Chemically Adjusting Plasma Temperature, Energy, and Reactivity (CAPTEAR) method using NOx and combustion for selective synthesis of Sc 3N@C80 metallic nitride fullerenes

Article abstract of DOI:10.1021/ja077305z

Goals are (1) to selectively synthesize metallic nitride fullerenes (MNFs) in lieu of empty-cage fullerenes (e.g., C₆₀, C₇₀) without compromising MNF yield and (2) to test our hypothesis that MNFs possess a different set of optimal formation parameters than empty-cage fullerenes. In this work, we introduce a novel approach for the selective synthesis of metallic nitride fullerenes. This new method is Chemically Adjusting Plasma Temperature, Energy, and Reactivity (CAPTEAR). The CAPTEAR approach with copper nitrate hydrate uses NO_x vapor from NO_x generating solid reagents, air, and combustion to tune the temperature, energy, and reactivity of the plasma environment. The extent of temperature, energy, and reactive environment is stoichiometrically varied until optimal conditions for selective MNF synthesis are achieved. Analysis of soot extracts indicate that percentages of C₆₀ and Sc₃N@C₈₀ are inversely related, whereas the percentages of C₇₀ and higher empty-cage C_2n fullerenes are largely unaffected. Hence, there may be a competitive link in the formation and mechanism of C₆₀ and Sc₃N@C₈₀. Using this CAPTEAR method, purified MNFs (96% Sc₃N@C₈₀, 12 mg) have been obtained in soot extracts without a significant penalty in milligram yield when compared to control soot extracts (4% Sc₃N@C₈₀, 13 mg of Sc₃N@C ₈₀). The CAPTEAR process with Cu(NO₃)₂· 2.5H₂O uses an exothermic nitrate moiety to suppress empty-cage fullerene formation, whereas Cu functions as a catalyst additive to offset the reactive plasma environment and boost the Sc₃N@C₈₀ MNF production.

Full text of DOI:10.1021/ja077305z

Published on Web 12/01/2007
Chemically Adjusting Plasma Temperature, Energy, and
Reactivity (CAPTEAR) Method Using NOx and Combustion for
Selective Synthesis of Sc3N@C80 Metallic Nitride Fullerenes
Steven Stevenson,* M. Corey Thompson, H. Louie Coumbe, Mary A. Mackey,
Curtis E. Coumbe, and J. Paige Phillips
Contribution from the Department of Chemistry and Biochemistry,
UniVersity of Southern Mississippi, Hattiesburg, Mississippi 39406
Received September 24, 2007; E-mail: Steven.Stevenson@usm.edu
Abstract: Goals are (1) to selectively synthesize metallic nitride fullerenes (MNFs) in lieu of empty-cage
fullerenes (e.g., C60, C70) without compromising MNF yield and (2) to test our hypothesis that MNFs possess
a different set of optimal formation parameters than empty-cage fullerenes. In this work, we introduce a
novel approach for the selective synthesis of metallic nitride fullerenes. This new method is “Chemically
Adjusting Plasma Temperature, Energy, and Reactivity” (CAPTEAR). The CAPTEAR approach with copper
nitrate hydrate uses NOx vapor from NOx generating solid reagents, air, and combustion to “tune” the
temperature, energy, and reactivity of the plasma environment. The extent of temperature, energy, and
reactive environment is stoichiometrically varied until optimal conditions for selective MNF synthesis are
achieved. Analysis of soot extracts indicate that percentages of C60 and Sc3N@C80 are inversely related,
whereas the percentages of C70 and higher empty-cage C2n fullerenes are largely unaffected. Hence, there
may be a “competitive link” in the formation and mechanism of C60 and Sc3N@C80. Using this CAPTEAR
method, purified MNFs (96% Sc3N@C80, 12 mg) have been obtained in soot extracts without a significant
penalty in milligram yield when compared to control soot extracts (4% Sc3N@C80, 13 mg of Sc3N@C80).
The CAPTEAR process with Cu(NO3)2‚2.5H2O uses an exothermic nitrate moiety to suppress empty-cage
fullerene formation, whereas Cu functions as a catalyst additive to offset the reactive plasma environment
and boost the Sc3N@C80 MNF production.
Introduction
Discovered in 1999,1 Metallic Nitride Fullerenes (MNFs) are
molecules which, arguably, are more exciting than their empty-
cage fullerene cousins (e.g., C60, C70) in that MNFs1-6 have an
entrapped trimetallic nitride cluster (Figure 1). The judicious
selection of transition or rare-earth metal imparts a unique set
of chemical and physical properties for MNFs. Examples include
the use of Gd3N@C80 for MRI agents,7,8 Lu3N@C80 for X-ray
contrast agents,9 Ho3N@C80 for radiotherapeutics, Er3N@C80
for optical properties,10 and Sc3N@C80 for electronic11 applica-
Figure 1. Structures of empty-cage and metallic nitride fullerenes produced
in electric-arc soot.
(1) Stevenson, S.; Rice, G.; Glass, T.; Harich, K.; Cromer, F.; Jordan, M. R.;
Craft, J.; Hadju, E.; Bible, R.; Olmstead, M. M.; Maitra, K.; Fisher, A. J.;
Balch, A. L.; Dorn, H. C. Nature 1999, 401, 55-57.
tions and as a “practice” molecule for chemical functionalization
of MNF carbon cages.7,8,12-19
(2) Krause, M.; Kuzmany, H.; Georgi, P.; Dunsch, L.; Vietze, K.; Seifert, G.
J. Chem. Phys. 2001, 115, 6596-6605.
(3) Stevenson, S.; Lee, H. M.; Olmstead, M. M.; Kozikowski, C.; Stevenson,
P.; Balch, A. L. Chem.sEur. J. 2002, 8, 4528-4535.
(11) Larade, B.; Taylor, J.; Zheng, Q. R.; Mehrez, H.; Pomorski, P.; Guo, H.
Phys. ReV. B 2001, 64, #195402.
(4) Stevenson, S.; Phillips, J. P.; Reid, J. E.; Olmstead, M. M.; Rath, S. P.;
Balch, A. L. Chem. Commun. 2004, 2814-2815.
(12) Cai, T.; Ge, Z. X.; Iezzi, E. B.; Glass, T. E.; Harich, K.; Gibson, H. W.;
Dorn, H. C. Chem. Commun. 2005, 3594-3596.
(5) Yang, S. F.; Dunsch, L. Chem.sEur. J. 2005, 12, 413-419.
(6) Yang, S. F.; Kalbac, M.; Popov, A.; Dunsch, L. Chem.sEur. J. 2006, 12,
7856-7863.
(13) Cardona, C. M.; Kitaygorodskiy, A.; Echegoyen, L. J. Am. Chem. Soc.
2005, 127, 10448-10453.
(7) Chen, Z.; Fatouros, P. P.; Corwin, F. D.; Broaddus, W. C.; Dorn, H. C.
Neuro-Oncology (Durham, NC, U.S.) 2006, 8, 492-492.
(8) Fatouros, P. P.; Corwin, F. D.; Chen, Z. J.; Broaddus, W. C.; Tatum, J. L.;
Kettenmann, B.; Ge, Z.; Gibson, H. W.; Russ, J. L.; Leonard, A. P.;
Duchamp, J. C.; Dorn, H. C. Radiology 2006, 240, 756-764.
(9) Iezzi, E. B.; Duchamp, J. C.; Fletcher, K. R.; Glass, T. E.; Dorn, H. C.
Nano Lett. 2002, 2, 1187-1190.
(14) Cardona, C. M.; Kitaygorodskiy, A.; Ortiz, A.; Herranz, M. A.; Echegoyen,
L. J. Org. Chem. 2005, 70, 5092-5097.
(15) Iezzi, E. B.; Cromer, F.; Stevenson, P.; Dorn, H. C. Synth. Met. 2002, 128,
289-291.
(16) Iezzi, E. B.; Duchamp, J. C.; Harich, K.; Glass, T. E.; Lee, H. M.; Olmstead,
M. M.; Balch, A. L.; Dorn, H. C. J. Am. Chem. Soc. 2002, 124, 524-525.
(17) Iiduka, Y.; Ikenaga, O.; Sakuraba, A.; Wakahara, T.; Tsuchiya, T.; Maeda,
Y.; Nakahodo, T.; Akasaka, T.; Kako, M.; Mizorogi, N.; Nagase, S. J.
Am. Chem. Soc. 2005, 127, 9956-9957.
(10) Macfarlane, R. M.; Bethune, D. S.; Stevenson, S.; Dorn, H. C. Chem. Phys.
Lett. 2001, 343, 229-234.
9
10.1021/ja077305z CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 16257-16262
16257

A R T I C L E S
Stevenson et al.
Application development for MNFs by academia and industry
has been hampered by poor sample availability. However, recent
advances in nonchromatographic purification methods20,21 and
further advances in achieving isomerically purified22,23 samples
have led to a flurry of research activity in MNF fundamental
research and application development. Historically, typical soot
extracts1,20,21 contain <10% MNFs with >90% of extract
samples being contaminant empty-cage fullerenes (Figure 1).
There is literature precedence for using chemical additives
in packed graphite rods for electric-arc generation of fullerene
extracts. Additives to plasma can be introduced as solids (e.g.,
calcium cyanamide,24 CoO,3,25,26 Cu27) or gases (e.g., NH3).28-30
The ammonia method uses a reactive atmosphere28-30 that
suppresses empty-cage fullerene formation and provides MNF
soot extracts with a high percent purity, but with a corresponding
reduction in milligram MNF yield.
Figure 2. Concept of using the CAPTEAR method to adjust and optimize
the temperature, energy, and reactivity of the plasma environment to “tune”
the type of fullerene produced.
In this work, we introduce a new approach (CAPTEAR) that
permits soot extracts of high MNF percent purity and without
the typical, corresponding penalty in milligram yield. The
CAPTEAR concept is based on the hypothesis that different
types of carbonaceous nanomaterials (e.g., nanotubes, empty-
cage fullerenes, metallic nitride fullerenes) have preferred
temperatures of formation and stability in a given chemically
reactive plasma (Figure 2). With CAPTEAR, we can “shift” or
“tune” the type of fullerene produced by changing and optimiz-
ing the temperature, energy, and reactivity of the plasma
environment.
that rods packed with Cu(NO3)2‚2.5H2O were NOT cured in a furnace
prior to use to avoid premature thermal decomposition.
Resulting soot was extracted with o-xylene, and the solvent was
removed under reduced pressure to furnish a dried extract, which was
washed with solvent (e.g., diethyl ether, acetone). Soot extracts were
weighed and characterized by HPLC to determine the type and amount
of fullerene material present. HPLC peak areas were obtained using
standard chromatographic integration software (Vernier, Logger Pro),
which did not account for variations of extinction coefficients.
Nevertheless, calculated amounts of fullerenes (e.g., Sc3N@C80)
obtained from peak integrations were comparable to isolated amounts.
The reactor and analysis methods had an error of ∼10%. HPLC
separations used a PYE column (Phenomenex, 4.6 mm × 250 mm)
with a 50 µL injection, 1.0 mL/min toluene mobile phase, and 360 nm
UV detection.
Experimental Section
The cylindrical electric-arc reactor had dimensions of 25 cm diameter
and a length of 43 cm. In preparation for vaporization, 1 in. (2.5 cm)
diameter graphite rods (6 in. length) were core-drilled to 4 in. (10 cm)
to leave an outer shell of 30 g of C as previously described.27 These
cored rods were then packed with Cu (Cerac), Cu(NO3)2‚2.5H2O, or
NH4NO3 (Aldrich) and various amounts of Sc2O3 powder (325 mesh,
Stanford Materials, CA). For control experiments without these
additives, 30 g of Sc2O3 were packed into cored, graphite rods. The
reactor was vacuum pumped with subsequent backfilling of He gas to
a reactor pressure of 300 Torr. An air flow of 6 Torr/min was introduced
into the reactor, and experiments were conducted under a dynamic flow.
Other reactor parameters were 220 A and a gap voltage of 38 V. Note
Results and Discussion
By adjusting the amounts and reagents supplied to the reactor,
the temperature, energy, and reactivity of the plasma can be
“tuned” toward the selective synthesis of one type of fullerene
(i.e., Sc3N@C80) versus empty-cage fullerenes (e.g., C60, C70).
The goal then is finding the optimal convergence of temperature,
energy, and reactivity for selective synthesis. We have selected
the thermal decomposition of Cu(NO3)2‚2.5H2O, whose reactive
byproducts have been previously investigated.31,32 Significant
heat release, chemical reaction, and subsequent decomposition
products have been reported31,32 and are shown in Scheme 1.
In our experimental design, we adjust the temperature, energy,
and amount of reactive, oxidizing NOx vapor by stoichiomet-
rically varying the amount of copper nitrate hydrate packed in
the Sc2O3 filled graphite rod.
(18) Stevenson, S.; Stephen, R. R.; Amos, T. M.; Cadorette, V. R.; Reid, J. E.;
Phillips, J. P. J. Am. Chem. Soc. 2005, 127, 12776-12777.
(19) Shustova, N. B.; Popov, A. A.; Mackey, M. A.; Coumbe, C. E.; Phillips,
J. P.; Stevenson, S.; Strauss, S. H.; Boltalina, O. V. J. Am. Chem. Soc.
2007, 129, 11676-11677.
(20) Ge, Z. X.; Duchamp, J. C.; Cai, T.; Gibson, H. W.; Dorn, H. C. J. Am.
Chem. Soc. 2005, 127, 16292-16298.
(21) Stevenson, S.; Harich, K.; Yu, H.; Stephen, R. R.; Heaps, D.; Coumbe, C.;
Phillips, J. P. J. Am. Chem. Soc. 2006, 128, 8829-8835.
(22) Cai, T.; Xu, L. S.; Anderson, M. R.; Ge, Z. X.; Zuo, T. M.; Wang, X. L.;
Olmstead, M. M.; Balch, A. L.; Gibson, H. W.; Dorn, H. C. J. Am. Chem.
Soc. 2006, 128, 8581-8589.
(23) Stevenson, S.; Mackey, M. A.; Coumbe, C. E.; Phillips, J. P.; Elliott, B.;
Echegoyen, L. J. Am. Chem. Soc. 2007, 129, 6072-6073.
(24) Wolf, M.; Muller, K. H.; Skourski, Y.; Eckert, D.; Georgi, P.; Krause, M.;
Dunsch, L. Angew. Chem., Int. Ed. 2005, 44, 3306-3309.
(25) Olmstead, M. H.; de Bettencourt-Dias, A.; Duchamp, J. C.; Stevenson, S.;
Marciu, D.; Dorn, H. C.; Balch, A. L. Angew. Chem., Int. Ed. 2001, 40,
1223-1225.
(26) Olmstead, M. M.; de Bettencourt-Dias, A.; Duchamp, J. C.; Stevenson, S.;
Dorn, H. C.; Balch, A. L. J. Am. Chem. Soc. 2000, 122, 12220-12226.
(27) Stevenson, S.; Mackey, M. A.; Thompson, M. C.; Coumbe, H. L.; Madasu,
P. K.; Coumbe, C. E.; Phillips, J. P. Chem. Commun. 2007, 4263-4265
(28) Dunsch, L.; Georgi, P.; Krause, M.; Wang, C. R. Synth. Met. 2003, 135,
761-762.
(29) Dunsch, L.; Krause, M.; Noack, J.; Georgi, P. J. Phys. Chem. Solids 2004,
65, 309-315.
(30) Dunsch, L.; Yang, S. Small 2007, 3, 1298-1320.
The mechanism, formation, and combustion reactions of NOx
have also been reported.33-36 A review of kinetics, mechanisms,
and roles of NO and N2O from N2 and O2 combustion is
available.35 The thermal mechanism of NO formation has been
(31) Ding, Z.; Martens, W.; Frost, R. L. J. Mater. Sci. Lett. 2002, 21, 1415-
1417.
(32) L’vov, B. V.; Novichikhin, A. V. Spectrochim. Acta, Part B 1995, 50,
1459-1468.
(33) Malte, P. C.; Pratt, D. T. Fifteenth International Symposium on Combustion
1975, 1061-1070.
(34) Monat, J. P.; Hanson, R. K.; Kruger, C. H. Combust. Sci. Technol. 1977,
16, 21-28.
(35) Tomeczek, J.; Gradon, B. Combust. Sci. Technol. 1997, 125, 159-180.
(36) Zeldovich, Y. B. Acta Physicochem., USSR 1946, 21, 577-628.
9
16258 J. AM. CHEM. SOC. VOL. 129, NO. 51, 2007

Synthesis of Sc3N@C80 Metallic Nitride Fullerenes
A R T I C L E S
Figure 3. Comparison of percent fullerene versus weight percent of added
Cu(NO3)2‚2.5H20 for (a) C60, Sc3N@C80 and (b) C70, higher fullerenes,
C2n (e.g., C76, C78, C82, C84).
Figure 4. Results from CAPTEAR experiments: (a) milligrams of fullerene
type produced versus weight percent of added Cu(NO3)2‚2.5H2O, (b) HPLC
chromatogram, and (c) MALDI mass spectrum of fullerene soot extract
corresponding to 80% copper nitrate hydrate/20% Sc2O3.
Scheme 1. Reactions and Decomposition Products of Volatile31
and Nonvolatile32 Species of Copper Nitrate Hydrate
were compared to control experiments having only Sc2O3 packed
(i.e., no Cu(NO3)2‚2.5H2O) in the cored graphite rods. The data
shown in Figure 3 indicate an inverse relationship between the
empty-cage C60 and Sc3N@C80 metallic nitride fullerene (MNF).
In contrast, the percentages of C70 and higher cage fullerenes,
C2n, appear to be largely unaffected for extracts prepared from
rods packed with 0 to 70 wt % added copper nitrate hydrate.
When conditions are optimized for MNF formation (i.e., 80%
copper nitrate hydrate), the percentages of C70, higher C2n, and
C60 are at their peak minima (∼0%). Beyond 80% copper nitrate
hydrate, MNF purity declines steadily to zero. Empty-cage
fullerene production, however, increases to comparable amounts
of C60 and C70 (i.e., C60/C70 ) 1:1) relative to control
experiments where C60/C70 ) 3:1. With CAPTEAR, MNF purity
increases from 4% in the control to 96% Sc3N@C80 at 80%
copper nitrate hydrate. The remaining 4% of this soot extract
is predominantly Sc3N@C78, with only traces of empty-cage
fullerenes observable. Also at 80% copper nitrate hydrate
additive, the amount and percent of C60 produced in the soot
extract decreased from 67% C60 (control) to ∼0%.
Scheme 2. NOx Formation33,35,36 in Combustion Process
described as a sequence of reactions (5)-(9) and are shown in
Scheme 2.35,36
To investigate CAPTEAR variables (i.e., plasma temperature,
energy, reactivity), experiments were performed using an array
of cored, graphite rods with increasing packing material content
of Cu(NO3)2‚2.5H2O relative to Sc2O3. These rods were
vaporized under comparable conditions, the soot was collected,
and their extracts were analyzed for fullerene type, percent
abundance (Figure 3), and milligram yield (Figure 4). Results
Note that when a rod packed with 80% copper nitrate hydrate
was vaporized, the plasma became noticeably larger. The
pressure in the reactor rose steeply due to the reactive gases
formed from CAPTEAR decomposition and combustion reac-
tions. Reactor flanges and electrode assemblies were signifi-
cantly hotter relative to control experiments. The amount of heat
generated was proportional to the amount of copper nitrate
9
J. AM. CHEM. SOC. VOL. 129, NO. 51, 2007 16259

A R T I C L E S
Stevenson et al.
Figure 5. Correlation of weight percent added Cu(NO3)2‚2.5H20 versus burn time.
hydrate added to the rod. Hence, we were able to “adjust” the
temperature in the reactor by controlling the amount of copper
nitrate hydrate (Scheme 1) added to the packing material. We
are attributing this CAPTEAR shift to extra temperature, energy,
and reactivity from the copper nitrate hydrate to the plasma,
which, as a result of the additive, is more optimal for selective
MNF synthesis (Figure 2). Too much copper nitrate, however,
results in a plasma unoptimized for MNFs, and the yield of
Sc3N@C80 drops precipitously.
burn times are associated with higher loadings of copper nitrate
hydrate and are consistent with rod vaporization under more
oxidizing and reactive environments. Note that control rods with
only Sc2O3 packing material needed 100 min to vaporize
compared to only 35-50 min for packed rods with high nitrate
content.
When the percent purity (Figure 6a) and milligram yield
(Figure 6b) are plotted versus burn time, another startling
discovery is made. The data suggest different optimal burn times
for various fullerene types. For extracts with higher MNF purity,
decreased burn times are optimal. Note the MNF purity drops
exponentially from 96% Sc3N@C80 (35 min) to only 4%
Sc3N@C80 (100 min) at longer burn times. In contrast to the
decay curve for Sc3N@C80, the percentage of empty-cage
fullerenes (e.g., C60, C70, higher C2n) exponentially increases
to a plateau with increasing burn time. Figure 6b shows the
corresponding milligram yields of soot extracts relative to burn
time. In contrast, the peak maximum for making the highest
milligram yield of Sc3N@C80 is not at 35 min (12 mg) of burn
time. Rather, the peak maximum occurs at 50-60 min (33 mg
of Sc3N@C80) followed by a steady decline in MNF yield
beyond 60 min. During the 60-100 min time frame, in which
Sc3N@C80 MNF yield is declining, empty-cage fullerene
production increases dramatically. This 60-100 min time frame
also corresponds to decreasing copper nitrate hydrate, which is
otherwise clearly deleterious for empty-cage fullerene produc-
tion. In general, increasing copper nitrate hydrate results in (1)
shorter burn times, (2) higher MNF percent purity, and (3)
suppression of empty-cage fullerenes.
The hypothesis of different optimal reaction conditions for
empty-cage fullerenes versus MNFs is also supported in Figure
4a. For the first time, we have experimentally shown a separate,
discrete peak maximum for MNF production relative to empty-
cage fullerenes. This optimal convergence at 80% copper nitrate
hydrate (20% Sc2O3) has resulted in an extracted soot sample
consisting of 12 mg of Sc3N@C80 at 96% purity. The corre-
sponding HPLC trace and MALDI mass spectrum are shown
in Figure 4b and c, respectively.
In a control experiment using no added copper nitrate(Figure
4a), the percent purity of MNF in the extract is only 4%
Sc3N@C80, corresponding to 13 mg of Sc3N@C80 in the extract
(305 mg). With only 10% added copper nitrate hydrate, the
percentage of Sc3N@C80 MNF tripled (12%), and the milligram
yield increased from 13 mg to 30 mg of Sc3N@C80. This factor
of 2.5 increase in MNF milligram yield is attributed to the Cu
metal from the Cu(NO3)2‚2.5H2O (Scheme 1). A similar boost
in MNF milligram yield by a factor of 2.5 from Cu metal
additive has recently been reported.27 With CAPTEAR, how-
ever, the MNF purity increases dramatically from 4% to 96%.
This remarkable boost in MNF yield is attributed to the effect
of the nitrate hydrate moiety of the Cu(NO3)2‚2.5H2O and its
subsequent decomposition and combustion products (Schemes
1 and 2).
Data for specific regions are summarized in Table 1.
CAPTEAR maximum enhancements in percent purity for soot
extracts are 4% Sc3N@C80 (control) to 96% Sc3N@C80 (80 wt
% copper nitrate hydrate). CAPTEAR maximum enhancements
in milligram yield for soot extracts are 13 mg of Sc3N@C80
(control) to 33 mg of Sc3N@C80 (70 wt % copper nitrate
hydrate). Although a higher yield of Sc3N@C80 (33 mg) was
obtained using 70 wt % additive, the corresponding purity was
only 47% Sc3N@C80, but still a factor of 10 improvement in
purity relative to the control (4% Sc3N@C80).
Also of interest is the correlation of burn time to the amount
of copper nitrate hydrate. Figure 5 shows that increasing
amounts of copper nitrate hydrate correspond to faster “burn
times,” as defined by the time needed to vaporize the graphite
rods packed with Sc2O3/copper nitrate hydrate. The more rapid
9
16260 J. AM. CHEM. SOC. VOL. 129, NO. 51, 2007

Synthesis of Sc3N@C80 Metallic Nitride Fullerenes
A R T I C L E S
Figure 7. Comparison of Sc3N@C80 purity (%) and Sc3N@C80 yield (mg)
from three soot extracts prepared from vaporizing packed rods containing
80 wt % of each of the following additives: Cu, Cu(NO3)2‚2.5 H2O, and
NH4NO3.
packed rods with 80% Cu, and 12 mg of Sc3N@C80 (96%
purity) were obtained from rods packed with 80 wt % Cu(NO3)2‚
2.5H2O.
Based on these results, we conclude that two types of plasma
additives promote the selective synthesis of Sc3N@C80 while
simultaneously maintaining comparable milligram yields of
MNFs. The high Sc3N@C80 MNF purity arises from a reactive
moiety (e.g., NOx from the copper nitrate), whereas the copper
boosts the milligram yields of MNFs. In this experimental
design, note that both the reactive moiety and catalyst additive
are contained within a single compound (copper nitrate hydrate).
Alternatively, another experimental design for MNF synthesis
could, in principle, introduce the plasma additives separately.
For example, a rod containing Cu metal additive (i.e., no nitrate)
could be vaporized in an atmosphere of NH3 (e.g., a different
type of reactive gas).
Figure 6. Comparison of (a) percent type and (b) amount of fullerene
produced versus burn time.
Table 1. Summary of Fullerene Yield Data from Soot Extracts
Obtained from Vaporizing Rods Packed with Various Amounts of
Cu(NO3)2‚2.5H20
Conclusions
70%
80%
90%
copper
nitrate
hydrate
copper
nitrate
hydrate
copper
nitrate
hydrate
We have investigated our hypothesis of different optimal
plasma conditions for empty-cage fullerenes versus MNFs.
Intuitively, optimal reactor conditions for C60 and other empty-
cage fullerenes should be different compared to MNF formation,
which requires encapsulation of a trimetallic nitride cluster. To
pursue the selective MNF synthesis idea, our experimental
design sought to chemically modify the plasma environment
until optimal MNF formation parameters (i.e., temperature,
energy, and reactivity) were obtained. We stoichiometrically
varied amounts of solid-phase reagent (i.e., copper nitrate
hydrate) to “tune” exothermicity and combustion in the plasma
reactor, thereby adjusting the corresponding temperature, energy,
and reactivity. Results indicate the percent composition of C60
and Sc3N@C80 in CAPTEAR soot extracts were inversely
related, whereas the percentages of C70 and other empty-cage
0% added
(control)
% Sc3N@C80
mg Sc3N@C80
mg C60
4
13
204
70
16
2
47
33
15
11
5
96
12
52
8
2
∼0
∼0
∼0
<1
13
mg C70
4
mg higher fullerenes, C2n
mg Sc3N@C78
mg extract
<1
<1
16
5
69
305
To investigate our hypothesis that the nitrate moiety permits
higher MNF purity while the Cu moiety of the copper nitrate
hydrate boosts MNF yield, we performed a set of experiments
using three types of additives, each containing copper and/or a
nitrate. Figure 7 compares soot extracts from rods packed with
Cu metal, Cu(NO3)2‚2.5H2O, and NH4NO3. The data indicate
that all rods packed with nitrates permitted high MNF purities
for Sc3N@C80. A purity of 96% Sc3N@C80 was achieved from
rods packed with 80 wt % Cu(NO3)2‚2.5H2O. A similar purity
of 98% Sc3N@C80 was obtained from rods packed with 80 wt
% NH4NO3, but only 1 mg of Sc3N@C80 was obtained. In
contrast, all rods packed with a copper source had significantly
higher Sc3N@C80 MNF yields (mg). In summary, 16 mg of
Sc3N@C80 (only 6% purity) were produced in soot extracts from
C2n fullerenes were relatively unaffected from 0 to 70% copper
nitrate hydrate additive. This finding suggests a “competitive
nature” for C60 and Sc3N@C80. With the CAPTEAR method,
purified Sc3N@C80 MNFs (96% purity, 12 mg) have been
achieved in soot extracts, while maintaining a comparable yield
(mg) relative to control soot extracts (4% Sc3N@C80, 13 mg of
Sc3N@C80). We believe the CAPTEAR process with copper
nitrate hydrate uses a twofold approach. First, the exothermic
nitrate moiety and decomposition vapors (i.e., experiments using
9
J. AM. CHEM. SOC. VOL. 129, NO. 51, 2007 16261

A R T I C L E S
Stevenson et al.
NH4NO3 and Cu(NO3)2‚2.5H2O) suppress empty-cage fullerene
formation to obtain soot extracts with high MNF purity. Second,
the Cu moiety functions as a catalyst additive for MNFs, offsets
reactive plasma conditions, and boosts Sc3N@C80 MNF produc-
tion.
0547988), the Department of Education GAANN Fellowship
Award No. P200A060323, and the Lucas Research Foundation
for financial support. The project described was also supported
by Grant No. R15AG028408 from the National Institute On
Aging.
Acknowledgment. We thank the National Science Founda-
JA077305Z
tion, NSF CAREER/EPSCOR Bimolecular Processes (CHE-
9
16262 J. AM. CHEM. SOC. VOL. 129, NO. 51, 2007