generation mechanism of carbon cluster, fullerene, and ...maruyama/papers/99/seoul_paper.pdf ·...
TRANSCRIPT
Generation Mechanism of Carbon Cluster, Fullerene, and Metallofullerene
Shigeo Maruyama
Engineering Research Institute and Department of Mechanical Engineering,
The University of Tokyo 2-11-16 Yayoi, Bunkyo-ku, Tokyo 113-8656, Japan
e-mail: [email protected] URL: http://www.photon.t.u-tokyo.ac.jp/~maruyama
Abstract:
The formation mechanism of empty and metal-containing fullerene was studied through molecular dynamics simulations and Fourier Transform Ion Cyclotron Resonance (FT-ICR) mass spectroscopy of laser vaporized carbon cluster. With classical molecular dynamics simulations using modified Brenner potential, the clustering process starting from 500 isolated carbon atoms in gas phase was simulated under the controlled temperature condition. When the control temperature was at Tc = 3000 K, imperfect caged clusters like C60 and C70 were obtained. Additional annealing simulations to compensate the short time-scale of the simulation resulted the perfect Ih-C60 structure through Stone-Wales transformations. A fullerene formation model featuring the random caged structure and annealing at the size range of C40 to C60 was proposed through the detailed study of the precursor structures in simulations. In order to incorporate metal atoms in the simulation, multi-body classical potential functions for metal-carbon and metal-metal interactions were constructed based on DFT (density functional theory) calculations of various forms of small clusters MCn and Mn (M = La, Sc, Ni). The classical potential was expressed with the Morse term and the Coulomb term as function of coordinate number of a metal atom. The simulated clustering process with addition of 1 % of metal atoms was compared with the pure carbon simulation. When La atoms were applied, the stable open-cap structure surrounding the La atom resulted in the La-containing caged cluster. For Sc-C system, the host carbon clusters were not affected so much as the La-C case due to the weaker Coulomb interaction, and the Sc atom was encapsulated in the host cage at the final stage of the growth process. Ni-C system was also simulated to explore the possible role of metal atoms in the generation of SWNT. The precursor clusters were similar to those in Sc-C system, although the Ni atom finally stayed on a face of 7 or 8 member-ring of the caged structure.
Fourier Transform Ion Cyclotron Resonance (FT-ICR) mass spectrometer directly connected to the laser vaporization cluster beam source was implemented to study the clustering process. With increase of cluster nozzle pressure, three different types of positive mass spectra were obtained for pure carbon clusters: smaller than about C60 with odd numbered clusters up to C40; almost only C60 and a trace of C70; well-known even atom mass with intense peaks at C60 and C70. Qualitatively the lower pressure condition of the cluster source corresponded to the earlier stage of the MD simulation. Through these comparisons, we speculated that the even-numbered clusters
corresponded to the annealed random caged clusters. The FT-ICR mass spectra of metal-carbon binary clusters were studied for sample materials normally used for generation of metal-containing fullerene and SWNT; La: 0.8%, Y: 0.8%, Sc: 0.8%, Gd: 0.8%, Ce: 0.8%, Ca: 0.3%, and Ni:4.2% - Y: 1%. Positive La-C, Y-C, Sc-C, Gd-C, Ce-C binary clusters commonly showed strong MC2n
+ signal in the range of 36 < 2n < 76 with intense magic numbers at MC44
+, MC50+ and MC60
+. Characteristics of these small clusters were compared with results of molecular dynamics simulations.
In order to further study the information of clusters appearing in the mass spectra, reactivity of negative carbon clusters and metal-carbon binary clusters to NO were measured by the FT-ICR spectrometer. For empty clusters, even-odd alternation of reactivity was clearly shown; even clusters were less reactive. Furthermore, carbon clusters with La atom such as LaC44
- were very much unreactive to NO. The reactivity of clusters contaminated with a hydrogen atom was very curious. One hydrogen atom made odd-numbered clusters less reactive and even-numbered clusters more reactive. These experimental results were perfectly explained by a consideration of number of dangling bonds based on the random-raged geometric structure predicted by the molecular dynamics simulations.
1
Congratulations to The 10th Anniversary of
the Institute of Advanced Machinery and DesignSeoul National University
Congratulations to The 10th Anniversary of
the Institute of Advanced Machinery and DesignSeoul National University
Generation Mechanism of Carbon Cluster, Fullerene,
and Metallofullerene
Generation Mechanism of Carbon Cluster, Fullerene,
and Metallofullerene
Shigeo Maruyama
Engineering Research Institute& Department of Mechanical Engineering
The University of Tokyoe-mail: [email protected]
URL: http://www.photon.t.u-tokyo.ac.jp/~maruyamaCo-Workers:Masamichi Kohno, Shuhei Inoue, Tetsuya Yoshida, Hideaki Hayashi, and Yasutaka Yamaguchi
Int. Symp. Micro/Nano Scale Mech. Eng.Dec. 1999, Seoul University
�Fullerene Formation Mechanism�Metal-Containing Fullerene �Hydrogen Storage by SWNT
�Laser-Vap. Cluster Source+ FT-ICR Mass Spectrometry+ Time of Flight
�Arc-Discharge Generator�Laser-Oven Generator (SWNT)
Fullerene & Carbon Nanotube
�Non-Adiabatic MD?�Laser Annealing of Clusters�Photo-Fragmentation
Light-Matter Interaction
�Liquid Droplet on Solid Surface�Vapor Bubble Nucleation�Droplet and Surface Tension�Thermal Boundary Resistance
Inter-Phase Phenomena
�Tight-Binding and Quantum MD�Dissociation Dynamics�Phase Change (Polycrystalline)
�Photo-Fragmentation in TOF�Chemical Reaction in FT-ICR�Quantum Dots?
Silicon Clusters, Surface & Bulk
Molecular Dynamics SimulationsExperimental
Current Research ProjectsCurrent Research Projects
2
Typical Structures of FullereneTypical Structures of Fullerene
(a) C60 (c) La@C82(b) C70
(e) SWNT (f) MWNT
C60
PVWin
(d) Sc2@C84
D S Ra c d
R R
e e= = = == = = == =
−6 325 129 15 13150 80469 0 011304 19 2 517 2 0
1
0 0 0
1 2
. eV . . � . �
. . .. � . �
βδ
A A
A A
Potential parameters
From D. W. Brenner: Phys. Rev. B, 42, 9458(1990)
1f
r0
R1 R2
Cut-off function
k
i j
θijk
Molecular Dynamics SimulationC-C Potential Function{ }��
<
−=i ij
ijAijijRb rVBrVE)(
* )()(
{ })(2exp1
)()( ee
R RrSSDrfrV −−−
= ��
��� −−
−= )(2exp
1)()( e
eA Rr
SSSDrfrV β
{ }δ
θ−
≠��
���
�+=
+= �
),(
* )()(1,2 jik
ikijkcijjiij
ij rfGBBB
B
���
����
�
++−+= 22
0
20
20
20
0 )cos1(1)(
θθ
dc
dcaGc
Total Energy Eb:
3
C10
C8
25001000 1500 2000500Time (ps)
C60C49
C28C26
C12 C15
C8
C330
20
40
60
C70C53
C8 Clu
ster
Siz
e
500 carbon atoms 342 Å cubic boxTc = 3000 K
Growth Process
PVWin
74
7
77
4 8
77
7
77
77 7
initial
215.81 ns 215.82 ns 216.35 ns 216.40 ns
216.45 ns 217.18 ns 218.39 ns 220.56 ns 221.70 ns (Ih C60)
A
215 ns
B
–6.72
–6.68
–6.64
–6.6
0
4
# of
dam
glin
gbo
nds
ND
B
Ep
NDB
pote
ntia
l ene
rgy
E p(e
V)
time (ns)190 200 210 220 230
Ih C60
Annealing Process to perfect C60
PVWin
4
Chain Ring FlatC10 C20 C30
Tangledpoly-cyclic
Closed cage Stone-Walestransformations
C50
C70C60
Fullerene(stable)
Opencage
Graphitic sheetToo low temperature
Chaotic 3-dimensionalstructure
Too hi
gh te
mperat
ure
Randomcage
Higherfullerene
Present Fullerene Formation Model
Studied Metal Atoms
Group 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18Period
1 21 H He
1.008 4.0033 4 5 6 7 8 9 10
2 Li Be B C N O F Ne6.941 9.012 10.81 12.01 14.01 16.00 19.00 20.18
11 12 13 14 15 16 17 183 Na Mg Al Si P S Cl Ar
22.99 24.31 26.98 28.09 30.97 32.07 35.45 39.9519 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
4 K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr39.10 40.08 44.96 47.88 50.94 52.00 54.94 55.85 58.93 58.69 63.55 65.39 69.72 72.61 74.92 78.96 79.90 83.80
37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 545 Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe
85.47 87.62 88.91 91.22 92.91 95.94 99.00 101.1 102.9 106.4 107.9 112.4 114.8 118.7 121.8 127.6 126.9 131.355 56 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86
6 Cs Ba * Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn132.9 137.3 178.5 180.9 183.8 186.2 190.2 192.2 195.1 197.0 200.6 204.4 207.2 209.0 210.0 210.0 222.0
87 88 104 105 106 107 108 109 110 111 1127 Fr Ra ** Unq Unp Unh Uns Uno Une Uun Uuu Uub
223 226 261 262 263 262 265 266 269 272 277----
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
138.9 140.1 140.9 144.2 145 150.4 152 157.3 158.9 162.5 164.9 167.3 168.9 173.0 175.089 90 91 92 93 94 95 96 97 98 99 100 101 102 103Ac Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr227 232 231 238 237 239 243 247 247 252 252 257 258 259 262
* Lanthanides
** Actinides
5
M-C and M-M Potential Function Expression
CARij VVVE ++=
VR: Repulsive term VA: Attractive term
VC: Coulomb term
N C: carbon coordinate number
B*: normalized bond order
cM, cC : charge of M (+) and C(-)
f(rij) : cut-off function
{ })(2exp1
)( eije
ijR RrSSDrfV −−−
= β { })(/2exp1
)( *eij
eijA RrS
SSDBrfV −−
−⋅−= β
ijijC r
ccerfV MC
0
2
4)(
πε−=
�≠
+=)(carbon
C )(1jk
ikrfN
{ } δ)1(1 C* −+= NbB
)exp(3 2C
1M kNkc +−−= CMC / Ncc =
�≠
+=)( metal
M )(1jk
iki rfN2
MMji
ijNNN +=
{ })1(exp)( 21 −−+= ijDeeije NCDDND
{ })1(exp)( 21 −−−= ijReeije NCRRNR
M-C M-M
N Mi: metal coordinate number
ij
jiijC r
ccerfV MM
0
2
4)(
πε=
Metal-Carbon Potential functions
2 3 4
–2
–1
0
–3
pote
ntia
l ene
rgy
(eV)
distance r (Å)
VC (N C=15)
(a) La-C
La2
VC (N C=5)
Eb (N C=15)
Eb (N C=5)
–3
–2
–1
0
2 3 4
pote
ntia
l ene
rgy
( eV)
distance r (Å)
(b) Sc-C
VC (N C=15)
Sc2
VC (N C=5)
Eb (N C=15)Eb (N C=5)
–3
–2
–1
0
2 3 4
pote
ntia
l ene
rgy
( eV)
distance r (Å)
(c) Ni-C
Ni2
Eb (N C=15)
Eb (N C=5)
Table 1. Poten tial parameters for metal-carbon in teractions.De (eV) S β (1/Å) Re (Å) R 1 (Å) R 2 (Å) b δ k1 k2
La-C 4.53 1.3 1.5 2.08 3.2 3.5 0.0854 -0.8 0.0469 1.032Sc-C 3.82 1.3 1.7 1.80 2.7 3.0 0.0936 -0.8 0.0300 1.020Ni-C 3.02 1.3 1.8 1.70 2.7 3.0 0.0330 -0.8 - -
Table 2. Poten tial parameters for metal-metal interactions.S β (1/Å) De1 (eV) De2 (eV) CD Re1 (Å) R e2 (Å) CR R 1 (Å) R 2 (Å)
La-La 1.3 1.05 0.740 2.64 0.570 3.735 0.777 0.459 4.0 4.5Sc-Sc 1.3 1.4 0.645 1.77 0.534 3.251 0.919 0.620 3.5 4.0Ni-Ni 1.3 1.55 0.74 1.423 0.365 2.520 0.304 0.200 2.7 3.2
6
time (ps)20001000 30000
0
20
40
0
10LaC2 LaC5 LaC6LaC9
LaC3
LaC5
LaC35 LaC46
60
80
20LaC15 LaC17
C12
C13
LaC19
LaC48
C20
La@C73LaC51
(a) Growth process of a La@C73
(b) Growth process of a LaC17
cluster size
Growth Process of LaCn Clusters
Control temperature Tc = 3000 K500 C & 5 La in (342 Å)3 cubic box
PVWin
Collision Process of La Attached Clusters
cluster size
time (ps)4000 5000
0
20
40
60
LaC24 La2C38
LaC14
(La@C54) La
C16
7
Cluster Source Nozzle for FT-ICR
Pulsed Valve
Target Disk
Waiting Room
Focused Laser Beam
He 10 atm
Expansion Cone
Cluster Beam
Cluster Source for FT-ICR
Window
To ICR CellFast Pulsed Valve
ExpansionCone
“Waiting” RoomTarget Disc
Gears
Gears
Window
Feedthroughfor Up-down
Feedthroughfor Rotation
Vapo
rizat
ion
Lase
r
8
FT-ICR (Fourier Transform Ion Cyclotron Resonance)Mass Spectrometer
Turbopump
Gate Valve
Cluster Source
6 Tesla Superconducting Magnet
100 cm
DecelerationTube
Front Door
Screen Door
Rear Door
Excite & DetectionCylinder
ElectricalFeedthroughGas Addition
Ionization Laser
Probe Laser
rmvqvBF
2
==
mqB
rvf
ππ 22==
Mechanism of Fourier Transform Ion Cyclotron ResonanceMass Spectrometry (FT-ICR)
Mechanism of Fourier Transform Ion Cyclotron ResonanceMass Spectrometry (FT-ICR)
Magnetic FieldDigital
OscilloscopeDigital
Oscilloscope
PreAmplifier
Arbitrary WaveformGenerator
Excite
Detect Ion
Back Door
ICR Cell
9
Example of Excite and Detect WaveformExample of Excite and Detect Waveform
0 500 1000Frequency (kHz)
Inte
nsity
(arb
. uni
ts)
C60+
Excite
Detect
0 10 20 30 40 50Time (ms)
Volta
ge (a
rb.)
Excite Detect
40 50 60 70 80
720 740Ion Mass [amu]
C60+
Number of Carbon Atoms
Inte
nsity
(arb
. uni
ts)
LaC44+ LaC50
+
LaC60+
Lanthanum-carbon binary clusters
10
Growth Process of a Sc Containing Cluster
ScC3
60
4000
ScC18 ScC37 ScC39
C11
C12
ScC43 ScC49 Sc@C54 Sc@C550
20
40
20000time (ps)
cluster size
1000 3000
Graphite-like Open cage
Metal outside Metal inside
Fan-type Closed cage
Sc: 5, C: 500 in 34.2nm box
PVWin
40 50 60 70 80
720 760 800Ion Mass [amu]
ScC60+
C64+
Sc2C56+
Number of Carbon Atoms
Inte
nsity
(arb
. uni
ts)
ScC60+
C60+
ScC50+
ScC44+
Scandium-carbon binary clusters
11
For Production of Single Wall Nanotube (SWNT)For Production of Single Wall Nanotube (SWNT)
Combination of Metals: Ni-Co or Ni-Y is necessaryfor Production of Single Wall Nanotube
TEM Pictures of SWNT RopesTEM Pictures of SWNT Ropes
Individual tube diameter: 1.3 nmSpacing: 0.34 nmMisalignments and Terminations
12
Molecular Simulation of Hydrogen Storage in SWNT
Hydrogen 5376 molecules
Bundle of SWNT560x8-140
=4340 C Atoms
Growth Process of Ni Attached Clusters
Graphite-like Open cageChain or Ring
Metal inside & outside
time (ps)20001000 3000 4000
0
20
40
NiC2
NiC18 NiC35 NiC49 NiC50NiC44
C16
cluster size
400 800 1200 1600
0
10NiC6 NiC10 NiC12 NiC14NiC9
Ni: 5, C: 500 in 34.2nm boxPVWin
13
Tc = 2500 KAnnealing of NiC78 and Ni2C54 Cluster
# of
pos
itive
inne
r pro
duct
s
time (ps)400 8000
10
20
30
insideoutside
400 800
NiC78 Ni2C54
500 700 900 1100 1300Ion Mass [amu]
Inte
nsity
(arb
. uni
ts)
Ni:4.2% Y:1.0%
YC50+
YC60+
YC44+
C60+
f1 time = 370µs
f1 time = 380µs (x5)
Y2C82+YC82
+YC70+
Nickel-Yttrium-carbon clustersNickel-Yttrium-carbon clusters
14
Yttrium-carbon binary clustersYttrium-carbon binary clusters
500 700 900 1100 1300Ion Mass [amu]
Inte
nsity
(arb
. uni
ts)
f1 time = 365µs
f1 time = 375µs (x2)
YC60+
Y:0.8%
YC50+
YC44+
C60+
YC70+ YC82
+
Y2C82+
As injected Negative Clusters
As injected Negative Clusters
400 600 800
20 40 60 80Number of Carbon Atoms
Inte
nsity
(arb
itrar
y)
(a) pure carbon
(b) La–C
(c) La–C
(1:130)
(1:130)
Cluster ion mass (amu)
15
Reaction with NO 10-5 Torr
Reaction with NO 10-5 Torr
440 460 480 500
36 38 40 42Number of Carbon Atoms
Inte
nsity
(arb
itrar
y)(a) as injected
(b) SWIFTed
(c) NO 1s
(d)NO 3s
(e) NO 10s
Cluster ion mass (amu)
C37–+C37H
–C40
–
C37H– C37NO–
C40NO–
Even-odd Difference of Reactivity for
Negative Pure Carbon Cluster
Even-odd Difference of Reactivity for
Negative Pure Carbon Cluster
400 500
30 33 36 39 42
Cluster ion mass (amu)
Inte
nsity
(arb
itrar
y)
(a) as injected
(b) SWIFTed
(c) NO 1s
Number of Carbon atoms
C30– C33
– C36–
C39–
C33NO– C39NO–
16
Even-odd Difference of Reactivity for
Negative Pure Carbon Cluster
Even-odd Difference of Reactivity for
Negative Pure Carbon Cluster
450 500 550 600
38 40 42 44 46 48 50Number of Carbon Atoms
Inte
nsity
(arb
itrar
y)
(a) as injected
(b) SWIFTed
(c) NO 1s
Cluster ion mass (amu)
C38– C41
– C44–
C47–
C41NO– C47NO–
LaC44- was not
Reactive to NO.LaC44
- was not Reactive to NO.
520 570 620 670 720
45 48 51 54 57 60Number of Carbon Atoms
Inte
nsity
(arb
itrar
y)
(a) as injected
(b) SWIFTed
(c) NO 1s
Cluster ion mass (amu)
C44–
C47– LaC44
–
C44NO– C47NO–
17
55 60Number of Carbon Atoms
Inte
nsity
(arb
itrar
y)(a) as injected
(b) NO 2s
LaC44–
LaCoddIs More
Reactive
LaCoddIs More
Reactive
Reaction with H During formation
(Dirty Source)
Reaction with H During formation
(Dirty Source)
42 44 46 48Number of Carbon Atoms
Inte
nsity
(arb
itrar
y)
3.6%6.4%43.0%47.0%n: odd1.6%7.2%19.6%71.6%n: evenCnH3CnH2CnH Cn
Average of n = 42-51
18
42 44 46 48Number of Carbon Atoms
Inte
nsity
(arb
itrar
y)LaC36
- has no Hydrogen Atom by Dirty Source.
LaC36- has no
Hydrogen Atom by Dirty Source.
LaC36-
Even Clusters with H
Reacts with NO.
Even Clusters with H
Reacts with NO.
52 54 56Number of Carbon Atoms
Inte
nsity
(arb
itrar
y)
(a) as injected
(b) NO 2s
19
Summary for Bare ClustersSummary for Bare Clusters
Ceven: Less Reactive
Codd: More Reactive
Effect of H: Unreactive
Effect of H: More Reactive
Reaction with NO
Reaction with H (Dirty Source)
Ceven: Less Reactive
Codd: More Reactive 3.6%6.4%43.0%47.0%n: odd1.6%7.2%19.6%71.6%n: evenCnH3CnH2CnH Cn
Average of n = 42-51
���
→+ −
−−
NOCHNOC
NOHCeven
eveneven
−− →+ NOCNOC oddodd
Possible Reaction SitesPossible Reaction Sites
Even Numbered Clusters# of Dangling Bond = 0, 2, 4, …
Odd Numbered Clusters# of Dangling Bond = 1, 3, 5, …
H or NO
Euler’s Theorem:223
323 42,2
2,
23,2 vfvvvvfvveevf +−=+=+=+=+=+
20
Revisit the Use of Euler’s TheoremRevisit the Use of Euler’s Theorem
Euler’s Theorem: 2+=+ evff: faces, v: vertices, e: edges
Usual Explanation of Even Numbered Positive Spectra
���
+=+=
+=
65
65
65
653652
ffvffe
fff
���
+==
6
5
22012
fvf
5f6f
23
2323
vvv
vve
+=
+=
More General TreatmentMore General Treatment
3vEuler’s Theorem: 2+=+ evf
When All Carbon has 3 Bonds
ve23=
2v
42223 −=→+=+ fvvvf
When Some of Carbon has Dangling Bond
23 422
2vfvvf +−=→+=
v must be even
v2 is even/odd when v is even/odd
21
Summary for La-C Binary ClustersSummary for La-C Binary Clusters
Ceven: Unreactive
Codd: Reactive
Effect of H: Unreactive
Reaction with NO
Reaction with H (Dirty Source)
Ceven: Less Reactive (less than 10% react)
Codd: More Reactive (about 40% react)
)36( ≥n
MD SimulationLa25+C2500, 3000K
MD SimulationLa25+C2500, 3000K
1
7
6
2
3
4
5
8
910
11
22
La@’’’C47
La@Cn Structures (n = Odd) Observed in Molecular Dynamics Simulation
La@Cn Structures (n = Odd) Observed in Molecular Dynamics Simulation
La@’C47
La@C54 La@’’C54
La@Cn Structures (n = Even) Observed in Molecular Dynamics Simulation
La@Cn Structures (n = Even) Observed in Molecular Dynamics Simulation