by various workers which gave a room temperature correlation time of 2 [6-8 ]. It, thus, follows that soliton dynamics can play x 1CT7 s when fitted by using an intramolecular dipole- a key role in the protonic conductivities in hydrogen- dipole interaction with one correlation time. Previous bonded systems like SCD. A model description has also low-temperature measurements yielded much longer cor shown that a permanent flow of solitons can take place, relation times [19,20]. The Tie values were extracted which is necessary for a nontransient proton current in a from the slope of the solid echo decay fitted to a single hydrogen-bonded network [6,13]. exponential form as represented in Fig. 2, which indi As in many hydrogen-bonded systems, the significant cates that the single-exponential fit can be made un DC conductivities in the SCD system have been at ambiguously. In Fig. 3 are shown our temperature- tributed to proton motions. Mognaschi et al. have sug dependent measurements of T^e obtained from the slope gested contributions to the protonic conductivities from of the solid echo decay, along with the conductivity data the translation of protons along a hydrogen bond in a taken from a previous work. A very similar temperature double well potential and the three-fold rotation of the dependence and an anomalous increase in the critical re water molecule and from the critical dynamics described gion are observed in the two quantities. by the pseudospin Ising model [22,23]. However, they were only able to explain the conductivities in the noncritical high-temperature region, where the protonic con 70 ___ ductivity follows an Arrhenius-type temperature depen dence, a oc e~E!RT. On the other hand, the normally ex '.ft _ pected critical slowing down can not explain the anoma lous increase of the conductivity near Tc, which indicates 511- a decrease in the proton correlation time. In other words, the electrical conductivity is expected to be proportional to the protonic jumping rate [22, 23] or inversely pro 40- portional to the correlation time of the proton motion. »(]- Then, a sharp increase in the protonic conductivity cor responding to the decrease in the correlation time around J the critical temperature cannot be explained by a critical slowing down of the collective proton motion. Concerning the collective proton motion in our sys tem, there has been a growing interest in the study of "S------i------'------1------!------'------1—----'------the dynamics and the thermodynamics of anharmonic HI 21) .111 41' 511 lattices with long-range interactions. For a long-range frequency (MHz) interaction potential in which the interaction falls off ex ponentially with the inter-particle separation, approxi Fig. 1. Larmor frequency dependence of the spin-lattice mate solutions can be obtained, making it possible to relaxation rate (l/Tf) at room temperature. The line shows put forward the role of the long-range interactions in the a fit to an intramolecular dipole-dipole interaction with one occurrence of critical phenomena [24-26]. correlation time.

~~II. EXPERIMENT~~

A powder sample of SCD was commercially avail able. While some workers have measured the temper ature dependence of the *H NMR spin-lattice relax ation time (Ti), we made room-temperature Larmor- frequency-dependent measurements in the barm or fre quency range 9-45 MHz by using the inversion recovery sequence. The spin-spin relaxation time measure ments were made using solid echo decay employing the f - r - £5 pulse sequence at 45 MHz in the temperature range of about 195 K to room temperature,

~~III. RESULTS AND DISCUSSION 21 (isj~~

Fig. 1 shows the room-temperature Larmor-frequency Fig. 2. Solid echo decay pattern at 273 K. The solid line dependence of the spin-lattice relaxation rate (I/7j), shows a single-exponential fit. -652- Journal of the Korean Physical Society, Vol. 47, No. 4, October 2005

ductivity, particularly the DC conductivity, is easily un derstood by using the soliton model. In a polarization n relaxation study, an anomalous decrease in the correla I tion time was observed near Tc, and a three-dimensional a long-range interaction near the critical temperature was c n suggested, which is responsible for the suppression of the critical slowing down [19], In this work, we show that soliton dynamics with a long-range interaction can account for the anomalous behaviors of the spin-spin re laxation time T2E and the electrical conductivity in SCD. It is also to be noted that specific cases, such as the mo tion of protons in an asymmetric double well, diffusion o Conductivity of Bjerrum defects, etc., can be treated in the soliton model. The protonic conductivity, particularly the DC elec Inverse temperature (1000/K) trical conductivity, in SCD is explained by solitons, such as ionic or Bjerrum defects, and the Bjerrum defect-like Fig. 3. Temperature dependencies of the spin-spin relax combined motion of proton can lead to an order-disorder ation time and the DC conductivity. The DC conductivity transition. In the presence of solitons, the critical dy data were taken from Ref. 22. The solid line shows a fit to namics is significantly affected by the central peak in the Eq. (4). low-frequency region of the spectral density, in which case the 1H nuclear spin relaxations will be dominated by the central peak. The soliton dynamical structure fac Since the limit uir q > 1 is well satisfied in our work, it tor in the presence of a long-range interaction has been is obvious that the anomalous critical behaviors can not reported in the form [24] be explained by a critical slowing down, or simply by the n(r)r q pseudospin Ising model, which predicts a decrease in the 5’(g,w)oc (1) 1 + U!STq2 ’ spin-spin relaxation time in the critical region. Thus, a new model that is compatible with the critical increase where = D(r)q2. n(r) and D(r) are the soliton in the spin-spin relaxation time and the conductivities is density and the soliton diffusion coefficient, respectively, required. If the soliton dynamics is responsible for the where the interaction range r varies between 0 and 1. proton conductivities in this system, the soliton model Then, it can be shown that the spectral density is given should explain the anomalies. In other words, if the by soliton model can be employed to explain the peculiar J( 1, It is worthwhile to compare our results of a maximum where the spin-spin relaxation time is expected to be in the spin-spin relaxation time at the critical tempera inversely proportional to the correlation time. ture, corresponding to a decrease in the correlation time, The soliton diffusion coefficient D is proportional to with those of other workers ’ spin-lattice relaxation time the soliton conductivity and is assumed to have a tem measurements. In fact, Kiriyaina et al.’s careful measure perature dependence D 1 apparently holds in 1 he laboratory frame and ficient l)(r) and the soliton conductivity will increase, the rotating frame at this temperature, the decrease in corresponding to a decrease in the collective correlation the spin-lattice relaxation time at the critical tempera time rq. In other words, when a long-range interaction ture unmistakably observed by Kiriyama et al. would occurs near Tc. the increasing soliton diffusivity near the correspond to a decrease in the correlation time, in good critical temperature will result in a decrease in the corre agreement with our spin-spin relaxation measurements. lation time, in contrast to the generally observed critical In SCD, the protonic order-disorder transition is ex slowing down cases. plained by the Bjerrum defect-like combined motion of The similar temperature dependence of the spin-spin translation and rotation of protons [19-23], and the con relaxation time and the conductivity, both of which are Critical Dynamics in SnCE 2H2O - Kyu Won Lee and Cheol Eui Lee -653- expected to be proportional to the inverse of the pro vation energy of E = 0.12 eV. Well below Tc, the spin- tonic collective correlation time, down to well below Tc spin relaxation time and the conductivity begin to show indicates that they are dominated by the same proton discrepancies, which indicates some additional contribu dynamics. If the spin-spin relaxation time is assumed tions to the spin-spin relaxation in the low-temperature to be inversely proportional to the correlation time, Fig. phase, which may arise from the three-dimensional order 1 suggests that the proton conductivity is given solely in the low-temperature phase. by the correlation time. In the critical temperature re In summary, our proton spin-spin relaxation time mea gion, a critical slowing down of the collective correlation surements in SCD showed a critical behavior quite com time is generally observed in systems described by an patible with that of the conductivity. We described the Ising model. In the critical slowing down case where the anomalous critical behaviors, which are not explained inverse of the collective correlation time shows a sharp by the generally known critical slowing down concept, in increase, both the spin-spin relaxation time and the pro light of the soliton model with a long range interaction. ton conductivity will show a sharp decrease, in contrast In other words, the critical behaviors in this work were to the actually observed behaviors. Thus, the critical interpreted as a signature of soliton dynamics. slowing down concept in the pseudo-spin Ising model is obviously not valid for the critical dynamics dominating the spin-spin relaxation and the conductivity in SCD. The fact that the electrical conductivity in SCD shows ACKNOWLEDGMENTS a strikingly similar temperature dependence and critical dynamics to those of T2e provides strong support for our This work was supported by the Korea Science and En interpretation that local and microscopic protonic corre gineering foundation (Proton Accelerator User Program lations associated with the formation of ionic or Bjerrum- No. M202AK010021-04A1101-02110 and No. RO1-20 type defects, which are understood in the framework of 05-000-10798-0) and by the Korea Research Founda the soliton model and which are sensitively reflected in tion (Grant KRF-2004-005-C00060 and Brain Korea 21 the T2e, are responsible for the global, or macroscopic, Project in 2005). The measurements at the Korean Basic protonic transport measured by the electrical conduc Science Institute are acknowledged. tivities. As such, this may be taken as decisive evi dence that soliton dynamics is directly responsible for the macroscopic protonic transport leading to the electrical conductivities. While it has been shown that Ising-type REFERENCES models, representing linear dynamics, cannot explain the critical dynamics in this system, it has been elucidated [1] A. V. Zolotaryuk, A. V. Savin and E. N. Economou, in this work and in the light of previous theoretical and Phys. Rev. Lett. 73, 2871 (1994). experimental achievements that an extended form of the [2] A. Scott, Phys. Rep. 217, 1 (1992); C. H. Lee, K. W. soliton model, representing nonlinear dynamics, compris Lee and C. E. Lee, Curr. Appl. Phys. 3, 477 (2003). ing the various specific cases mentioned above, is capa [3] Ferroelectrics 71 (1987) and 72 (1987) (Special issues ble of accounting for the anomalies associated with the on the KH2PO4 type ferro- and antiferroelectrics) ; C. H. critical dynamics of the proton motions in this hydrogen- Lee, K. M. Lee, C. E. Lee and W. Kang, Curr. Appl. bonded system. Phys. 3, 359 (2003). As discussed above, in the presence of a soliton long- [4] A. Gordon, Phys. Rev. B 52, R6999 (1995); K. W. Lee, range interaction near Tc [24], a critical decrease in the C. H. Lee and C. E. Lee, J. Korean. Phys. Soc. 42, 170 collective correlation time can simultaneously account for (2003). [5] J. Z. Xu, J. Phys.: Condens. Matter 7, 269 (1995). the critical increase in the spin-spin relaxation time and [6] St. Pnevmatikos, A. V. Savin, A. V. Zolotaryuk, Yu. in the protonic conductivity. In that case, the critical be S. Kivshar and M. J. Velgakis, Phys. Rev. B 43, 5518 havior of the collective correlation time can be assumed (1991); C. E. Lee, K. W. Lee, C. H. Lee and D. K. Oh, to scale as J. Korean Phys. Soc. 42, S1182 (2003). [7] G. P. Tsironis and St. Pnevmatikos, Phys. Rev. B 39, T,(XT.|(T - TJ/T/, (3) 7161 (1989): J. W. Jang. S. H. Kim, C. E. Lee, T. J. Lee. C. ,1. Lee, II. S. Kirn, E. It. Kim and S. J. Noh, J. where 0 > 0. Then, in the limit u/r, > 1. assuming an Korean Phys. Soc. 42, S985 (2003). Arrhenius temperature dependence for [8 ] Y. Kaslmnori. T. Kikuchi and K. Nishimoto, J. Ghent. Phys. 77. 1901 (1982). T2 oc r,-' oc - n/r,r". (4) [9] C. M. Van 11a, Phys. Rev. B 14, 244 (1976). [10] J. B. SokololT and A. Widow, Phys. Rev. B 18, 2824 where E is the soliton activation energy. In fact, Fig. (1978). 1 shows that the spin-spin relaxation time and the con [11] J. A. Krumliansl and J. R. Schrieffer, Phys. Rev. B 11, ductivity are well described by this expression near and 3535 (1975); S. H. Kim O. H. Lee, K. W. Lee, C. E. above the critical region. The solid line in Fig. 1 shows Lee. W. Kang and K. S. Hong, Curr. Appl. Phys. 4, 452 the fit with a critical exponent of 0 = 0.25 and an acti (200-1). -654- Journal of the Korean Physical Society, Vol. 47, No. 4, October 2005

[12] A. L. Sukstanskii and K. I. Prinak. Phys. Rev. Lett. 75, tion , edited by F. J. Owens, C. P. Poole, Jr. and H. A. 3029 (1995). Farach (Academic Press, New York, London, 1979). [13] St. Pnevmatikos, Phys. Rev. Lett. 60, 1534 (1988). [20] H. Kiriyama, O. Nakamura and R. Kiriyama, Chem. [14] H. Kiriyama and R. Kirivama, ,1. Phys. Soc. Jpn. 28, Lett. 10, 689 (1976). 114 (1970). [21] K. W. Lee, C. E. Lee and J. K. Kang, J. Korean Phys. [15] T. Matsuo, M. Yatsmi, H. Suga and S. Seki, Solid State Soc. 28, 383 (1995). Common. 13, 1829 (1973); C. E. Lee, N. S. Dalai and R. [22] E. R. Mognaschi, A. Chierico and G. Parravicini, J. Fu, Curr. Appl. Phys. 3, 355 (2003). Chem. Soc., Faraday Trans. I 74, 2333 (1978). [16] R.. Youngblood and K. Kjems, Phys. Rev. B 20, 3792 [23] E. R. Mognaschi and A. Chierico, Solid State Commun. (1979). 33, 103 (1980). [17] R. Kiriyama, H. Kiriyama, K. Kitahama and O. Naka [24] A. M. Dikande and T. C. Kofane, Physica D 83, 450 mura, Chem. Lett. 7, 1105 (1973). (1995). [18] H. Kiriyama, K. Kiriyama, O. Nakamura and R. [25] S. K. Barker and J. A. Krumhansl, Phys. Rev. B 23, Kiriyama, Bull. Chem. Soc. Jpn. 46, 1389 (1973). 2374 (1980). [19] E. R. Mognaschi, A. Rigamonti and L. Menafra, Phys. [26] P. Woafo, T. C. Kofane and A. S. Bokosah, J. Phys.: Rev. B 14, 2005 (1976); F. Borsa and A. Rigamonti ’s Condens. Matter 3, 2279 (1991). Contribution in Magnetic Resonance of Phase Transi Electron Paramagnetic Resonance Study of Pmton-Beam-Irradiated Til l PCX

~~a' »~~

~~I l.rirP I'm .r , ' I:;, i ' !n p:' : ' = « , . = % t . I e ; -~~

~~K e v a;:{ I 'r» C . (r\l'« ; ,f\'. ; !'!*;( ( * . . "I I:~~

~~I. IN 1 R.Or.il.:< I ION~~

IS ! C't' P« " - P': - I I' « ' :.'' ) \ . j I ' * 14 (is. 'A\:,'|, i;.u:'"'y |; l, ", l/i ?" '(uli i: I !i..l I . n ' u ii' mu!.^ .ix ' i< . m • v .*! e j t.;t i • ‘ • ’i i ■ /\‘ i ; « *> -s • ■ In.;, y « > n > . ;' :iv l:i ih''-;i:-('rvyi:u-" lln. nn'i'/U-: .i , .\f j p.« I; ;i= :nis ini Iir .'O.n. . ~ ; ' : " ||\,N ) it I':.) '' ? '!.% 'I;- r ;d.

~~I Ai;in 'i.uixiliiiii I i: - :j, fy, pi..,. -'o-u ; :~~

I- '! 1 • !■ ' I : i A i : ■ 1 . 1 :■ ! 1 . I 1 ; ■!: i v V. I : I. -M ; ":i IM 'II ' ! i U; ■ s i n-i-irrl - ,iis, s, a, 1 - ; '• v ■ s "i. 1 1 - I - > . ' - ■ - - . .(I : . II: ■ ; : ■ ■ I ■ • ■ • ■■■ ‘ ’ i ■ 'ii.-i : : - ■ , ;■ : ■

~~■ vt!,a; a. ■:I.u ii:V ii vs '~~

,\l 1 ui.: - is i'i )" , v. 'i ih a .-a , ui !':■ :i i :■■■•.ii: ; '! i i> Ml Ail'A: A. lA(l s A SI A A : ' ' i i V 1 sA. A i’Ai I I"I I .1 1. ' A I ■ - ' -A i; n c . : M i , ^ :i« fit Uni V, . I'l I H,| I , | • I : AA'aH' S' M'Sl ■- . I ■ ; I , - \ ""xX; r ;:i. I V.ii/iMl'iA'.; ■ I! A AIMS1. V A A. MAASiM! s A AM" 'll s ' : i -:i:= 1.1)' - r rv-

>|'! Ms a i r-i MsM !'i| i is 1 .:: 1 ■■ > i iss ii ■ ; s j,i \ L. s. , si ; ; i 1 1 h - ' I," i-i 'f' ; n (uUi: ' it I n's. -.1! I !, ,;is< ' : ■ ‘: ' !. :' : i si1 ;, . ' i I..1.1/ t '.,1-' l :!! T ’ i I ‘ f ' 11 ’I K ! ii ’ S -s' ! ' s,i, it 'l. si ■ .,r , , r .. s' :.'i«- - : : «- - i «- u ! - : ,

~~• ii M-'l A«M s . s, s I” r' L r i ? i -.1 1 ■ a ‘ ■ a : i " , . ,■ n s~~

~~it? -m.mi- . | > | -n-.-i ii., ; i,- ‘ : - • s~~

! -!' _ /.-.'t: ■ i;, , : *.sii . a a ' ss . ■ : i. -- ■: ! ii' !:•:■; E! A ! i ! ,•.A I 1' A ! ' Ml'( M ■ I' I !" ‘ t : .!. >1,1!.- 'SI . i ' " I. U i: ' I ! I w IAIMMM, Is, AAMAAMi. iSil i isi, 1 i A ; I ■ ; i V. s .AS !' 4 " I' ' \ U ;i! i.-

~~‘ •,' i Si I-’ .a is. ': i S: s I ■ a i'' is ':!' s. -si .. i a . I: a .Isii i ■ :; ;ill j ! '.I S' 'Ll I ' I -'S' : j t.~~

~~! ’S > ■ i ■ : ,|s s ; i 'All'.. ' s : - s 1 s - A . ' - ss . i~~

~~; i 111 li i"' A '.s s. -s. s',- : , s v.-. • «(, i;~~

~~■Si i;: i I sill ; I s - s Is, s ! - s si 11 i • s - ' -. , .: '. I ;~~

~~I IS ISA I- A. si i A.A. . i." . | ' ,1 . . 1~~

~~i'.s A.!!', 'A . i is I.'i A, ;ii'AA, A A A A |.s. : A. ■ ■ . •~~

iillls' si i ASS' A: IAA',:'.' II I A AS : AS ". A S , I ■ . I '. :m AI is 1 'ASH !• : AS i II : S . i ' i si .."-A , , - a i ; a, ' ! i is 'SMI villi il A " ' ,i is, A I . A : . SSI", is ' ' If.

~~Jil-1' I: I siijis! 11 ' . ' A , ,. . Ii. : ,■ .-> A. ' s I : .~~

~~! A' A ' |A| II ASA!!'!’. Ml” A'A '.S' is Ill'S V : 'A..- I''-. "■ i hi~~

~~A " I ' II - SMI I I."I S'. ' ' A I s ■ . ■ • t" l-p-~~

~~■ I i~~

~~i is sis. A - si.i' - . si i . i.~~

~~AS. -S ", S sis A ,1 , ■ • .. I I';:' .S’ i'll "IS . i i - s . i is i A '.'S' - 1 s' - ii . A ' 1 ' " • '~~

~~VI'S ■ I'M Ass ; III Is, As AS .A. : is • ' A A ; A. A.; AHer~~

~~‘A- L A A ' " ' A A: - AS s A.A AS ■'! I ' A S'. ■ ' ! i ii i' « ' I i ; \ ) /: : i , , . - - .~~

~~i' r .!. / :: « ( ' ''~~

~~1: • \~~

~~i .~~

' i ' 'M Cl|i ( > ' i " - :. ' :i! c «I ! I ! »- : i - r1 IT ;'l f r.risu -iv * r) - 'll'!! : I i ! : . I '. i . " " n \ '. .. | :|...... i :>;1:1'. W' ! Xu- i: ' ii !i.! ! . i. .1 x . ' ,; 'i' ; " ! . . . ! . i; i : i I r . un ,i'if' u.r. I. : ' ' ! i' I , » \ | \ i irv'l''/'; "/;T- " ii. ..''r- ; ! : ' 'c i n. 4 !r : I I,''". . ' I :i'- - i, \ i/ : : . lii'ii (;u!\ hvikuyv | i' i\ = i- - ^ itiALTv^opk ivwl. r:n.i:u.v;.' :uu;« I ? 'nru n.x.n scupir t,x»! tu TinMiA" A: K/uni k'HijuTal!): I ;i l.m : v SUi)i)r Iryl I - ;tu :mU t^wrrvr'.-! Lv KSi\ H-io 1 r» ci -;;,i i 1H. I(hS% T.TS A\T) DlFiCl'SSlOX rt^ronm is IXX . n/ud. uri^m in- fr- n 01 u\(hiu(i'l. UI irui T n :i :r =.u * I rlH:, I - II, l in O' ' Uin'U- l,|.\,;; ' , in,-. u '.u.I' ' i y: ' x:-U _ V.u.-^nrirs 'ulj.iw \-r.iy il l .t j!fi" . '! ; .i'l : */ i ' i - i\\ "i , IN' C'|iLir^u n/nuiltn-U l

^uni'd U/ r,'/ rriiiii'd I- uvXuiy :i v/.. "hi- - \ i , ., ' ! ' s - :LllL'' llir «'! !r, ih'/.'/'U ni l\ 1:< !»' l!r |- J. M, KL 1 I" 0 I =, I:U ii ...'! tin I: r I* » I. 'T ' h/i.-llv Tiri'l.fi' (! i.'. ;v \ u :i :iru;i.v.-; ; - . r, ! . ' 11.. !. i -'' i (l:H c 1 ii'i.'i .: O': - /i ,] 'i , m i - i. * ' h ! '! ! : : ' ', u U"l'i !t "i/ ' ,1 i \ i - .1' I'' '.' In Uiis v.\nk wr r.ici'.il c n ; -n ITU' I.. I'lmjlO tl : I. : I I : . ' . !' •..: I I'l.. ! mui'i" .or '!)i' ' in K !;)' v. x «j - n i :i r i 1 "A i: x (in sr.'&iui/ nn^ri'.y ,nx' h\c: " y r* c ' n - : ' ' : ■ ■ !or TUP ix *i;v (i-n r.ni l: ';'": X '' .».i Kv-omiui ''. Krh!. :..i,iv - ; k x s|n -f I run wii!- I%;n:r:;x s;,.u:v- (rpurl.i'i). n . 1^)' . I,''!"'. || i , iufo'O wviiumi'.'J !'\ iiu K:r. !::/j..«n U., ! 'Ur : !.=.. . n O') ! i'l' . - .- '. '! U'u^' uni 'nci.iil - ii m l /*' - | \in . ;;,on ^iv- n . :n :n.i.r. - 41 :a o f\ I '. v - < i .r « r... = . /II : II f n:i. ".:U !-0': i i I (>i

u.i'i n .. !%i; !, ^ . i. - i I- ;. . ! II ! I ' ; " . ' " I ' I' I' n' I' ! - ' : ' Ui I I'!-." 1: 0 'F'' I: - ' - % ! . O' ' - «.n'H - i,l - \ I I n - - »I;n< = ' .r y . i :.m 'U ;rr : : r \* ' '« n 1 . r i|il' ; i\ ;

~~,rni I \ . #U '' '.' ' - .. .1. - x - . ! / \'! f H i., ' I = I ';. . . r/ X'' ;. ' /I'V '' \ . \ ! . . . ,~~

I i ' ,1 ^ ' !. < < I. \\ i ». i I ' - ; ' : ' : ! I ! ,, , _ . <*h. .1 -'!//>. s,,, 12. 'i ' \ I . r.- ». ' ^ :l !: "% \ i hi. 1' ,n \ '»r ,n ; i ir I <»:' I i. (* %:'x ; r. . ( #1 "" " Serf nit.l W.I.rr ( H i I '1 . \ \ ! ..., TX \) . h I-rr. J kr.mxn S'x" 42, ! \!«\}; i- IT# 1"; ' #! ii.J. \f!rr.es AM# II. T' l\ "v. ! '.f.h < m. x \ . I M-^ «_ -x mim XX,-T/l | ~.x: . ' : , M K,.irx*, \ <:, ;7| I \ . Su^yih:. I( ;( 1 / \ ,\ixI .% i ". x."- «' i ^ h/ . / x. . fr,.~ H >. H. )\ i L ^ h * hr "I V \ X;ul '-riH .ikil .1 .« 11,-Y« m-.n, . I< .17 r: . x !.%u .\ Kli \l. I« I x ,i I \. ' I - c";:: : r.-.s-z : - \! T- 1.4. - / 11 ' .1 '. Journal of the Korean Physical Society, Vol. 47. No. 2, August 2005, pp. 294~29G

~~NMR Study of the Anionic Motion in the (TMTSF)2PF6 Organic Conductor~~

~~S. H. Kim, K. W. Lee, In-Mook Kim and Cheol Eui Lee* Department of Physics and Institute for Nano Science, Korea University, Seoul 136-713~~

~~W. Kang Department of Physics, Ewha Womans University, Seoul 120-750~~

~~K. S. H ong Korea Basic Science Institute, Daejeon 305-333~~

~~(Received 15 June 2005)~~

The i9 F nuclear magnetic resonance spin-lattice relaxation in an organic superconductor, (TMTSF^PFe, was investigated in view of the molecular motions of the anionic sublattice struc ture. The temperature dependence of the spin-lattice relaxation was described well by the anionic molecular motions in unequal potential wells with distinct activation energies and correlation times.

~~PACS numbers: 74.70.Kn, 75.30.Fv, 76.60.-k, 78.30.Jw Keywords: (TMTSFRPFg, Organic superconductor. NMR~~

I. INTRODUCTION onset of PFg rotation [10, 11]. Jacobsen et al. re ported that the polarized reflectance in (TMTSF)2PFq for £? || b arose from the infrared absorption band of the The Bechgaard salts, (tetramethyltetraslenafulvalcni PF(7 ion, representing a crossover from ID to 2D metallic um-perphosphorate) (TMTSF)2X, show various ground behavior [12]. Some attempts have been made to explain states, in particular, a superconducting one [1]. At low the anomalous behavior in relation to the anion poten temperatures, some of the Bechgaard salts undergo a tial and its motion [9], In this work, the PFg anion mo metal-to-insulator transition via the Peierls mechanism, and others experience antiferromagnetism, spin density tion, in view of its anion potential, in (TMTSF)2PFg was studied by using 19 F nuclear magnetic resonance (NMR) waves (SDW), or superconducting phase under pressure measurements. [2-6]. Superconductivity has been found in (TMTSF)2X for monovalent inorganic anions X such as PFg, ReCL, and CIO4. The TMTSF molecules as radical cation donors play an important role in electrical conductivity. II. EXPERIMENT (TMTSF)2PFg has a superconducting transition temper ature of 0.9 K at a pressure of 12 kbar [7]. The PFC An electrochemically grown sample of (TMTSF)2PFg anions are placed in centrosymmetrical cavities confined was investigated by using a 188-MHz 19 F NMR spec by the methyl groups of the TMTSF molecules. trometer in the temperature range of 120 to 290 K. In At low temperatures, the mct.al-SDW instabilities are order to investigate the PFg molecular motion, we ob correlated with the anion potential in the nesting model tained the spin-lattice relaxation time (T;) data by using for magnetic field-induced spin density waves [8 j. The the inversion recovery method. The lineshapes were ob electrical behavior of (TMTTF)2X compound, with oc tained by Fourier-transforming I tie free-induction decay tahedral anions MFe (M = P. As, Sb) shows a high- (FID) signals. temperature anomaly near 160 K, with the temperature coefficient of resistance (TC’R) abruptly increasing [9]. Also, in the vicinity of ICO K. (TMTSF)2PFg has been reported to undergo a structural phase transition associ III. RESULTS AND DISCUSSION ated with an interlayer shift [10]. Nevald and cowork ers observed a change in the 19 F NMR linewidth of The anionic PFg sublatlice of the TMTSF molecules (TMTSF)2PF6 at 70 K, which they attributed to the has an octahedral structure consisting of one phosphor atom and six liuorine atoms (Fig. 1). The phos "Corresponding Author : rsccl'&korca.mc.kr phor atom has a six-coordination of six fluorine atoms. -29)- NMR Study of the Anionic Motion in the (TMTSF^PFg Organic Conductor - S. H. Kim et al. -295-

~~Expei iment ...... litling line~~

Fig. 1. PFg anion sublattice in the (TMTSF^PFg struc 1000/T (K ture. Fig. 2. Temperature dependence of the 19 F NMR spin- lattice relaxation rate (T\~1) in (TMTSF)2PF6. The octahedral sublattices have slightly different bond lengths and angles between the phosphor and fluorine atoms, giving rise to a reorientation of inequivalent lat and tice sites in unequal potential wells [13,14], The asym T2 = TQ2 exp^/AT). (3) metry in the anion potential is related to the intramolec ular interaction within the PFg octahedral potential sur 7, M2, Tg, and E are the protonic gyromagnetic ratio, face [15], and the anisotropy in the NMR measurements the second moment, the motional correlation time at infi results from intramolecular interactions giving ruse to nite temperature, and the activation energy, respectively, hindered rotational motions [16]. The energies of the and R is the universal gas constant. A, and A2 refer equilibrium positions of the molecule may be different to the energy differences between equilibrium and the when various positions are inequivalent with respect to mcta.sta.ble states. For At — A2 and E\ =£2, these ex the lattice symmetry [17]. The NMR spin-lattice relax pressions reduce to the usual spin-lattice relaxation rate ation time is given by the modulation of the dipole-dipole for a double-well potential with equal minima. The cor interactions between the spins belonging to a molecule. relation times are for the re-orientational motion of the In order to investigate the nature of the PFg motions PFg octahedra. The rapid rotational motion of the octa and intramolecular interactions, we calculated the flu hedral anions is anisotropic; i.e., the distorted octahedral orine second moments theoretically by using the struc shape of the PFg anion includes a deviation from spher tural parameters [13, 18]. The fluorine homo-nuclear ical symmetry and an anisotropy in the libration tensor dipolar interaction makes a much greater contribution [10]. In effect, the anisotropy of the anions can be at to the spin-lattice relaxation than the F-P hetero-nuclear tributed to the inequivalent lattice sites of the ellipsoidal interaction as the fluorine resonance frequency is far from rotation. The parameters obtained from the fit of the re the phosphor one. laxation data are shown in Table 1. The two correlation The temperature dependence of the spin-lattice relax times ti and r2 are quite distinct so that the unequal ation rate cannot be fitted by a simple the Bloembergen- potential has two metastable potential wells with A, = Purcell-Pound (BPP) description. The spin-lattice re 22 ± 4 meV and A2 = 34 ± 2 meV, respectively. laxation shows an asymmetric temperature dependence The second moment M2 obtained from the fitting of with respect to the T\ minimum, as shown in Fig. 2, in the spin-lattice relaxation rate above is 2.49 ± 0.29 G2, dicating that the relaxation may arise from fluctuations which is in very good agreement with our experimental caused by an unequal potential-well surface. If we use the unequal potential model with powder averaging, the spin-lattice relaxation rate can be expressed by [14] Table 1. The parameters obtained from a fit. of the spin- lattice relaxation data for the octahedral PFg anion motion. IT' = ^fM2[3exp(-A,/RT)(^^ (1) Parameter/unit Value

~~M2/G2 2.49 ± 0.29 ' ' + cxp(-Az/RT)' E| /meV 334 ± 30 1 + 4 w2t2 1 + CV2T2 Til / S 7.70 < 10‘20 ± 1.56 x hr21 ___ A] /riieY 22 - 4 Ee /meV 121 + 3 where rm/s 2.41 x 10-'3 ± 5.66 x ttr"~~

~~A2/ meV 31 ± 2 ti = T0i exp(Ei/ET) 12) -296- Journal of the Korean Physical Society, Vol. 47, No. 2, August 2005~~

~~REFERENCES 2.7- "T in (TMTSFJ^RF~~

<5. 2.4- [1] V. J. Emery and R. Bruinsma, Phys. Rev. Lett. 48, 1039 ■ ■ £■?> #■■■■ ■ (1982); K. W. Lee, C. H. Lee and C. E. Lee, J. Korean I ■ Phys. Soc. 42, 170 (2003). O [2] D. Jerome, A. Mazaud, M. Ribault and K. Bechgaarrd, J. Physique Lett. 41, L95 (1980); C. E. Lee, K. W. Lee, XJ 2.1 ' s C. H. Lee and D. K. Oh, J. Korean Phys. Soc. 42, SI 182 (2003). [3] K. Andres, F. Wudl, D. B. McWhan, G. A. Thomas, D. Nalewafek and A. L. Stevens, Phys. Rev. Lett. 45, 1449 1.8 - 120 140 160 180 200 220 240 260 (1980); J. W. Jang, S. H. Kim, C. E. Lee, T. J. Lee, C. J. Lee, H. S. Kim, E. H. Kim and S. J. Noh, J. Korean Temperature (K) Phys. Soc. 42, S985 (2003). Fig. 3. Second moment of the 19 F NMR measured as a [4] R. L. Greene and E. M. Engler, Phys. Rev. Lett. 45, 1587 function of temperature in (TMTSF^PFg. (1980); C. E. Lee, N. S. Dalai and R. Fu, Curr. Appl. Phys. 3, 355 (2003). [5] M. Ribault, J. P. Pouget., D. Jerome and K. Bechgaarrd, C. R. Acad. Sci. Ser. B 291, 145 (1980); C. H. Lee, K. W. data in Fig. 3. The second moment of the NMR. line was Lee and C. E. Lee, Curr. Appl. Phys. 3, 477 (2003). calculated by using [6] D. Jerome, Science 252, 1509 (1991); C. H. Lee, K. M. Lee, C. E. Lee and W. Kang, Curr. Appl. Phys. 3, 359 M2= / (w-wjVMdw/ / (4) (2003). [7] K. Bechgaard, C. S. Jacobson, K. Mortensen, J. H. Peder son and N. Thorup, Solid State Commun. 33, 1119 (1980); where w is a frequency and f(uj) is the NMR lineshape S. H. Kim, C. H. Lee, K. W. Lee, C. E. Lee, W. Kang and with a maximum at a frequency cv„ [19], In fact, our K. S. Hong, Curr. Appl. Phys. 4, 452 (2004). experimental M2Xp value remains almost unchanged at [8 ] K. Sengupta and N. Dupuis, Phys. Rev. B 65, 35108 around 2.4 G2 with little temperature dependence, indi (2001). cating that the motional narrowing for the PFg rotation [9] C. Coulon, S. S. P. Parkin and R. Laversanne, Phys. Rev. is hindered, as expected, for unequal potential wells [15- B 31, 3583 (1985). 17]- [10] V. J. Mcbrierty, D. C. Douglass and F. Wudl, Solid State In summary, our 19 F NMR spin-lattice relaxation data Commun. 43, 679 (1982). for the organic superconductor (TMTSF^PFg were an [11] V. J. Mcbrierty, D. C. Douglass, F. Wudl and E. Aharon- alyzed in terms of unequal potential wells for the PFg - Shalom, Phys. Rev. B 26, 4805 (1982). sublattice. As a result, correlation times for the re- [12] C. S. Jacobsen, D. B. Tanner and K. Bechgaarrd, Phys. Rev. Lett. 46. 1142 (1981). orientational motions and distinct potential well depths [13] N. Thorup, G. Rindorf, H. Soling and K. Bechgaard, were obtained by fitting the data. Acta Cryst. B 37, 1236 (1981). [14] M. Polak and D. C. Ailion, J. Chem. Phys. 67, 3029 (1977). ACKNOWLEDGMENTS [15] N. E. Ainbinder, G. E. Kibrik, A. N. Osipenko and G. B. Soifer, Phys. Stab Sol. 39, 780 (1966). [16] R. Bersohn and II. S. Gutowsky, J. Chem. Phys. 22, 651 This work was supported by the Korea Science and En (1954). gineering Foundation (Proton Accelerator User Program [17] D. C. Look and I. J. Lowe, J. Chem. Phys. 14, 3437 No. M202AK010021-04A1101-02110 and RO1-2005-000- (1966). 10798-0) and by the Korea Research Foundation (Grant [18] S. H. Kim. O. H. Lee, K. W. Lee, C. E. Lee, W. Kang No. KRF-2004-005-C00060 and Brain Korea 21 Project and K. S. Hong, Curr. Appl. Phys. 4, 452 (2004). in 2005). The measurements at the Korean Basic Science- [19] A. Abragam. Principles of Nuclear Magnetic Resonance Institute (KBSI) are acknowledged. (Oxford. New York, 1983). Journal of the Korean Physical Society, Vol. 47, No. 1, July 2005, pp. 130~132

~~Photoluminescence Study of the Proton-Irradiated MEH-PPV Conjugated Polymer~~

~~Kyu Won Lee, Kyu Hyun Mo, Jae Won Jang and Cheol Eui Lee* Department of Physics and Institute for Nano Science, Korea University, Seoul 136-713~~

~~(Received 26 April 2005, in final form 6 March 2005)~~

We studied the proton-irradiation effect on the photoluminescence in the MEH-PPV conjugated polymer. The integrated PL intensity was severely weakened by low-energy irradiation, but was affected only weakly by high-energy irradiation. The integrated PL intensity decreased with increas ing irradiation dosage and the decrease was steeper for lower-energy irradiation. The reduction in the integrated PL intensity can be attributed to proton-irradiation-induced hydrogen loss from the MEH-PPV polymer.

~~PACS numbers: 78.70.-g, 78.55.Kz, 78.66.Qn Keywords: Photoluminescence, Proton irradiation, MEH-PPV polymer~~

There has been great interest in light-emitting diodes organic materials will mainly be caused by ionization of (LEDs) based on conjugated polymers because of their the target atoms. potential applicability to large-area flat panel dis The MEH-PPV conjugated polymers used in this plays operating at a relatively low voltage, along with study were purchased from ALDRICH. A solution of their promise of low cost and easy fabrication [1], MEH-PPV dissolved in chlorobenzene was spin-coated The poly(phenylenevinylene) (PPV) conjugated poly onto 1 cm x 1 cm indium tin oxide (ITO) glass to a mer, which is one of the most studied polymers for thickness of about 200 nm. To avoid oxygen contami PLEDs (polymer LEDs) due to its excellent luminescent nation, we injected argon gas into the spin coater. The and mechanical properties [2,3], is known to be a hole- MEH-PPV films were irradiated with proton beams with transport material because its hole mobility is one to a spot radius of 0.5 cm at room temperature. The pro three orders of magnitude higher than its electron mo ton energy was varied from 0.5 MeV to 3 MeV with bility [4,5]. PPV and its derivatives have been inten doses from 1013 to 1015 ions/cm 2. The photolumines sively studied [6-8 ], and the use of conjugated polymers cence (PL) spectra of the MEH-PPV films were obtained as possible radiation detection materials has also been two days after the irradiation. studied [9-12]. A recent study of the gamma radiation Figure 1 (a) shows the PL spectra of the pristine MEH- effect on the optical properties of poly[2-methoxy-5-(2- PPV films and of the films irradiated with a dose of 1013 ethylhexyloxy)-p-phenylenevinylene] (MEH-PPV), one ions/cm 2. The line shapes of the PL spectra were typical of the PPV derivatives, showed that a MEH-PPV so of MEH-PPV and were not changed after the irradiation. lution is very sensitive to low-dose radiation [12]. Interestingly, the PL intensity reduction increased with In this work, we have studied the proton-irradiation decreasing irradiation energy. The integrated intensity / effect on the luminescence properties of MEH-PPV thin of the PL spectrum normalized to the integrated inten films. Ion-irradiation is well known to induce hydrogen sity I0 of the pristine MEH-PPV is shown in Figure 1 loss from organic materials, and some theoretical models (b). have been developed for the hydrogen loss [13,14], An Figure 2 (a) shows the PL spectra of the pristine MEH- energetic ion incident on a solid imparts its energy to the PPV films and of the films irradiated with 3-McV pro target nuclei in elastic collisions or spends its energy in tons at various dosages. The line shapes of the PL spec the ionization of target atoms [13-15], For high-energy tra were typical of MEH-PPV and were not, changed by ions, where the ions traverse the target, ionization is the irradiation up to a dose of It)14 ions/cm 2, above which dominant process of energy loss for the irradiating ions, they were broadened with a blue-shift. With 1-MeV pro leading to hydrogen loss from the organic materials. In ton irradiation, a similar tendency for the PL spectra was the case of organic thin films, relatively low-energy ions observed, but the PL intensity of the MEH-PPV films ir can traverse the films, and the hydrogen loss from the radiated with high dosages was too small to be observed. Figure 2 (b) shows the integrated intensity, ///13, of the * E-mail: [email protected] : Fax: +82-2-927-3292 PL spectra as a. function of the irradiation dosage, where 113 is the integrated intensity corresponding to a dose of -130- Photoluminescenc.e Study of the Proton-Irradiated- - - - Kyu Won Lee et al. -131-

~~la) / V~~

~~MEH-PPV y/r .1 MeV ...... 2 MeV t MeV~~

~~07 MeV~~

~~0,5 MeV~~

~~500 550 600 650 700 750 5 30 550 600 6 50 700 7=~~

~~Wave length (run) Wavelength~~

~~1.0 * tb) ■ 1 MeV 1 0- “ c J MeV 0.B- ■~~

~~n . ■ 0.6- 08- n _~~

~~0.4- ■ - ■~~

06- 0.2- ■ 101' ion s/cm *

~~0,0- 04- 0.0 0.5 1.0 1.5 2.0 2.5 3.0 1013 \ o' 4 Proton energy (MeV) Dosage~~

Fig. 1. (a) PL spectra after irradiation with a dose of Fig. 2. (a) PL spectra after 3-MeV proton beam irradi 101013 *ions/c.m * 2 at various irradiation energies, (b) Normalized ation at various proton dosages, (b) Normalized integrated integrated intensity I/Iq of the PL spectra as a function of intensity I/Ii3 of the PL spectra as a function of the irra the irradiation energy, where To is the integrated intensity of diation dosage, where 113 is the integrated intensity corre the pristine MEH-PPV. sponding to a dose of 1013 ions/cm 2. The solid and the open symbols correspond to 1-MeV and 3-MeV proton irradiation, respectively. 1013 ions/cm 2. The integrated intensity decreased with increasing dosage much faster for the lower-energy irra tra, indicative of a shortening of the polymer conjugation diation. length, which may arise from a large hydrogen loss in The loss of hydrogen induced by ion irradiation was MEH-PPV. observed to increase with increasing irradiation dosage In summary, we studied the proton-irradiation ef and was much faster for lower-energy irradiation [13-15], fect on the photoluminescence in the MEH-PPV light- According to a model for ion-irradiation-induced hydro emitting polymer. The photoluminescence was sensitive gen loss, secondary electrons are emitted along the ion track, and the total number of secondary electrons in to low-energy and low-dose proton irradiation, suggest ing a possible application for proton dosimetry. creases with decreasing irradiation energy; their kinetic energies also decrease [14]. The total number of sec ondary electrons appears to play an important role in the hydrogen loss. The hydrogen loss induced by ion irradi ACKNOWLEDGMENTS ation was consistent with our results for the PL intensi ties, which decreased with increasing irradiation dosage, the decrease being much faster for lower-energy irradia This work was supported by the Korea Science and En tion. Therefore, the proton-irradiation-induced diminu gineering Foundation (RO1-2005-000-10798-0 and Pro tion of the PL intensity can be attributed to hydrogen ton Accelerator User Program No. Ml02KS010001- loss induced by proton irradiation. At low dosages, the 02K1901-01814) and by the Korea Research Founda proton irradiation changed the integrated PL intensities, tion (Grant No. KRF-005-C000G0 and Brain Korea 21 but did not change the. line shapes of the PL spectra, Project in 2005). We thank Dr. H-.J. Woo at the KIGAM which indicates that the energy band of MEH-PPV was for the proton-beam irradiation. The measurements at not significantly affected. At high dosages, however, the the Korean Basic Science Institute (KBSI) are acknowl proton irradiation gave rise to blue-shifts of the PL spec- edged. -132- Journal of the Korean Physical Society, Vol. 47, No. 1, July 2005

REFERENCES and J. Kim, J. Korean Phvs. Soc. 45, 505 (2004). |8 ] S. H. Park, J. Y. Kim, H. Kim and K. Lee, J. Korean Phys. Soc. 46, 1049 (2004). [1] J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. [9] K. Yoshino, S. Hayashi and Y. Inushi, Jpn. J. Appl. Phys., Marks, K. Mackay, R. H. Friend, P. L. Burns and A. B. Part 1 21, L569 (1982). Holmes, Nature 347, 539 (1990). [10] H. Kudoh, T. Sasuga, T. Seguchi and Y. Katsumura, [2] R. H. Friend, R. W. Gymer, A. B. Holmes, J. H. Bur Polymer 37, 2903 (1996). roughes, R. N. Marks, C. Taliani, D. D. C. Bradley, D. [11] S. C. Graham, R. H. Friend, S. Fung and S. C. Moratti, A. Dos Santos, J. L. Bredas, M. Logdlund and W. R. Synth. Met. 68 , 903 (1997). Salaneck, Nature 397, 121 (1999). [12] E. A. B. Silva, J. F. Borin, P. Nicolucci, C. F. O. Graff, T. [3] P. K. H. Ho, J. S. Kim, J. H. Burroughes, H. Becker, S. F. G. Netto and R. F. Bianchi, App. Phys. Lett. 86 , 131902 Y. Li, T. M. Brown, F. Cacialli and R. H. Friend, Nature (2005). 404, 481 (2000). [13] M. E. Adel, O. Amir, R. Kalish and L. C. Feldman, J. [4] H. C. F. Martens, J. N. Huiberts and P. W. M. Blom, App. Phys. 66, 3248 (1989). Appl. Phys. Lett. 77, 1852 (2000). [14] M. P. de Jong, A. J. H. Maas, L. J. van Ijzendoorn, S. [5] B. K. Crone, I. H. Campbell, P. S. Davids and D. L. Smith, S. Klein and M. J. A. de Voigt, J. App. Phys. 82, 1058 Appl. Phys. Lett. 73, 3162 (1998). (1997). [6] S. Cho, E. Kim, C. Kim, J. Noh, H. Choi, D. Park, T. [15] J. Gonzalez-Hernandez, R. Asomoza, A. Reyse-Mena, J. Kwon, D. Yoo, I. Kim, S. Kim and D. Chung, J. Korean Rickards C., S. S. Chao and D. Pawlik, J. Vac. Sci. Tech- Phys. Soc. 45, 623 (2004). nol. A 6, 1798 (1988). [7] I. Chang, S. Kim, J. Song, S. Choi, M. J. Cho, D. H. Choi Journal of the Korean Physical Society, Vol. 46, No. 1, January 2005, pp. 242~244

~~Impedance Spectroscopy of Ionic Conduction in Hydrogen-Bonded TIH2PO4~~

~~Se Hun Kim, I~~

~~(Received 5 August 2004)~~

TIH2PO4, a hydrogen-bonded ferroelectric, was investigated by AC impedance measurements at various temperatures. Both the real and the imaginary impedance values showed anomalous behaviors at the ferroelastic-paraelastic transition. On the complex impedance plane, the Cole- Cole plots evolved into two regions with increasing temperature. The low-frequency region shows Warburg impedance (n ~ 0.5), as in the case of surface conduction in solid electrolytes. The high- frequency conduction mechanism appears to be proton ionic transport in the bulk. The proton conduction for TIH2PO4 consists of two-phase elements, corresponding to a glass-like surface region and a normal bulk region.

~~PACS numbers: 64.70.Kb, 76.60.-k, 77.90. +k Keywords: TIH2PO4, Hydrogen bond, Proton conductor~~

I. INTRODUCTION the c axis. Protons of these bonds are at special positions at a center of inversion and undergo an order-disorder phase transition through the phase transition temper TIH2PO4 (TDP) is closely related to the KH2PO4 ature Tc. The longest bond, 0.25 nm, is asymmetric (KDP)-type crystals, which are interesting hydrogen- along the b axis and the protons are at a general posi bonded materials undergoing structural phase transi tion both above and below Tc [12-16]. The very short tions accompanied by ferroelectricity or antiferroelectric- hydrogen bond lengths and the very heavy mass of the ity [1]. In these crystals, it is known that protons in T1+ ion are peculiar to TDP and play an important role double well potentials on the hydrogen bonds undergo in the phase transitions. In a TDP sample, two types of a phase transition accompanied by displacements in the domains are known to coexist in the ferroelastic phase. heavy atom (K,P,0) structure. One outstanding phe One is the (001) /(100) ferroelastic domain capable of nomenon is the proton-deuterou “isotope effect” that switching by external stress, disappearing in the parae raises the transition temperature by about 100 K and lastic phase. The other is the (201) plane which is not a decreases the pressure dependence of this transition tem ferroelastic domain: it persists in the paraelastic phase perature. TDP undergoes two major phase transitions and no switching occurs [17-20], This intriguing domain and three more complex phase transitions [2-9]. The structure may affect the proton transport. In this work, room temperature phase (phase II) is known to be para- complex impedance measurements were used for the pro electric and ferroelastic, while the low-temperature phase ton dynamics. (phase III) is believed to be antiferroelectric. The high- temperature phase (phase I) is known to be paraelectric and paraelastic. TDP has a mouoclinic crystal structure in phases II and III, and an orthorhombic structure in II. EXPERIMENTS phase I [2-4,9,10], In TDP the low temperature (1I-III) antiferroelectric phase transition occurs at Tc = 230 K, A sample of TDP powder pellet was used for the mea and the high temperature (I-II) ferroelastic phase tran surements in the temperature range of 160 to 400 K. The sition at T'r = 357 K. complex impedance was measured by using a Quadtech TDP has three different crystallographic hydrogen 7600 impedance analyzer in the frequency range between bonds as determined by X-ray and neutron scattering, 10 Hz and 2 MHz. and the crystal structure of TDP illustrating the three inequivalent H sites can be found in the literature ! 11]. The two shorter bonds. 0.213 mil and 0.215 11m. respec tively, are centrosymmetric and form zigzag chains along III. RESULTS AND DISCUSSION

* E-mail: rsccl'S'korea. ac. kr; Figure 1 shows the real (Z') and the imaginary (l") Tel: 4-82-2-3290-3098; Fax: 4-82-2-927-3292 parts of complex impedance as a function of frequency -242- Impedance Spectroscopy of Ionic Conduction in Hydrogen-Bonded TIH2PO4 - Se Hun Kim et al. -243-

~~T emperature = 392K~~

~~~^\f~ W-~~

~~c N1~~

~~n = 0 576~~

~~6 9 12 15 Z- (kfi) Fig. 2. Cole-Cole plot of complex impedance. The dotted line is a fit to an equivalent circuit.~~

0.85 ev Frequency (Hz) Fig. 1. The real (a) and imaginary (b) parts of complex impedance as a function of frequency. at several temperatures above 300 K. Z' in the low fre 0,78 eV I quency region decreases with increasing temperature and in the high frequency region decreases with increasing frequency, approximately as yiv dependence of the ad mittance Y = j? [21]. The peaks of the relaxation dis 1000ZT (K) persion of Z" move toward higher frequencies with in creasing temperature. In the low frequency region, a Fig. 3. Activation energy of TIH2PO4 new dispersion appears above 317 K. The two disper sions suggest that two different conduction mechanisms evolve with increasing temperature and that the diffusion limited ionic motion, in which the concentration of the process is a function of frequency and temperature. mobile carriers are distinct as in a double layer structure. Figure 2 shows the Cole-Cole plot above T'c. The The correlation time is the product of resistance and impedance spectra show a semicircle in the high fre capacitance which are obtained by the fits in Figure 2 quency region and show a Warburg impedance in the low [22]. The correlation time follows the Arrhenius law, frequency region. The Warburg impedance corresponds r = r0exp[- Ea/ksT], where Ea is the activation energy, to the diffusion process in an infinite transmission line and r as a function of reciprocal temperature is shown in composed of resistors and capacitors and is described as figure 3. Below and above T'c, En was found to be 0.85 eV and 0.78 eV, respectively. Above 7%, the decrease of Ea 1 may be attributed to the proton disordering [23]. The Z* = (1) GW Warburg impedance in the low-frequency region seems to be related to the boundary between the paraelastic with an index of n = 0.5. Below 308 K, the Cole- region and the (201) domain, which persists even in the Cole plot was well fitted to a simple equivalent circuit paraelastic phase. model of parallel combination of resistive and capacitive elements. Between 308 K and I]', some more compli cated equivalent circuit models were required. O11 the IV. CONCLUSIONS other hand, above 7’r', only an equivalent, circuit including the Warburg impedance can fit the experimental data. An impedance spectroscopy study of TIH2PO4 shows The Cole-Cole plot shows two separate regions: one is a a new diffusion process of proton ionic conduction above bulk-limited proton transport and the other is a surfa.ce- 317 K in the low frequency region. A Warburg diffusion -244- Journal of the Korean Physical Society, Vol. 46, No. 1, January 2005

(n ~ 0.5) was manifest above the ferroelastic phase tran [7] K. S. Lee and K. L. Kim, J. Phys. Soc. Jpn. 60, 3207 sition temperature T'c. Above T'c, the proton ionic con (1991). duction may be frustrated near the boundary of the (201) [8 ] J. Seliger, V. Zagar and R. Blinc, Phys. Rev. B 48, 52 domain, which persists even in the paraelastic phase. (1993). [9] R. Blinc, M. Rozmarin, F. Milia and M. Melisaropoulou, Solid State Common. 27, 999 (1978). ACKNOWLEDGMENTS [10] R. H. Nelms and R. N. P. Choudhary, Solid State Com mon. 38, 321 (1981). [11] Z. Ouafik, D. Brule, F. Remain, B. Pasquier and N. Le This work was supported by KISTEP (National Re Calve, J. Raman Spectrosc. 27, 1 (1996). search Laboratory and Proton Accelerator User Program [12] J. Seliger, V. Zagar, R. Blinc and V. IL Schmidt, J. No. M102KS010001-02K1901-01814) and by the Korea Chem. Phys. 88 , 3260 (1988). Research Foundation (KRF-2004-005-C00060/D00057 [13] C. E. Lee, C. H. Lee, J. H. Kirn and K. S. Lee, Phvs. and Brain Korea 21 Project in 2004). Measurements Rev. Lett. 75, 3309 (1995). at the Korean Basic Science Institute (KBSI) are ac [14] C. H. Lee, K. W. Lee, C. E. Lee and K. S. Lee, Phys. knowledged. C. E. L. gratefully acknowledges the Korea Rev. B 55, 11088 (1997). [15] C. E. Lee, N. S. Dalai and R. Fu, Curr. Appl. Phys. 3, University Research Fund. 405 (2003). [16] K. W. Lee, E. M. Lee, C. E. Lee, C. H. Lee and K. S. Lee, J. Korean Phys. Soc. 39, 394 (2001). REFERENCES [17] H. Yoshida, M. Endo, T. Kaneko, T. Osaka and Y. Makita, J. Phys. Soc. Jpn. 53, 910 (1984). [1] R. Blinc and B. Zeks, Ferroelectrics 72, 193 (1987); J. [18] K. Irokawa, M. Komukae, T. Osaka and Y. Makita, J. H. Jeong, Y. H. Kim, J. S. Kim, I. W. Kim, B. M. Jin Phys. Soc. Jpn. 63, 1162 (1994). and S. H. Bae, J. Korean Phys. Soc. 44, 1521 (2004). [19] E. M. Lee, K. W. Lee, C. H. Lee, C. E. Lee and K. S. [2] K. S. Lee and D. H. Ha, Phys. Rev. B 48, 73 (1993); Y. Lee, J. Korean Phys. Soc. 35, S1434 (1999). Seo and S. Park, J. Korean Phys. Soc. 45, 769 (2004). [20] K. S. Lee, J. H. Park, K. B. Kim, J. B. Kim and, J. N. [3] K. Irokawa, M. Komukae, T. Osaka and Y Makita, J. Kim, J. Phys. Soc. Jpn. 66, 1268 (1997). Phys. Soc. Jpn. 63, 1162 (1994); J. W. Hyun, J. Korean [21] J. R. Macdonald, Impedance spectroscopy, (Wiley Inter Phys. Soc. 44, 381 (2004). science, New York, 1987). [4] A. Matsuo, K. Irokawa, M. Komukae, T. Osaka and Y. [22] S. H. Kim, J. W. Jang, K. W. Lee, C. E. Lee and S. W. Makita, J. Phys. Soc. Jpn. 63, 1626 (1994); C. H. Lee, Kim, Solid State Common. 128, 143 (2003). A. Han, J. Kim, J. Korean Phys. Soc. 45, 1123 (2004). [23] W. T. Lee, E. K. H. Salje and LL Bismayer, J. Phys.: [5] R. J. Nelms, Solid State Common . 39, 741 (1981). Condens. Matter 15, 1353 (2003). [6] K. S. Lee, S. M. Ju, K. L. Kim, S. K. Lee, J. H. Kim, J. B. Kim, B. C. Choi and J. N. Kim, Ferroelectrics 137, 123 (1992). Journal of the Korean Physical Society, Vol. 46, No. 1, January 2005, pp. 245~249

~~NMR Study of Distinct Phase Transitional Behaviors in (CnH2n+iNH3)2SnCl6~~

~~Kyu Won Lee and Cheol Eui Lee* Department of Physics and Institute for Nano Science, Korea University, Seoul 136-723~~

~~J. Y. Choi Department of Computer Science, Korea University, Seoul 136-723~~

~~Joon Kim School of Life Sciences and Biotechnology, Korea University, Seoul 136-723~~

~~(Received 5 August 2004)~~

Phase transitions in bis-(n-CnH2n+iNH3)2SnCl6, where the hydrocarbon part is analogous to lipid membrane, were investigated by means of 200-MHz 'H nuclear magnetic resonance. As a result, critical fluctuations and molecular dynamics associated with the phase transitions, an order- disorder and a conformational phase transition, were distinguished in a wide temperature range. The critical dynamics, observed in the long-chain compounds but not in the short-chain compounds by laboratory frame spin-lattice relaxation measurements, is discussed in view of the chain length dependence of molecular dynamics.

~~PACS numbers: 64.70.kb, 76.60.-k Keywords: Nuclear magnetic resonance, Phase transition, Model membrane~~

I. INTRODUCTION low the order-disorder transition temperature, the alkyl chains are rigid and adopt an all-trans conformation. In the intermediate (IT) phase between the order-disorder The bis-n-alkylarnmonium hexachlorost'annates (n- and the conformational transition temperatures, the uni CnH2n+iNH3)2SnCl6 (CnSn for short) are. layered com axial rotation of the rigid alkyl chain about the chain pounds of alternating organic and inorganic layers. The axis as well as the chain-end trans-flonc/re isomerization SnCle2- octahedra do not form a 2D macroanion but ex take place. In the high temperature (HT) phase above ist separately [1—3]. The NH3 group of alkylammonium the conformational transition temperature, the average ion links the three closest octahedra through equivalent chain axis is destroyed and the chains undergo liquid hydrogen bonds of the N-H- • Cl type, forming an inor like isotropic motions. The three phases and the two ganic layer. The distance between the ammonium groups phase transitions are quite analogous to those of a low- or between the tin atoms in CnSn is great enough (7.3 ~ hydration lipid membrane [10-14], 7.5 A, depending on the chain length) to accommodate In contrast to the case of bis-n-alkylammonium tetra- a monolayer of interdigitated alkyl chains. The alky chlorometallates (n-CnH2n+iNH3)2MCl4 with M = Cd, lammonium groups are statically disordered around the Cu, Mn and Zn (CnM) [15-19], little information is avail three fold-axes at (1/3, 2/3, z) and (2/3, 1/3, 5), with able on the phase transitions in CnSn [4-7,20-22], The the alkyl chains alternately pointing upwards and down first systematic studies on the layer-structure intercala wards [3], tion compounds including ClOCd were made by Blinc In our previous 'H NMR studies of the CnSn systems, and coworkers by using magnetic resonance. Measure two typical successive phase transitions were observed [ 1- ments of the order parameters by 3o Cl and 14N NQR and 9]: a) order-disorder transition of the rigid alkyl chains *H and 13C NMR made possible a consistent understand accompanied by a uniaxial reorientation about their long ing of phase transitions and dynamics of the NII3 and the axes along with the flipping of the polar groups, such hydrocarbons by using a Landau theory similar to that in as NH3, between the potential wells, and b) conforma the liquid crystal [15,16], In a 2H NMR study of C4Cd tional transition leading to a partial melting of the alkyl [19], the evolution of orientational and conformational chain part. In the low temperature (L'T) phase lie- orders of the hydrocarbon chain was shown to be similar to those in lipid membranes, although the order parame * E-mail: rscel@korea. ar. kr; ter tensor was not uniaxial and the value in the high tem Tel: 4-82-2-3290-3098: Fax: 4-82-2-027-3292 perature phase was two times greater than that in the La -246- Journal of the Korean Physical Society, Vol. 46, No. 1, January 2005

phase. Critical fluctuations have been observed in a few (free induction decay) signals. studies near the order-disorder transition temperature of a lipid layer intercalated between the inorganic layers [4,5,23]. In hydrated lipid membranes, (pseudo)critical swelling has been intensively studied near the main tran III. RESULTS AND DISCUSSION sition temperature, mostly near the La <-> Pp transition temperature [24-26], and much less frequently, near the Figure 1 shows the evolution of the NMR line shape in La <-> Lp transition temperature [27]. C16Sn with increasing temperature, which is typical in While ClOSn and CISSn show the same transition se CnSn for all n [6,7]. At low temperatures it consists of a quence, the transitional behaviors reflected in 'H NMR relatively narrow component and a broad one. The broad spin-lattice relaxation time at 200 MHz are considerably line can be mostly attributed to the rigid part of the long different. In ClOSn as well as in C12Sn, the critical alkylammonium chains, corresponding to the methylene fluctuations were not reflected in the spin-lattice relax groups, as the chains are supposed to be well ordered ation [6,7], but were observed in C18Sn near the order- at low temperatures. The narrow line is attributed to disorder transition temperature [4,5]. In C18Sn, the spin- the mobile part of the alkylammonium chains, presum lattice relaxation pattern near the order-disorder tran ably corresponding to the methyl and ammonium groups, sition temperature depends on the Larmor frequency. whose fast reorientation averages out the dipolar field, re At 200 MHz, the spin-lattice relaxation pattern is a sulting in the sharp line. The low-temperature line shape double-exponential form, for which the shorter time con was well fitted into a dipolar powder spectrum (Pake stant only reflects a critical fluctuation near the order- doublet) and a Lorentzian line [23]. The Lorentzian disorder transition temperature [4]. The spin-lattice re line corresponds to the reference frequency and the dipo laxation is a single-exponential form at 45 MHz, where lar powder spectrum shows two symmetrically separated the time constant also shows a divergent anomaly near shoulders about the central Lorentzian line. The separa the order-disorder transition temperature [5]. While the tion, which is ascribed to a dipolar splitting of the rigid spin-lattice relaxation time of C12Sn at 200 MHz does hydrocarbon chain, gradually decreased with increasing not reflect a critical fluctuation [7], the rotating frame temperature. spin-lattice relaxation time at 55 kHz shows a divergent Figure 2 shows the FWIIM (full width at half maxi anomaly near the order-disorder transition temperature mum) linewidth of C16Sn as measured directly from Fig [28]. Thus, it appears that the transitional behaviors re ure 1 as a function of temperature, in which two phase flected in *H NMR spin-lattice relaxation time are asso transitions are apparent at about Tcl = 329 K and Tc2 = ciated with the observing frequency (Larmor frequency). 355 K. The linewidth shows a continuous and a discon In other words, the dynamics near the order-disorder tinuous change at Tc\ and Tc2, respectively. The contin- transition temperature appears to be the origin of the different transitional behaviors. In this work, the chain length dependence of the transitional behavior was in vestigated by employing *H NMR for CnSn in view of the chain length dependence of molecular motions.

~~II. EXPERIMENTS~~

The CnSn sample used in this work was synthesized with much care to avoid impurities by the chemical re action: 2(n-C„H2n + iNH3Cl) + SnCR- 5H20 —> (n- CTiH2n +1NH3)oSiiClf, + 5H20. After filtering and two recrystallizations, white sugar-like crystals were finally obtained and then vacuum-dried and kept in a dry con 370 K dition for further work. The stoichiometry and the 350 K structure were checked by elemental analysis and X- 340 K 320 K ray diffraction (XRD). Differential scanning calorimetry 270 K (DSC) carried out between 123 K and 453 K shows two 220 K reversible phase transitions. The line shape and the spin- 1----- '----- 1----- '----- 1----- '----- 1----- 1----- 1 lattice relaxation time measurements were made by us -200 -100 0 100 200 ing 200 MHz *H NMR in the temperature range 150 - Frequency (kHz) 400 K. The spin-lattice relaxation times were measured by the conventional inversion recovery method. The line Fig. 1. Line shape evolution of CT6S11 as a function of shapes were obtained by Fourier-transforming the FID temperature. NMR Study of Distinct Phase Transitional Behaviors- • • - Kyu Won Lee et al. -247-

~~100-~~

~~Tempereture (K3 00 300 K Temperature (K) ■80 $~~

~~-60 P~~

a T. -o n 15 0 200 25 0 3 00 350 Temperature (K) Temperature (K) Fig. 2. FWHM linewidth of C16Sn as a function of temper ature. Inset: Dipolar splitting obtained from the line shapes Fig. 4. Temperature dependence of the spin-lattice relax as a function of temperature. ation rate in Cl6Sn. Inset: Fraction of the Tis component, As/{As + Al). The dotted line corresponds to Eqs. (1) and (2). The solid line corresponds to Eqs. (3) and (4). near T 1- Cl ■ C12Sn 1 analysis, the spin-lattice relaxation near Tc 1 is quite dif If X C14Sn n C16Sn ferent for n below and above n = 14. Figure 4 shows E the spin-lattice relaxation rates in C16Sn as a function ° 0.1 of temperature. As observed in C18Sn [4], T\s is be 1 G lieved to reflect the critical fluctuation around the phase • CZ3 -it- transition. The variation of the fraction of T\s, shown in the inset of Figure 4, supports the supposition, as the f n -□ fraction increases approaching TcA. 0.01 ■ □ As in C18Sn, the kinetic Ising model was introduced f for T“5] in CIGSn [4]: ■ □ T^ = - |(|^),T < (1) ■ 1 E-3 10 15 20 Til/=A2|tM-A-\T>T:, (2) X (S) in the limit wr(

ClOSn, C12Sn, C14Sn, and C16Sn. The activation en ergy of the NH3 group increases with increasing chain length, which indicates that the NH3 contribution to the spin-lattice relaxation rate shifts toward the high- 2 0 ■ temperature region. While the activation energy of the NH3 motion in C16Sn, 80 kj/mol, may be unreliable due to its large error, the activation energy of the NH3 motion can be taken to be very large, indicating that the NH3 motion contributes little to the spin-lattice relaxation around Tci. In practice, the large error simply originates from the insignificant contribution to the spin-lattice re laxation. Therefore, with increasing chain length, the molecular contribution to the spin-lattice relaxation rate decreases and the critical contribution increases, as if the 0 0 ■ two contributions compete with each other. On the other hand, near Tcl, the spin-lattice relaxation in C16Sn and C18Sn is of a double-exponential form in spite of the Temperature (K) strong dipolar coupling, where the strong spin diffusion is expected to induce a single-exponential type of relax Fig. 5. Temperature dependence of the spin-lattice relax ation curve. Therefore, it seems that the critical region is ation rate of Cl4Sn. The solid line corresponds to Eqs. (3) fairly well separated from the normal region, and in the and (4). The open squares are the difference between the critical region the molecular and critical contributions measured (solid square) and the fitted (solid line) spin-lattice to the spin-lattice relaxation coexist. When the molec relaxation rate near Tc\. ular contribution is dominant in the critical region, as in the case of short-chain systems, the spin-lattice relax TCi = TQleE'/RT(i = 1,2,... ,n), (4) ation is of a single-exponential form and the spin-lattice relaxation rate is determined by the molecular motions. where 7 is the proton gyromagnetic ratio, M-n the second When the critical contribution is dominant in the criti moment, uj the Larmor frequency, and S,: the activation cal region, as in the case of long-chain systems, the spin- energy. Three different types of the molecular motions lattice relaxation is of a double-exponential form and the (n = 3) were introduced, and the results of the fitting spin-lattice relaxation reflects the critical fluctuations. according to Eqs. (3) and (4) are shown in Figure 4 as a solid line. As in ClOSn and C12Sn, CH3 (M2 = 2.5 G2) and an unknown defect (M2 = 1.6 G2) dominate the spin-lattice relaxation below 300 K. However, the NH3 IV. SUMMARY (M2 = 3.1 G2) contribution below Tcl is very small. Figure 5 shows the spin-lattice relaxation rate 7j~* of In summary, two successive phase transitions in CnSn C14Sn, which was fitted to Eqs. (3) and (4) with three were investigated by means of ^H NMR. An order- types of molecular motions. The best fit is shown in Fig disorder transition and a conformational transition were ure 5 as a solid line. However, near Tcl, the spin-lattice characterized by the line shape evolution and the spin- relaxation rate was not well fitted to Eqs. (3) and (4). lattice relaxation. Around the order-disorder transition The open squares show the difference between the mea temperature, the spin-lattice relaxation was well sepa sured and the fitted T’1~1. In ClOSn and C12Sn [6,7], rated into the molecular motion and critical fluctuation the chain end gauchy defect largely contributes to the contributions. The distinct transitional behaviors, de spin-lattice relaxation near Tcl. We have attempted to pending on the chain length, were found to have a dy include one more motion, i.e. chain end defect motion in namical origin. the fitting, which did not improve the fitting. Consider ing that the critical contribution to the spin-lattice relax ation rate becomes dominant for long-chain compounds, the residual relaxation rate in C14Sn ma,v be the critical ACKNOWLEDGMENTS contribution, although other possibilities such as molec ular motions in the anisotropic potential wells cannot be excluded [17,18]. This work was supported by KISTEP (National Re The activation energy of the CII3 group is 1(1, 11, 10. search Laboratory and M102KS010001-02K1901-01814) and 7(± 3) kj/rnol in ClOSn. C12Sn. ClJSn, and ClOSn. and by the Korea Research Foundation (KRF-2004-005- respectively, which values are 110I much different. On the C00060/D00057 and Brain Korea 21 Project in 2004). other hand, the activation energy of the NH3 group is 40 C. E. L. gratefully acknowledges the Korea University k.J/mol. 55 kJ/mol. 58 kj/mol, and 80( ± 60) k.l /mol in Research Fund. NMR Study of Distinct Phase Transitional Behaviors- - - - Kyu Won Lee et al. -249-

REFERENCES [16] R. Kind, S. Plesko, H. Arend, R. Blinc, B. Zeks, J. Seliger, B. Lozar, .1. Slak, A. Levstic, C. Filipic, V. Zagar, G. Lahajnar, F. Milia and G. Chapuis, J. Chem. Phys. [1] O. K. Knop and W. J. Westerhaus, Can. J. Chein. 58, 71, 2118 (1979). 270 (1980). [17] J. Fenrych, E. C. R.eynhardt, S. Jurga and K. Jurga, [2] K. Kitahama, H. Kiriyama and Y. Baba, Bull. Chem. Soc. Molec. Phys. 78, 1117 (1993). Jpn. 52, 324 (1979). [18] S. Jurga, K. Jurga, E. C. Reynhardt and P. Katowski, [3] M. H. B. Ghozlen, A. Daoud, T. Molk, H. Poulet, M. Le Z. Naturforsch 48, 563 (1993). Postllec and N. Toupry, J. Raman Spec. 16, 219 (1985). [19] H. Zhang, N. P. Kulshrestha, S. E. Woehler and R. J. [4] K. W. Lee, C. H. Lee, C. E. Lee and J. K. Kang, Phys. Wittebort, J. Chem. Phys. 105, 2891 (1996). Rev. B 54, 8989 (1996). [20] J. Kroupa, A. Fuith, K. J. Shenk, H. Warhanek and M. [5] K. W. Lee, C. H. Lee, C. E. Lee and J. K. Kang, J. Chem. Ceramak, Ferroelectrics, 159, 109 (1994). Phys. 104, 6964 (1996). [21] M. Ceramak, F. Fuith, P. Vanek, J. Silha and J. Maikova, [6] K. W. Lee, M. W. Park, C. Rhee, C. E. Lee, J. K. Kang, Phys. Stat. Sol. (b) 182, 289 (1994). K. W. Kim and K. S. Lee, J. Chem. Phys. 108, 3019 [22] H. Elleuch, M. Kamoun, A. Daoud, J. M. Reau and J. (1998). Senegas, Phys. Stat. Sol. (b) 2 1 4, 141 (1999). [7] K. W. Lee, C. E. Lee, J. Kim and J. K. Kang, Solid State [23] K. W. Lee, D. K. Oh, C. E. Lee, J. K. Kang, C. H. Lee Commun. 124, 185 (2002). and J. Kim, J. Chem. Phys. 117, 8004 (2002). [8 ] C. E. Lee, K. W. Lee, C. H. Lee and D. K. Oh, J. Korean [24] R. Zhang, W. Sun, S. T. Nagle, R. L. Headrick, R. M. Phys. Soc. 42, 1182 (2003). Suter and J. F. Nagle, Phys. Rev. Lett. 74, 2832 (1995). [9] K. W. Lee, C. E. Lee, J. K. Kang and J. Kim, J. Korean [25] J. L. Lemmich, K. Mortensen, J. H. Ipse.n, T. Hpger, Phys. Soc. 41, 287 (2002). R. Bauer and O. G. Mourisen, Phys. Rev. Lett. 75, 3958 [10] M. J. Janiak, D. M. Small and G. G. Shipley, J. Biol. (1995). Chem. 254, 6068 (1979). [26] P. C. Mason, B. D. Gaulin, R. M. Epand and J. Katsaras, [11] R. Koynova and M. Caffrey, Biochim. Biophys. Acta Phys. Rev. E 61, 5634 (2000). 1376, 91 (1998); J. Sung, K. Park and D. Kim, J. Ko [27] P. C. Mason, J. F. Nagle, R. M. Epand and J. Katsaras, rean Phys. Soc. 44, 1394 (2004); C. Ay das and M. Kork Phys. Rev. E 63, 030902 (2001). in az, J. Korean Phys. Soc. 45, 975 (2004); Y. H. chang, [28] K. W. Lee, D. K. Oh and C. E. Lee, J. Phys. Soc. Jpn. C. H. Park and K. Matsuishi, J. Korean Phys. Soc. 44, 72, 2398 (2003). 889 (2004); K. Lee and T. C. Phong, J. Korean Phys. Soc. [29] B. I. Halperin and P. C. Hohenberg, Phys. Rev. 117, 952 43, 807 (2003). [12] P. Meier, E. Ohmes and G. Kothe, J. Chem. Phys. 85, (1969). [30] B. I. Halperin and P. C. Hohenberg, Phys. Rev. B 10, 3598 (1986). 139 (1974). [13] R. M. Venable, B. R. Brooks and R. W. Pastor, J. Chem. [31] Z. Racz and M. F. Collins, Phys. Rev. B 13, 3074 (1976). Phys. 112, 4822 (2000). [32] C. P. Slichter, Principles of Mganetic Resonance , [14] S. Konig, E. Sackmann, D. Richter, R. Zorn, C. Carlile (Springer-Verlag, Berlin, 1990); A. Abragam, Principles and T. M. Bayerl, J. Chem. Phys. 100, 3307 (1994). of Nuclear Magnetism, Oxford University Press (1983). [15] R. Blinc, M. I. Brugar, V. Rutar, B. Zeks, R. Kind, H. Arend and G. Chapuis, Phys. Rev. Lett. 43, 1679 (1979). Journal of the Korean Physical Society, Vol. 48. No. 1, January 2006, pp. 146^149

~~Blending MEH-PPV Conjugate Polymers with Single-Walled Carbon Nanotubes for Polymer Light Emitting Diodes~~

~~Sung Pyo Lee, H. Choi , Kyu Won Lee, Kyu Hyun Mo, Jae Won Jang , Eunmo Lee, In-Mook Kim and Cheol Eui Lee* Department of Physics and Institute for Nano Science, Korea University, Seoul 136-713~~

~~(Received 27 October 2005)~~

We report the effects of blending polv[2-methoxy-5-(2 ’-ethyl-hexyloxy)-p-phenylene-vinylene] (MEH-PPV) conjugate polymers with single-walled carbon nanotubes (SWNTs). The blend ratio of MEH-PPV/SWNT composites for polymer light-emitting diodes (PLEDs) ranged from 0 wt% to 1 wt% SWNTs. From time-of-flight. (TOP) measurements, the electron mobility was observed to increase gradually with the weight percent of SWNTs while the hole mobility remained almost unchanged. According to the Poole-Frenkel model, an increase in the weight percent of SWNTs lowers the zero-field activation energy, suggesting that SWNTs effectively reduce the oxygen trap density. At a 0.2-wt% SWNTs concentration, the electron and the hole mobilities could be tuned for the highest external quantum efficiency.

~~PACS numbers: 78, 72 Keywords: Polymer light-emitting diode, Time-of-flight. measurement, MEH-PPV polymer, Single-walled carbon nanotube~~

I. INTRODUCTION as Ca (2.9 eV), instead of A1 (4.3 eV), can be used to lower the electron injection barrier and, thus, to increase the electron injection current [6], Further improvement There has been great interest in light-emitt-ing diodes can be made by either i) introducing an electron trans (LEDs) based on conjugated polymers because of their porting layer [9] or ii) incorporating an electron-deficient potential applicability to large-area flat-panel displays moiety, such as oxadiazole (OXD), via physical blending operating at a relatively low voltage, along with their [10] or chemical bonding on the side chains of polymers promise of low cost and easy fabrication [1], Despite [11] . great progress in the development of polymer light- In spite of the aforementioned efforts, there are still emitting diodes (PLEDs), however, there still remain wild debates about the mechanisms to obtain highly ef fundamental limitations that have to be overcome for ficient PLED devices [3,6,7]. A device with a lower in wide application of such devices. Poly (phenyleneviny- jection barrier does not always have a higher efficiency. lene) (PPV) conjugate polymer, which is one of the most Nonetheless, it is generally accepted that the highest ef studied polymers in PLEDs due to its excellent lumines ficiency can only be obtained by balancing the carrier cent and mechanical properties [2,3], has been known densities and the carrier mobilities. Recently, it has as a hole transport material because its hole mobility is been demonstrated that the copolymerization of poly[ 2- one to three orders of magnitude higher than its electron metlioxy-5-(2 ‘-ethyl-hexyloxy)-p-phenylene-vinylene] (M mobility [4,5], Thus, an imbalance of electron and hole EH-PPV) with the electron-deficient triazole (TAZ) moi mobilities results in a shift of the recombination zone to ety can tune the electron and the hole current densities ward the cathode, which lowers the device efficiency due for improved efficiency of PLED devices [12]. However, to exciton quenching by the metal electrode. TAZ has both electron-transport and hole-blocking char Another reason for the low device efficiency is the elec acteristics so that a trade-off invariably occurs between tron injection current that is usually too low due to the the electron and the hole mobilities [13]. high electron injection barrier as in the case of TTO- This work reports the effects of blending MEH-PPV anode/PPV/Al-catliocle devices [6, 7], For a more effi conjugate polymers with single-walled carbon nanotubes cient LED device, it is necessary to enhance the injection (SWNTs). MEH-PPV is a hole-transporting material in efficiency of the electron carriers. In order to improve the which the high density of electron traps, such as oxygen electron injection properties, several methods have been impurities and conjugal tonal defects inhibits the electron proposed [8 ]: A cathode with a low work function, such transport. [14], As SWNTs have been demonstrated to be efficient oxygen-capturing materials [15], blending MEII- "Corresponding Author : [email protected] -146- Blending MEH-PPV Conjugate Polymers with Single-Walled- ■ • - Sung Pvo Lee et al -147-

PPV with SWNTs could possibly decrease the oxygen trap density and, thus, increase the electron mobility, which initiated the present study. • MEH.prVh 0.01 wi %_e v 0.01 wt 0 10 wl %_e II. EXPERIMENT 0.10 wl %_h

The MEH-PPV conjugated polymers were purchased from Aldrich, and the SWNTs were provided by Prof. C. ■ ■ 1.00 wl %_e 1.00 wl %_h J. Lee’s group (Korea University). MEH-PPV/SWNT composite solutions with SWNT concentrations of 0.01 to 1 wt% were prepared by blending the MWH-PPV 10.00 11.25 Electric field f 1Q5 V/cm with the SWNTs in chlorobenzene and dispersing the SWNTs by using a magnetic stirrer. The PLEDs were fabricated by spin-coating the composite solution onto O— ITO glasses of 1 cm x 1 cm to a thickness of about 'o 200 nm. For the cathode metal contact, a thin layer of A1 was deposited on the sample by thermal evaporation. Thus, PLEDs fabricated in a thin-sandwich configura tion, ITO/composite/Al, were obtained. For the time-of-fiight (TOF) measurements, 355-nm electron Nd-YAG laser pulses with durations of 7 ns were illu hole minated on one electrode. The photo-created carriers drifted across the sample under a reverse bias field (5 OX) ' 02 ’ 04 ' 06 ' 08 ' U0~ x 10-5 -lx 10~6 V/cm) to be collected on the oppo SWNTs concentration [wt %] site electrode. The incident photon energy (355 nm) was chosen to be well above the bandgap of the polymer. The Fig. 1. (a) Electron (filled symbols) and hole (empty sym optical absorption coefficient was large enough to ensure bols) mobilities versus electric field (5 x 1CT5 -lx 1(DC that the incident light would be absorbed within a very V/cm) for different SWNT concentrations and (b) averaged shallow penetration depth close to the illuminated sur mobilities as a function of SWNT concentration. face. The hole mobility was measured by illuminating the cathode (A1 electrode) with a laser pulse whereas the electron mobility was measured by illuminating the -lx 1CV8 cm2/Vs) [4], By increasing the concentra anode (ITO electrode). The I-V characteristics and the tion of SWNTs, we gradually raised the zero-field mo quantum efficiencies of the PLEDs were measured using bility as shown in Fig. 2 (b). The activation energy a PC-based assembly consisting of a Keithley 2400 source A was obtained from the zero-field electron mobility meter, a Newport optical power meter (model 1830-C), and a Newport low-power detector (model 818- UV). He=o = Hoe A/fcBT with /z0 = 1 x 1CV5 m2/Vs [17] and T = 300 K. The activation energy for MEH-PPV was ~0.28 eV, also in agreement with the earlier reported values (0.21 - 0.36 eV) [18]. As the concentration of III. RESULTS AND DISCUSSION SWNTs was increased, the activation energy was low ered, indicating that SWNTs, indeed, act to decrease Fig. 1 shows the carrier mobilities as a function of the the trap density in MEH-PPV. Oxygen has been consid electric field and as a function of the SWNTs concentra ered as one of the main traps in MEH-PPV [18,19], and tion; the electron mobility increases gradually with the SWNTs have been reported to capture oxygen very effi concentration of SWNTs whereas the hole mobility re ciently [20]. Thus, by blending MEH-PPV with SWNTs, mains almost constant. As the logarithmic plot of the the oxygen-trap density in MEH-PPV can be reduced, electron mobility vs E1'2 in Fig. 2 (a) shows, the elec giving rise to an enhanced electron mobility. Another tron carrier mobility // followed a Poole-Frenkel-like form piece of evidence for a reduction in trap density is the ft = ME=oexp(qv /E), a universal behavior in conjugated fact that the electric field coefficient decreased with the polymers, where /tE-0 is the zero-field mobility and q SWNT concentration (see the inset of Fig. 2 (b)). The the electric field coefficient [16]. electric field coefficient reflects the width of the energetic The zero-field mobility He=o and the electric field co disorder caused by the traps [21]. Hence, the decrease in efficient 7 were obtained by fitting the electron mobility the factor indicates a reduction in the energetic disorder to the Poole-Frenkel form. The zero-field mobility (~1.5 width, or equivalently, a reduction in the trap density

x 1CU10 cm2/Vs) for MEH-PPV is in good agreement [21]- with the previously reported values (5 x 1(U12 cm2/Vs For practical applications. PLED devices need good -148- Journal of the Korean Physical Society, Vol. 48, No. 1, January 2006

~~• MEH-PPVc 0.01 wi%_e 0.05 wt % c 0.10 wi %_« 0.15 wt%"c 0.20 wt %_c £ 0.30 wl % c 0.50 wt % c 1.00 wt % c I - Linear Fit~~

~~BOO 900 1000 Em rv^/cm1''2! Current density [mA/mm"'~~

I ■' (b)> z I /- - r 107 2000- 1 ^0.20 vV f 26 MEH-PPV 1 0.0! wl% 0.05 wl % -1" 1.. 3tj E 1000- 0.I5 wt% i L1 w r» < 0.20 wl % D □ 0.0 0 2 04 0.6 08 1 0 t 500- < 010- 0.50 wl % 1 V SWNTs concentration [wl %] 1.00 wt% ■ ~~-----~D------

~~0.2 0.4 0.6 0.8 1.5x10 SWNTs concentration [wt %] Electric field [V/m]~~

Fig. 2. (a) Logarithmic plot of electron mobility versus Fig. 3. (a) External quantum efficiency versus current den S1/2 with Poole-Frenkel-like linear fits and (b) zero-field elec sity and (b) current density versus electric field for different tron mobility and activation energy versus SWNT concen SWNT concentrations. tration. The inset shows the electric field coefficient versus SWNT concentration. characteristics at concentrations above 0.2 wt% SWNTs are distinct from those below it. As Fig. 3 (b) shows, the external quantum efficiencies, defined as the ratio of the current density increased by several orders of magnitude number of emitted photons to the number of injected for SWNT concentrations above 0.2 wt%. The abrupt electrons. The quantum efficiency measured for MEH- downfall of the quantum efficiency in spite of such a high PPV/SWNTs composites is shown in Fig. 3 (a). As current density would indicate that within the compos the concentration of SWNTs was raised from 0 to 0.2 ite polymer the injected carriers flow simply from one wt%, the quantum efficiency was improved by one order electrode to the other without recombination. of magnitude, which is attributed to the mobility bal A recent study on polymer/SWNTs composite pho ancing by SWNTs as observed in the mobility measure tovoltaic devices has demonstrated that SWNTs have ments. At 0.2 wt% SWNTs, where the highest external a large electron affinity (ranging from 3.4 to 4 eV), so quantum efficiency was obtained, the electron and the they can act as electron acceptors [22]. This large elec hole mobilities were almost equal (see Fig. 1), indicat tron affinity allows preferential transfer of the electrons ing that good mobility balancing is required for a high into the SWNTs while leaving the holes to be preferen quantum efficiency in PLEDs. tially transported through the polymer. In particular, if As the SWNT concentration was increased above 0.2 the SWNT concentration is high enough to provide con \vt%, however, the external quantum efficiency dropped ductivity (or percolation) paths, Ohmic contacts can be drastically. In view of mobility balancing, this is very formed bet ween 1 he conductivity path and the cathode surprising. Even at 1 wt% SWNTs. the mobility bal [22,23], allowing I lie transferred electrons to be efficiently ancing (the ratio of the electron mobility to the hole transported through I lie conductivity paths to the anode. mobility) is ~2, which is as good as the mobility balanc Now, our experimental results are well explained by ing of ~0.5 at 0.10 wt% SWNTs (see Fig. 1). Still, the supposing that above 0.2 wt% SWNTs, the SWNTs quantum efficiency at 1 wt% SWNTs is much smaller create electron conductivity paths between the elec (about three orders of magnitude smaller) than that at trodes, whereby the injected electrons are preferentially 0.15 wt% SWNTs. In fact, quite interestingly, the 7-1' transported via the conductivity paths while the holes Blending MEH-PPV Conjugate Polymers with Single-Walled- - ■ - Sung Pyo Lee et al. -149- are transported through the MEH-PPV polymer: This C. Scott, G. G. Malliaras and P. J. Brock, Appl. would reduce the radiative recombination rate between Phys. Lett. 74, 1132 (1999). the carriers, giving rise to a rapid decrease in the quan [5] B. K. Crone, I. H. Campbell, P. S. Davids and D. L. Smith, tum efficiency. Thus, it may be inferred that below Appl. Phys. Lett. 73, 3162 (1998); C. H. Lee, K. M. Lee, 0.2 wt% SWNTs, the MEH-PPV/SWNT composites C. E. Lee and W. Kang, Curr. Appl. Phys. 3, 359 (2003). are dominated by the oxygen-capturing property of the [6] I. D. J. Parker, Appl. Phys. 75, 1656 (1994); S. H. Kim C. H. Lee, K. W. Lee, C. E. Lee, W. Kang and K. S. Hong, SWNTs whereas above that, the composites are domi Curr. Appl. Phys. 4, 452 (2004). nated by the electron-accepting property of SWNTs. [7] A. J. Campbell, D. D. C. Bradley and D. G. J. Lidzey, J. In conclusion, by blending MEH-PPV with SWNTs, Appl. Phys. 82, 6326 (1997); C. E. Lee, N. S. Dalai and the oxygen trap density in MEH-PPV was effectively re R. Fu, Curr. Appl. Phys. 3, 355 (2003). duced. The electron mobility was, thus, increased while [8 ] T. W. Lee and O. O. Park, Adv. Mater. 12, 801 (2000). the hole mobility remained almost unchanged, giving [9] T. Fukuda, T. Kanbara, T. Yamamoto, K. Ishikasw, rise to much improved mobility balancing and enhanced H. Takezoe and A. Fukuda, Appl. Phys. Lett. 68 , 2346 quantum efficiency. The highest quantum efficiency was (1996); D. O'Brien, M. S. Weaver, D. G. Lidzey and D. obtained at 0.2 wt% SWNTs, where the electron and D. C. Bradley, Appl. Phys. Lett. 69, 881 (1996); H. M. the hole mobilities were tuned to equal levels. Above 0.2 Lee, K. H. Choi, D. H. Hwang, L. M. Do, T. Zyung, J. W. Lee and J. K. Park, Appl. Phys. Lett. 72, 2382 (1998); J. wt% SWNTs, the quantum efficiency rapidly decreased, W. Jang, D. K. Oh, C. H. Lee and C. E. Lee, J. Korean which is attributed to the preferential transport of elec Phys. Soc. 38, LI (2001). trons through conductivity paths created by the SWNTs. [10] Y. Cao, I. D. Parker, G. Yu, C. Zhang and A. J. Heeger, The present study demonstrates that blending MEH- Nature 397, 414 (1999). PPV with SWNTs, in comparison to earlier copolymer [11] Y. Z. Lee, Z. W. Chen, S. A. Chen, P. K. Wei and S. ization methods, is a relatively easy and cheap alterna J. Fann, Am. Chern. Soc. 123, 2296 (2001); J. W. Jang, tive for obtaining PLED devices with improved quantum C. E. Lee, D. W. Lee and J. I. Jin, Solid State Commun. efficiency. 130, 265 (2004); J. W. Jang, D. K. Oh, C. H. Lee, C. E. Lee, D. W. Lee and J. I. Jin, Appl. Phys. 87, 3183 (2000); S. J. Chung, K. Y. Kwon, S. W. Lee, J. I. Jin, C. H. Lee, ACKNOWLEDGMENTS C. E. Lee and Y. Park, Adv. Mater. 10, 1112 (1998). [12] L. S. Yu and S. A. Chen, Adv. Mater. 16, 744 (2004). This work was supported by the Korea Science [13] J. Kido, C. Ohtaki, K. Hongawa, K. Okuyama and K. Nagai, Jpn. J. Appl. Phys. 32, L917 (1993); M. Strukelj, and Engineering Foundation (M60504000021-05B0400- F. Papadirnitrakopoulos, T. M. Miller and L. J. Rothberg, 02110 and Proton Accelerator User Program No. Science 267, 1969 (1995). M202AK010021-04A1101-02110) and by the Korea Re [14] M. S. A. Abdou, F. P. Orfino, Y. K. Son and S. Holdcroft, search Foundation (Grant No. KRF-2004-005-C00060 J. Am. Chem. Soc. 119, 4518 (1997); E. J. W. List, C. H. and Brain Korea 21 Project in 2005). The measure Kim, J. Shinar, A. Pogantsch, G. Leising and W. Graup- ments at the Korean Basic Science Institute (KBSI) are ner, Appl. Phys. Lett. 76, 2083 (2000); N. V. Malm, J. acknowledged. Steiger, R. Schmechel and II. V. Seggern, J. Appl. Phys. 89, 5559 (2001). [15] P. G. Collins, K. Bradley, M. Ishigami and A. Zettl, Sci REFERENCES 1 2 3 4 ence 287, 180 (2000). [16] G. G. Malliaras, J. R. Salem, P. J. Brock and J. C. Scott, [1] J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. Phys. Rev. B 58, R13411 (1998); G. G. Malliaras and J. N. Marks, K. Mackay, R. H. Friend, P. L. Burns and A. C. Scott, J. Appl. Phys. 85, 7426 (1999). B. Holmes, Nature 347, 539 (1990); D. Braun and A. J. [17] S. J. Martin, M. J. Lupton, I. D. Samuel and A. B. Heeger, Appl. Phys. Lett. 58, 1982 (1991); E. Kim and S. Walker, J. Phys.: Condens. Matter 14, 9921 (2002). Jung, J. Korean Phys. Soc. 45, 1361 (2004); S. Kim and [18] V. Kazukauskas, H. Tzeng and S. A. Chen, Appl. Phys. H. R. Fetterman, J. Korean Phys. Soc. 42, 305 (2003); I. Lett. 80, 2017 (2002). Chang, S. Kim, J. S. Song, S. Choi, M. Cho, D. H. Choi [19] II. Antoniadis, L. J. Rothberg, F. Papadirnitrakopoulos, and J. Kim, J. Korean Phys. Soc. 45, 505 (2004). M. Yan, M. E. Galvin and M. A. Abkowitz, Phys. Rev. B [2] R. H. Friend, R. W. Gymer, A. B. Holmes, J. H. Bur 50, 14911 (1994). roughes, R. N. Marks, C. Taliani, D. D. C. Bradley. D. A. [20] L. Valentini, I. Armentano, L. Lozzi, S. Santucc.i and J. Dos Santos, J. L. Bredas, M. Logdlund and W. R. Sala- M. Kenny, Mater. Sci. k Eng. C 24, 527 ( 2004); N. Park, neck, Nature 397, 121 (1999); P. I\. H. Ho, J. S. Kim. J. S. Han and J. [Inn, Phys. Rev. B 64. 125401 (2001). H. Burroughes, H. Becker, S. F. Y. Li. T. M. Brown, F. [21] D. H. Dunlap, P. E. Parris and V. M. Kenkre, Phys. Rev. Cacialli arid R. Id. Friend, Nature 404, 481 (2000). Lett. 77, 542 (1996). [3] P. W. M, Blom and M. C. J. M. Vissenberg, Mater. Sci. [22] E. Ixymakis and G. A. ,J. Arnaratunga, Appl. Phys. Lett. k Eng. R. 27, 53 (2000); C. H. Lee, K. W. Lee and C. E. 80,112(2002). Lee, Curr. Appl. Phys. 3, 477 (2003). [23] J. N. Coleman, S. Curran, A. B. Dalton. A. P. Davey, B. [4] H. C. F. Martens, J. N. Huiberts and P. W. Blom, Appl. McCarthy, W. Blau and R. C. Bark lie, Phys. Rev. B 58, Phys. Lett. 77, 1852 (2000); L. Bozano, S. A. Carter, J. R7492 (1998). Journal ul

~~Derivation of dc X(tiiiivos in a Maonvliv Mullilaxvr { ndt'r^oiny Fcrrom ay notie Resonance~~

~~Doti'j l\mm (>ii nmi (. lunl I ui ! it~~

III (ills "(ilk. n v |>i cm'iiI i vi'iiij>ivhci)'ivc anil '\>lcili:iii; u.ivli l-u Niv Uvi nation ui 1 hr ilv wiil.iui- ;;vm i :llrd I iv ;i K nmlukv-ii' nm Ivi'Uuinv In iyvh M.isii'v . i-n^l It.ilh ihrix til l fx|iivwllv in tin. limii .if ihv >i>in ilil'luvimi Iviinili miivlt In uprt ihe vanavr niv.in ii'tv p:iili.

~~Kev xtni'ds : Vii!i %!./■-'.v'i:v;hi.;:-' _• r, i,- ...~~

~~I. introduction~~

I n :i nxvm prior. Ovl; ‘..ii :if ravvm i:: •*> vnilv- i 'iic Sv\ ■ i-ii'.'ii i-:t ;h. -.pin under i.hu lait'Oipnc'iL t:: >n< i |if|! I'J .. V : li;.' Hi: .1 :ullii winsuienx: --vlcvnoi iii lh ii l.ji 11 v : i::uv;:l I"/-/' iu! h\ ii ;..i o:i -pm imeriK-iii'". n- 'll n;. n sI i I’ll x;vr,-'-i;Ulu liv7 I1-" lo !t'i'"ui!.,!i'.'i, ;r ' : p;ni

~~'ll.'~~

~~2. !■ oi iimlatiun~~

~~n , dt 1 %~~

~~A' ■Hi .~~

~~I in I I - HI~~

;i i;i|i 'I ;i; . :i. '! . = 1 - "'I'.' : \ "l" ' - ..:r . ;u;: , I ..rl,: lly:; : "U ./u.'/k in \*ri| ; n: . ",i: ;:-r, iJ - i ' Ji;. \v( ' '//'%. "'I.'' H.-ru . -it; j ii i.-'-'.v I - :;.u, ,A: . H. ' I 1 | i ;i. ■ \> ‘.i.\- i ‘ II . Li’* ’ «‘V 1 ’• t ■ - \|'i.'!' -•; , ' - A Air. ' , | , I " | ,A' | ''U(-U H, . .'CA("\ r;\ i .i,\ 1- . U\T;r .-,i| :i nu^Hx: \\ r-:.ll| \l .1 ' X| ' | I I U'UU pur uru iu: ' wo;;-' ; .^iuri{ /; .|i. v, .\\r- - .=; |i-r unit * i u-. \ '.Y AMiMp j'.p. pAHtl.jiAj !J.V|,’i hi]

~~\ \ A ■------' '~~

~~iih: !i;:>i’.;|ip ."n,; finpiuiii .^v,~~

\ \ \ ■■’■'■‘.i I'-'.i ii| ipi:.i "'ij i''m i.-i i.;i:;- ,yh m:i|i ...Y.iuni ijr*".uf • i i!|.i .| :.ulMip;p a p uni v, 'Will Minjf , pir: uoi-.r;;11i) 11-.If .-u; v.i l Y' mm.ym .\| him a .*:, ; i’: i >.:!ii.'hL--Lin> iinoj' . i’-"x -iil’.i- \in-iiivi> HM ; i:.\ui.X po.,; op ' ' .

~~r luniHi.' vu|i-,iiv- •du-uul-• ; ' : ;v,:.'t| uviifp;~~

~~X A ;v-;uij H|', n: , :p Ma H upu ; ii.hxhj ,i.ji ■< \ H .< ; i:ii'.:'ix.' r>;ii;.-ip pur i.ir i;iik p -Sy~~

~~.ri| I{. .~~

~~hHAi.."' unln ur ! i~~

~~!: I -.nil ; !,i; :iiihh h ; - ui:| ji | j'-~~

.••ini'/’".pH ’ ii U|| j. U| , H '\ i 11’mi i m- ”i; . uA'' .. "r : ' ' ' •( ; I % - I ' -A ! • i ' 5 i j : - j IK'X H) ', .i ;//, : .;nu , -II . ' (HUS ' U; U p.\H ur U.U i-U , : -u\

~~r r(| .1! is.i-iijii: |n S;:-P|S.\< ;il4i I.'l l||vll.-i|~~

~~a "111 f!; oi .-ipiiui-,isii./ui ."k 'Pin ii •Xiisu.'p iu.\un.i~~

(i.'ir.i.i/i:ji i,,4'.i..!i 'i.-1 .111| U.MX:.X| UUiil:|P.I V'-nt' it ;l v.iiihiS,': ii:.xn | ,'iiji Mi; .XU Li'li.pi'. •Ili'lllflih.' LKiilsU!:.!) 3lij Ml : Si M 1 n11 |V LU'J ;-|KM,iiov:i;ii: ,ti|t jo UI.IO: j ('Hi! iir ■■ 11 XMS ui". ; i :| Ml ;h ui>nni]mup nmiA’p iii: .x. Aim o.i;m 11 v- i v ;. tii, ; l||i:d ' VonI' (if ^ '-M.l' ||-.| \ pU:' | : ; | >iiup .riXuui i|.-Hi;:: ipiun| uoiMisjip -Mu - ' 1 I f,g| i^r:)^ iij'^\ uJ .) 1' :l i .taw iimi-vp .1'|i ("P'i A'Jfi ( 6i 'll ' "S Ail|,| f 1 , ;:K:' '^(i _|u -* ji|i jr i A(tlth* Mimii;-|nui iu| inQ tjuuhui

~~ijUA n.iivu i' ui >u * i-1ii!.' .x| ;nix si|ns,xi iii; nriii~~

~~-J :uji u| .i’!:i:uox.ii itkkptuiviiti~~

~~iinn?inpnn :.^\r|ii;;iui ■'I.'iir .ui pi'ir.i.iu.W .">di’l|ui up~~

~~• .->11. .•.iriiAuj ;:.X1 M.-'t,' •i>i 11 i;i| %i|:ivr>; iuv:u«ul'. 11 vu'i-s.tid iu U"i)i'Vixp~~

~~-'■•'I 1 '-ufiHi:- •-:! -ir p.: imi;| u.w 'uiMMiiuiii'i ii[~~

~~"Xu A 'MK.KO ■-0D ruo: -IH>: 'i:\ -l'vM: .! }xii;v;ni.">iii™o|i: i.iPliui! us i.vu :i~~

~~•:v hi ywMi ooi.i yxu .'iiwun'riiiiiivi Y'.i(i i.:; S i.| j|,'llli, 1,1,1 I'Ui.lU! |nU:£UIJ~~

~~i f 1M P- 0! i:;i>-1oi ! vt'O "(jiilONVrilMX >\\ •XU w % - % sr ,p V’. sc ■;■>,>;.>(Irl-is ,xj i'-iiui at:|Mini n:;/;(::".''Y iirhiij; u ■'!■!! ',ui- ,ui. in liinimuoi 31|) MPH~~

~~VI ' ON .1||'| ’ l, p \~~

~~mu )\ A~~

~~,l!ii: = "\ III IV IMUimf Journal ol" ■~~

~~Nuclear Magnetic Relaxation Study of flu- Organic-Inorganic Hybrid Sxsteins (( J[:,,.1\IH)vSn(i„~~

~~kvti \\tin Ixv ami ( herd taii Lee~~

~~riiv H N\|R spin -lattice i-vlaMiliiot in :< writs of i he .irjjaiiiv-inoripmie Inliriii sisimis u;-( „l l:, Ml,rNn( X.~~

I ft - S. Ml. 12. 141 imdctijnlni: two suerc.ssivv phase iransilious w :x sluflivd. A disctunimiio vha rat teiisik of a first order phase transition was observed at ihe hi»li temperature ctmfm'mniiutial transition. Uv-idex the spin- lattice relaxation rale be Itm the eiittl'ormalional iransilitni temperature «

~~kc\ words: Xi.cieai r. .reliv rvia -..av, i i|a,i ,. r.i v.v IK M-. i o. a." - I’i-a-v a 11. > r-~~

~~1. Inlmtlitetiun s'' -1 ' S'- ' i lsn .ltd < ' .S" • is'cr- .i ,',!• ■ ;:'s.‘ i • a111,11' 1 liv i ivxa! the eel lead " •x mxllv \ \' \ v cv ix idvnh'ivd and w~~

i \ K Rh ., ( V \l l : M . liv. S:. IV ir. iK'bAX. .i! ! am. Mi: I'I I'lC six I a s' . i r vivnirs. (As. —: X \ \\ i lien:'...Sis cry ski Hi/ in me cnNv x 'iv ;si:!-;u :vn>V‘■ i" -.mv'll plia'-,,- jirvlini'iin met ill' spa ixPi > ' v :;iv i'Hl.h IVmpe -1 lllx'X 11 Sl-S.’i \\ i rx:-T ;'C'

~~;d phase Us.ill siileitis 1, '■IT.: lii.'ir A i\,.' ' nisi' a ;:isI i VII II , I |, s US'S~~

~~ilccre.ix ini; icmr'erMi:'. C 1 *- ‘v .XI’’: i".V lILx -~~

.riniciiuim ion Mi: V. \i\.. I> PV . lie:: n-II ii It Is .'IV 2. k Xpert men I Jonneiiand Uie Cl*ne nsil',!];;! 1 Ihe , ,xc I'JI slnic'mv is teduced fl'n D "i > X Mr ha-, v Me it x. !: i. id. I - i xi'iivlv.-. ii'Cu m ll'ii - work 1 Cl i Kir ihe Sri. 1' I in-.i : c" me < •-:tail:inioiiii,' n 'si,,i mm . ■, ,- v "vii :;:,i: iv. ivvi'ii A: I I , \ Midi

v.n 11 f S;""( - . ' ( -is:: (or sii.il ’ : .11 • i - I ! Mi .Slit. 1 '■! i t '>■ A;tv.• p: HillJs, .'IV ! • V ;V'.. - IC0"V "Ml - UK'. It hit 1-si in' hl.l | ,1, ir . III!. K\; :ll 'v.iiir ■" ' ' V Ii vuivl'x iiul |!k tin* Me, LiinmrnCM I'll ir r. ' 'ill . _ V VI'I II 11ns nii'ti i I: S'.': i v'. .' . :llV: X ‘ ill.:!'., v.s II ill \ IXIX ; i x v. ! Ik lit lill lit1

~~V [s es 1111 x ami I CiM lixxrmi r '•!.! K,~~

~~n r 1~~

~~C Ul/rr' Cl !:?u~~

~~" OT;:cd'-'J'c i'Kl~~

~~lia. 1 1 ha 1 ik-vt il,Ti v ! ’ 1 : u; in: -d % 1.1 I i:l 2 iii. it XMR 1 I 1.a.... ai,:n..v- n i.Hv. :a \n ,1- .~~

1 mi!a;id van!,i'l.iia ij vi' . m. ma ddm. :m : m-da i 'a ddnidd a 1 > inae,. a ma v dar di'kidai " (M' mia".a:il aian 1 ( IS'I i k'i’.k" i1 ; ivr.ii/.n.: and '• avd, : i.a- 11 a ... a An a.. 1 : lanad :a!:ia-. a. a a-. -,v:ms .!■ c:il n::v;i a: ilk i nad X\' R i.'c.a- ivonl>. A Liv:xr;ii k-mJvilv-. < >■' iio'Oj.-nrj pluyv :i.uisi:""i 1 ' - : I. In a' !. "■■ a 'li'ii : a:N - la, I 1 -a I Sn. V. Via : i 1’IOC 1 ,nad

~~iOlV pCT-liU! C« A11 i K'lv:i:-i 'a: V v.i'li IviiLti . m.|mn WlR 'Vi ns ida :ii . a d !ha- I I:iaa - '".a avaa’-.1:1 m a : : ! ' ;~~

ihv ■.>''..k'i-;.l;v AIknisili" ikUma.'d dm , -.: 11 m- ■ ' 'a aran :ai ::I aa.' ■:s>'t:;-jn! ajnn pd.isV ll';:nS|l;vii !v;d nvi'.ihil'iv. an -m ii-lainn.; 1 a: 1 >.,n :i 1: ■ a a a :' a.a a 1 a: a , .1 a 1 a. 1 an an a,a 1 a.:| ilia :a1 1.1.i -

V :hv i , Sn siimp’i-s s'iiv . ..d V -n:viIi.■-1/%:k 1;iv:■ 1 ]; 1! anda: 1 II ( X.n a -. a a .is m i i a ..a :-a i /• \. 1 'a...... : Ida i)\vr ill: anpi-J *iivp.:r:i ’iir. a\a..:a- aa: dm Valaa!! a! vln 'lv a. a" a . ■■ -.i a. n'U '.a ! I a :sv ia \ :! allil

~~ini’sniv-mi IviriXTiinrus. ’.‘.iivrc nvi'lh.'-h:..- it.nivxvMi '. ’ I :'P. I'aps iii an a-v v.ai'aa,, a a'a." MYaiil inniiirnk~~

* 11i 1 i- v-.vrx iiiuiui Ilk spi; l.iama ivl.jxiiXinii : iX: 1 MaViiVil a an i:i a ; . , , XI C . ,;nv 2 ' (a, a, am measurement-, m } :d. sRcv ;i disnvannmlv a' 1 . a..!"!;;, I'.ssi:' 'a..: 11: ihn \ 1: a ,aia and d'v 1. II. ai'oiiiX diarnctV! isr-c I'l a in si amRai ia.aa 'r :k :; i iI.:nA .:::a ' ka'aa'a ad iik.l dm 1. a ma;'r::r!'n1 .ml:i".iii anci'aies rephSCnl- ;i:c :iiUvr-dlsvivvi !r.a ail:, a; 'riiTViuliiiv 1 iRa. a.am aAkiimn dva; da- dm a, - ,a - ' ; ana 121. \n slVIli. MV ..Uiaiiiii!;, is a, a | :.i: a:;' i: - ! la .: ,s jin:: mu - ■ !■■ : n mi lit i md iiviio' M:.v.i,.a! a i 1 i 11-.: n'mi; laliina

K-poik-J ! X. ' al.a.aiiva. 'ilia iik ki k Xll .111.1 ' .fi ii I’l 11 i.id la .1 l!i:I!II Ilia ypin-UiliiCe rekt.viiivn kite dial;; Rv Iv -a nv;e veil dm: vvl i:' aaaR i: I d:a I an... ami Ida aiiai iaaidaia ;,.Ti laniialnim ihk'i.1 lv ilia lii’MHi'ikvu:,!: il:j'i'k-ilii'"''lv lain! ,ad. a a. iwi.>dh,.ik''.i In ,i! ;■ 'ii> i: p:. • ■' :::.a:av; X: aaiama a •: a - - la ' nan-' I' ’ V a n '. I: ■ a k'n ..am: a a. I VI ilk' I 1 llld i:’:i V.lilvlc:.. 1:' :'il:-i -a. Vail 11.| \i I : a a i: | a- a '' i a. I k' id.; ad:! ii ! :v a; 11 i '1 i” : 'a \ i ii ii: a.a : ' a: i I da X i ; a a ;. a. : a a 11 ! i n •' a' a. \ I' . a i a - ; ■ ' \ ■ >. / 1 _ ' li . a a ..." ' , aid n IX a da ( I ' i a. ad." via. i d-a . . i ■■ I.. a a: ■ ai ■ da: d : ■ • a a id d, ... j .. a

~~■ ■ :: .; -. a , ' a_ - . a..:,a.- d , d, a . | di j., i ■.!,,! d I N ’ . ■ -m a; ! ■ ; : ; d a - d I ai i ji'.dd ' a !~~

~~ill:..''~~

~~1 v : ■ d: 1 In a. I Ji aa a. a' a ■ i. a; a: I da linn Jnurnnl nl Miiajiel k >~~

~~Arkrntv, ledi’ments~~

~~I "i'-: v.,.'i;k qi:;v« Ale,.: O'e die Km on .Sue nee and ioainioe nni ; •ni.niJ.Anin , HU I • 2(u",i5-~~

~~References~~

~~1 I lin-M ii k H I'm, n, ’ ' M l.vniivioi: R H Nui Id ,ee: Id W. SI.:,1 1 (I ... -1 on,-. A Eila I'Xi~~

>■ if*, X 1 nc ;n:ni ;!li in : ( i :.v'- 1 LA l ' b!. urncJ Old':. 1 mm ihc '1 i nm : . pin - !;.!;ii.v Sii.1!• Hi meHSUU.'ITIL'II: Iv k.r;, -|' and X\ 1 V, e-.iiviotta. '■ do. J i lien: 58 (1 K XX i a,-. d i 1 1 no, avj Id 1: 1 ee, .1. ,4 7i .e 1 K XX. 1 oe. ; II hx. T. T: 1h;u nl K in i ' unida; hold : m won 'h:\ir 141 II. 1",. Vi i- X i; ,.vd. i XaOk dl' v'vi. VI 1 bus Ci.' 111! r u.mil In n> Mid ,he ! i i-e.Je,' .i>!,:1 N h,.,n: : ddn-i-a: Sfi'.-i Kn4>. did tlh; LA 1v. m mpe; ;-.i "b 1.1! l (o ,n u W: ir. U V imciri.liii.i i m; CI O'. I.IO n. pic

\V;u imvract mn. V. 1 !d i n, i AM Sl d him IdH.b Ii .Ml •ia iei-,. ! e rr; heeL'iu- 1 59, 1 1. n.K 1 1994 : thus b\:: mi erred : ill:-- r u;.ii v. h:nr) ilk i eMi-vl, 1 lanui d Xocc;. , .1 SViha. and : Maikova. .turn pi-.illl' li i Aim ||X .1 ; lee led \ ilk '\Sim. N i:l. del IH2: ; 2.S9 , ■■1)94; d>.ii in i;iccrui. 'H ns 1in 111, l\ ( "'! v droimn XV ; ,lv. X 1 XX'. I'arl:. Rhce. C H 1 .vc. J. K. Karat. tod K s 1 : ■a. 1 ri-A-ni Phyx. KlHi?:. mid ; i n k in Hr •- n >: ■> 'WlJ -A i Hi hub )l le in n<; !k, (he XX1 Kun, , Xu x : it’ ' Nl 1; Viump vie spi!e W. i A v. t 1 1 w. an, 1 I. K kang, 1‘hyv. !h'.' | m' ‘own; il'lv -ni:‘l!v if K ' :m s ."' I.i 54:• v.:! h hsise S:i.;hlcr. /'.O', 1 ,'n'ie,'fa:d'e.mik,',':, e {Sponger tuns m :nid ;i v wit irmnli;mu( irans: 1. n ere ; .r- 'lie,.' Oedti a :'did 1 X A;'-l aaiar-l, d-"aI'pin' ,Xu A iV R ■ i'.l'Cl )\ Ml 1 ;l -H- H vm : ? i - L (Hie; .1X: laloid ’ on, 1 a 1:

~~i emi-iiK Id, ii:; i:ii' ill;' spi nl,un k':1 idum:!o n iuie\ and id idlain a-, i . "hern Phvx 92,; 25 . 'A'ilfl 1,,'ur !!;, l*iv . rrn 4; ei. L'llCl S t'-l ill.' 0, : ,1 'll! un n>. m;u~~