Thermal decomposition mechanism of single-molecule precursors forming metal sulfide nanoparticles

Article abstract of DOI:10.1021/ja905353a

Several zinc alkyldithiocarbamates [Zn(S₂CNR₂) ₂] were synthesized, and their thermal decomposition in the presence of alkylamines was studied in order to understand the formation of metal sulfide nanoparticles. The major

Full text of DOI:10.1021/ja905353a

Published on Web 12/14/2009
Thermal Decomposition Mechanism of Single-Molecule
Precursors Forming Metal Sulfide Nanoparticles
Yun Ku Jung, Jae Il Kim, and Jin-Kyu Lee*
Department of Chemistry, Seoul National UniVersity, Seoul 151-747, South Korea
Received July 10, 2009; E-mail: jinklee@snu.ac.kr
Abstract: Several zinc alkyldithiocarbamates [Zn(S2CNR2)2] were synthesized, and their thermal decomposi-
tion in the presence of alkylamines was studied in order to understand the formation of metal sulfide
nanoparticles. The major intermediates and side products were isolated under various conditions and
characterized by NMR spectroscopy and LC-MS. The analysis results showed that nucleophilic attack of
the metal-coordinated amine on the most electron-deficient thiocarbonyl carbon of the alkyldithiocarbamate
ligands at high temperature initiated the decomposition to generate thiourea, hydrogen sulfide, and solid
metal sulfide nanoparticles. From the proposed decomposition mechanism, the role of amines in thermolysis
was studied, and methods for synthesizing various metal sulfide nanoparticles were generalized. Formation
of the desired products was confirmed by carrying out model experiments.
amines promote nanoparticle formation at low temperatures.12,13
Recently, metal dialkyldithiocarbamates have been used as a
Introduction
Because of their unique properties and potential applications,
nanoparticles have been extensively studied by several research
groups. Thermal decomposition (i.e., thermolysis or pyrolysis)
of organometallic precursors is the representative method of
nanoparticle synthesis. Thermal decomposition allows easier and
simpler control of the shapes and sizes of the nanoparticles than
do other methods. In the past decade, many research groups
have investigated the synthesis and characterization of nano-
particles of various compositions, sizes, shapes, and properties.1-4
single-molecule source for the synthesis of semiconductor QDs
and rare-earth metal sulfide nanoparticles.8,9,14-16 Zinc dieth-
ylthiocarbamate [Zn(DETC)2] has also been used as a single-
molecule source for QD synthesis; in this method, ZnS shells
are grown around the surface of CdSe QDs in a microfluidic
system.17 The growth kinetics and shape evolution of the
nanoparticles synthesized by thermolysis have been actively
studied. However, the role of the amine in nanoparticle synthesis
and the exact mechanism of conversion of the precursor
molecules into intermediates and the subsequent decomposition
of the intermediates into nanoparticles have not been investigated
in detail.18-23
Because of their size-dependent optical and electrical proper-
ties, semiconductor quantum dots (QDs) have attracted the
attention of researchers.3 Since Bawendi and co-workers5
synthesized metal chalcogenide QDs (which first showed band-
edge emission at room temperature) by carrying out high-
temperature pyrolysis of the organometallic and chalcogenide
precursors in a coordinating solvent, extensive research on the
synthesis and characterization of QDs has been carried out. Peng
and Peng6 developed a new method of metal chalcogenide QD
synthesis using cadmium oxide and phosphonic acid, which
dramatically resolved a safety issue with respect to materials,
but high temperature was required for thermal decomposition.
Hence, the use of single-molecule precursors for the synthesis
of QDs was proposed,7-11 and has been reported that primary
Metal dialkyldithiocarbamates have been widely used as
active accelerators in the sulfur vulcanization of rubber24 and
(10) Cumberland, S. L.; Hanif, K. M.; Javier, A.; Khitrov, G. A.; Strouse,
G. F.; Woessner, S. M.; Yun, C. S. Chem. Mater. 2002, 14, 1576–
1584.
(11) Pradhan, N.; Efrima, S. J. Am. Chem. Soc. 2003, 125, 2050–2051.
(12) Pradhan, N.; Katz, B.; Efrima, S. J. Phys. Chem. B 2003, 107, 13843–
13854.
(13) Zhang, Z.; Lee, S. H.; Vittal, J. J.; Chin, W. S. J. Phys. Chem. B
2006, 110, 6649–6654.
(14) Trindade, T.; O’Brien, P. J. Mater. Chem. 1996, 6, 343–347.
(15) Lee, S.-M.; Jun, Y.-W.; Cho, S.-N.; Cheon, J. J. Am. Chem. Soc. 2002,
124, 11244–11245.
(16) Mirkovic, T.; Hines, M. A.; Nair, P. S.; Scholes, G. D. Chem. Mater.
2005, 17, 3451–3456.
(1) Cushing, B. L.; Kolesnichenko, V. L.; O’Connor, C. J. Chem. ReV.
2004, 104, 3893–3946.
(17) Wang, H.; Nakamura, H.; Uehara, M.; Yamaguchi, Y.; Miyazaki, M.;
Maeda, H. AdV. Funct. Mater. 2005, 15, 603–608.
(18) Peng, X.; Wickham, J.; Alivisatos, A. P. J. Am. Chem. Soc. 1998,
120, 5343–5344.
(2) Yin, Y.; Alivisatos, A. P. Nature 2005, 437, 664–670.
(3) Alivisatos, A. P. J. Phys. Chem. 1996, 100, 13226–13239.
(4) Lu, A.-H.; Salabas, E. L.; Schu¨th, F. Angew. Chem., Int. Ed. 2007,
46, 1222–1244.
(19) Manna, L.; Scher, E. C.; Alivisatos, A. P. J. Am. Chem. Soc. 2000,
122, 12700–12706.
(5) Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc. 1993,
115, 8706–8715.
(20) Peng, X.; Manna, L.; Yang, W.; Wickham, J.; Scher, E.; Kadavanich,
A.; Alivisatos, A. P. Nature 2000, 404, 59–61.
(21) Manna, L.; Milliron, D. J.; Meisel, A.; Scher, E. C.; Alivisatos, A. P.
Nat. Mater. 2003, 2, 382–385.
(6) Peng, Z. A.; Peng, X. J. Am. Chem. Soc. 2001, 123, 183–184.
(7) Trindade, T.; O’Brien, P. AdV. Mater. 1996, 8, 161–163.
(8) Trindade, T.; O’Brien, P.; Zhang, X.-m. Chem. Mater. 1997, 9, 523–
530.
(22) Kanaras, A. G.; Sonnichsen, C.; Liu, H.; Alivisatos, A. P. Nano Lett.
2005, 5, 2164–2167.
(9) Trindade, T.; OBrien, P.; Zhang, X.-m.; Motevalli, M. J. Mater. Chem.
1997, 7, 1011–1016.
(23) Kwon, S. G.; Hyeon, T. Acc. Chem. Res. 2008, 41, 1696–1709.
9
178 J. AM. CHEM. SOC. 2010, 132, 178–184
10.1021/ja905353a  2010 American Chemical Society

Thermal Decomposition Mechanism of Single-Molecule Precursors
A R T I C L E S
LC-MS and NMR spectroscopy after the solvent was removed in
vacuo. The solid precipitate was characterized by TEM and XRD.
In Situ Monitoring of Thermal Decomposition. A colorless
toluene solution containing Zn(DMTC)2 and an appropriate amount
of octylamine was heated at 50 °C and then cooled to room
temperature, after which the solution was centrifuged to separate
Zn(DMTC)2 · 2NH2Oct. The isolated Zn(DMTC)2 · 2NH2Oct was
dissolved in DMSO-d6 and analyzed by variable-temperature NMR
(VT-NMR) spectroscopy. Using the NMR heating system, the
solution was heated for 30 min from 60 to 100 °C in intervals of
10 °C to achieve thermal equilibrium.
Synthesis of Sodium Octyldithiocarbamate (NaS2CNHOct).
NaS2CNHOct was prepared using a slightly modified version of
the method reported in the literature.28 To the preheated aqueous
solution of 0.01 mol of octylamine and an equimolar amount of
NaOH at 100 °C, 25 mL of toluene containing 0.012 mol of carbon
disulfide was added dropwise with cooling and vigorous stirring.
The reaction mixture was stirred for 1 h, diluted with 100 mL of
petroleum ether, and then stirred for an additional 2 h. The mixture
was filtered on a sintered glass funnel, and the residue was washed
with petroleum ether and dried to obtain the pure of NaS2CNHOct
(1.78 g; 82.7% yield).
Synthesis of Zinc Octyldithiocarbamate [Zn(S2CNHOct)2].
Zn(S2CNHOct)2 was synthesized using the method reported in the
literature.8 Equimolar amounts of aqueous ZnCl2 and NaS2CNHOct
solutions were mixed and stirred for 10 min. The precipitate was
filtered, washed with a large amount of distilled water, and dried
in air to obtain Zn(S2CNHOct)2 (90.6% yield). 1H NMR (300 MHz,
DMSO-d6): δ 9.95 (br s, -NH-, 2H), 3.29 (q, -NHCH2-, 4H),
1.50 (m, -CH2CH2-, 4H), 1.25 (s, -C5H10-, 20H), 0.85 (t, -CH3,
6H).
Thermal Decomposition of Zn(S2CNHOct)2 in the Presence
of Secondary Amine. Zn(S2CNHOct)2 was mixed with 2 equiv of
dioctylamine in toluene, and the solution was heated to 100 °C.
The gaseous and solid products as well as the organic side products
were characterized as described previously.
as metal sulfide precursors in metal-organic chemical vapor
deposition (MOCVD) applications.25,26 This is because they are
moisture-insensitive, air-stable, less toxic, and easy to synthesize
and handle. Interestingly, Reedijk and co-workers27 have
proposed a mechanism for thermal decomposition of zinc
dimethyldithiocarbamate to explain the cross-linking of cis-1,4-
polybutadiene with a cyclic disulfide in the presence of zinc
dimethylthiocarbamate [Zn(DMTC)2] and a primary amine.
They carried out a reaction between Zn(DMTC)2 and hexy-
lamine to explain the generation of hydrogen sulfide (H2S),
which helps in the cross-linking of the unsaturated rubbers. They
observed the generation of dimethylamine (Me2NH), H2S, a solid
ZnS precipitate, and asymmetric and symmetric thiourea by the
thermolysis of Zn(DMTC)2 with hexylamine and proposed
several transition states generated by the interaction of coordi-
nated hexylamine with the DMTC ligands. However, they
focused on the generation of H2S and proposed a decomposition
mechanism on the basis of only the final decomposed products
obtained in sealed-tube reaction systems.
Here we propose a new mechanism for the decomposition
of Zn(DMTC)2 in the presence of alkylamines on the basis of
the results of studies conducted on the isolated intermediates
and the side products obtained at 1 atm. In this mechanism,
nucleophilic attack of the metal-coordinated amine on the most
electron-deficient thiocarbonyl carbon of the alkyldithiocarbam-
ate ligands at high temperature initiates the decomposition of
Zn(DMTC)2; steric hindrance in the alkylamine plays a very
important role in the thermolysis. This reaction mechanism could
be extended to other metal sulfide nanosystems and used as a
good basis for the synthesis of high-quality metal sulfide and
related nanoparticles.
Experimental Section
Chemicals and Instrumentation. Zn(DMTC)2 was purchased
from TCI. Zinc chloride (ZnCl2), lead nitrate [Pb(NO3)2], sodium
diethylthiocarbamate (NaDETC), octylamine (NH2Oct), oleic acid,
and the solvents (hexane, chloroform, toluene, and ethanol) were
purchased from Aldrich. NMR solvents [toluene-d8, dimethyl
sulfoxide-d6 (DMSO-d6), and chloroform-d1] were purchased from
CIL. All of the chemicals were used as received without further
purification. UV-vis spectra were recorded on a S-3100 spectro-
photometer (Scinco). Powder X-ray diffraction (XRD) measure-
ments were carried out on D8 Advance diffractometer (Bruker).
Transmission electron microscopy (TEM) and energy-dispersive
spectroscopy (EDS) were conducted using Hitachi-7600 (Hitachi)
and JEM-3000F (JEOL) instruments, respectively. NMR spectra
were recorded on NMR 500 (Varian) and DPX 300 (Bruker)
spectrometers. Liquid chromatography-mass spectrometry (LC-MS)
measurements were carried out using a Finnigan Surveyor MSQ
LC/MS Plus instrument (Thermo Electron Corporation).
Synthesis of Cadmium Diethyldithiocarbamate [Cd(DETC)2].
Cd(DETC)2 was synthesized using the method reported in the
literature.8 Equimolar amounts of aqueous CdCl2 and Na(DETC)
solutions were mixed and stirred for 10 min. The precipitate was
filtered, washed with a large amount of distilled water, and dried
1
in air to obtain Cd(DETC)2 (89.5% yield). H NMR (300 MHz,
DMSO-d6): δ 3.74 (m, -CH2-, 8H), 1.22 (s, -CH3, 12H).
Synthesis of Lead Diethyldithiocarbamate [Pb(DETC)2].
Pb(DETC)2 was prepared in a manner similar to that for the
preparation of Cd(DETC)2, which afforded the product in good yield
(89%). 1H NMR (300 MHz, DMSO-d6): δ 3.74 (m, -CH2-, 8H),
1.22 (s, -CH3, 12H).
Synthesis of CdS and PbS Nanoparticles. The appropriate
precursor complexes were mixed with 2 equiv of octylamine, and
the solution was heated at 100 °C. The obtained nanoparticles were
characterized in a manner similar to that described for the ZnS
nanoparticles.
Synthesis of Zinc Sulfide. Zn(DMTC)2 was used as a precursor
for the synthesis of ZnS nanoparticles. Zn(DMTC)2 was mixed with
an appropriate amount of octylamine in toluene, and the solution
was heated to 100 °C and maintained at this temperature. The gas
evolved during heating was trapped by an aqueous Pb2+ solution.
Next, the reaction mixture was cooled to room temperature and
then centrifuged to separate the organic side products and the solid
precipitate. The organic side products were characterized by
Results and Discussion
Zn(DMTC)2 powder was mixed with 1 equiv of octylamine
in toluene and heated at 50 °C to afford a homogeneous and
colorless solution, which was then cooled to room temperature
to obtain a white solid powder. The NMR spectrum of this solid
in DMSO-d6 showed representative peaks, including one at 2.74
ppm due to NH2-CH2C7H15 in the coordinated octylamine. The
peaks observed in the spectrum matched well with those reported
for the pentacoordinated Zn(DMTC)2 ·NH2Oct complex, whose
crystalline structure was also resolved by single-crystal X-ray
(24) Higgins, G. M. C.; Saville, B. J. Chem. Soc. 1963, 2812–2817.
(25) Frigo, D. M.; Khan, O. F. Z.; O’Brien, P. J. Cryst. Growth 1989, 96,
989–992.
(26) Pike, R. D.; Cui, H.; Kershaw, R.; Dwight, K.; Wold, A.; Blanton,
T. N.; Wernberg, A. A.; Gysling, H. J. Thin Solid Films 1993, 224,
221–226.
(27) Dirksen, A.; Nieuwenhuizen, P. J.; Hoogenraad, M.; Haasnoot, J. G.;
Reedijk, J. J. Appl. Polym. Sci. 2001, 79, 1074–1083.
(28) Sawant, P.; Kovalev, E.; Klug, J. T.; Efrima, S. Langmuir 2001, 17,
2913–2917.
9
J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010 179

A R T I C L E S
Jung et al.
Figure 1. In situ NMR monitoring of the thermal decomposition of hexacoordinated Zn(DMTC)2 ·2NH2Oct at (a) 50, (b) 60, (c) 70, (d) 80, (e) 90, and (f)
100 °C.
crystallography. When Zn(DMTC)2 was mixed with 2 equiv of
octylamine, a waxy solid was obtained and confirmed by NMR
analysis to be Zn(DMTC)2 ·2NH2Oct. The NMR spectrum of
this compound showed a peak due to NH2-CH2C7H15 at 2.66
ppm (Figure 1a); this δ value is between those for the free amine
(2.53 ppm) and the pentacoordinated complex. All efforts to
isolate the crystalline product to resolve the absolute configu-
ration of Zn(DMTC)2 · 2NH2Oct were unsuccessful; this was
probably a result of the weak interaction of the amines with
the Zn metal site and the mixed structure of possible isomers.
If more than 3 equiv of octylamine was mixed with Zn(DMTC)2
or an excess amount of octylamine was added to the solution
of Zn(DMTC)2 ·2NH2Oct, only one peak at ∼2.64 ppm, due to
the NH2-CH2C7H15 group bound to Zn, was observed. The peak
due to free NH2-CH2C7H15 was not observed. This confirmed
that octylamine is not tightly bound to Zn in the solution but
exists in a dynamic equilibrium state on NMR time scale. The
instability of Zn(DMTC)2 ·2NH2Oct can be explained on the
basis of this observation, which is supported by the fact that
the structures of hexacoordinated compounds have been reported
only in the case of chelating diamines such as bipyridine and
phenanthroline.29
New peaks appeared at 3.44 and 3.12 ppm at 60 °C and became
dominant at 100 °C (Figure 1b-f). For the analysis of the
structural changes in the intermediates, the isolated
Zn(DMTC)2 ·2NH2Oct was heated in toluene for 1 h at 100 °C.
Subsequently, the solution was cooled to room temperature to
obtain a white solid precipitate. The precipitate was dissolved
in DMSO, and the tiny amount of insoluble material was
removed by filtering through a 0.4 µm membrane filter; the
solvent was then evaporated in vacuo. The spectrum of the solid
residue showed peaks corresponding to NH2-CH2C7H15 and
-N(CH3)2 at 3.47 and 3.12 ppm, respectively, in DMSO-d6.
From the integration ratios, the composition of the residual solid
was confirmed to be a Zn(DMTC)2 unit plus 2 equiv of NH2Oct
(Figure 2c, which is very similar to Figure 1f). The simple NMR
spectrum suggested the formation of a symmetric intermediate
(intermediate II; Figure 2c, lower panel) that resulted from the
reaction between the two coordinated octylamines and the most
electron-deficient thiocarbonyl carbons of the alkyldithiocar-
bamate ligands. On the other hand, the NMR spectrum of the
solid isolated from the toluene solution of Zn(DMTC)2 ·
2NH2Oct heated at 80 °C (Figure 2b) showed two different
peaks corresponding to -N(CH3)2 at 3.12 and 3.41 ppm in
DMSO-d6, arising from an asymmetric structure (intermediate
I; Figure 2b, lower panel) inferred to contain an unreacted
DMTC ligand and a DMTC ligand that had reacted with
octylamine. The relative intensities of the two -N(CH3)2 peaks
changed with the heating time, and finally, the spectrum
resembled that of the symmetric structure isolated in the
The isolated waxy solid of Zn(DMTC)2 ·2NH2Oct was
dissolved in DMSO-d6, and the changes in its structure were
monitored using VT-NMR analysis. With an increase in
temperature, all of the peaks in the spectrum broadened slightly.
(29) Bell, N. A.; Johnson, E.; March, L. A.; Marsden, S. D.; Nowell, I. W.;
Walker, Y. Inorg. Chim. Acta 1989, 156, 205–211.
9
180 J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010

Thermal Decomposition Mechanism of Single-Molecule Precursors
A R T I C L E S
Figure 2. NMR spectra of Zn(DMTC)2 ·2NH2Oct after heating for 60 min at (a) 50, (b) 80, and (c) 100 °C.
experiment conducted at 100 °C. It is possible that the sample
obtained from the experiment conducted at 80 °C was a simple
mixture of hexacoordinated Zn(DMTC)2 · 2NH2Oct and fully
converted symmetric intermediate II obtained by the consecutive
migration of the two coordinated octylamines to the DMTC
ligands. However, it is speculated that the structural change
proceeds in a stepwise manner via the formation of the
asymmetric intermediate I, as suggested in Figure 2b. Two peaks
(at 207 and 182 ppm) corresponding to the thiocarbonyl carbon
were detected in the 13C NMR spectrum of intermediate I in
DMSO-d6. However, only one peak was detected in the 13C
NMR spectra of hexacoordinated Zn(DMTC)2 ·2NH2Oct (at 207
ppm) and fully converted symmetric intermediate II (at 182
ppm).
Single-crystal X-ray crystallography would have been useful
in resolving the absolute configurations of the obtained inter-
mediates. However, we were unable to prepare good-quality
crystalline solids for this purpose. This was probably because
intermediates I and II were structurally similar and unstable in
the solution during recrystallization, even at room temperature
for a long period of time.
Therefore, the spectrum cannot provide any information regard-
ing the nonprotonated thiocarbonyl carbon center, and unam-
biguous structural assignments cannot be made when the proton
resonances exactly overlap. On the other hand, the gHMBC
spectra help determine long-range or multiple-bond correlation,
i.e, the correlation between a given carbon and the protons on
the nearest two or three neighboring carbon atoms.
There are no cross-peaks in the gHSQC spectra of intermedi-
ates I and II because protons are not directly attached to the
thiocarbonyl carbon (data not shown). However, there are
several cross-peaks in the gHMBC (JC-H ) 8 Hz) spectra of
intermediates I and II. These are due to the long-range coupling
between the thiocarbonyl carbon and the protons of the
neighboring dimethylamine and octylamine moieties (Figure 3).
The spectrum of intermediate I, which was obtained at 80 °C,
shows a cross-peak due to the C1-H1 correlation between the
protons of the -N(CH3)2 group (3.40 ppm) and the thiocarbonyl
carbon (207 ppm). Another set of cross-peaks, one correspond-
ing to the C2-H2 correlation between the protons from
-N(CH3)2 (3.10 ppm) and the reacted thiocarbonyl carbon (182
ppm) and the other to the C2-H3 correlation between the
protons from -NH2CH2C7H15 (3.55 ppm) and the reacted
thiocarbonyl carbon, also appear (Figure 3a). However, the
spectrum of intermediate II, which was obtained at 100 °C,
shows only two cross peaks, C2-H2 and C2-H3, because of
its symmetric structure, which is attacked from both sides by
the octylamine nucleophiles (Figure 3b).
We used two-dimensional (2D) NMR inverse C-H correla-
tion techniques such as gradient heteronuclear single-quantum
n
correlation (gHSQC; JC-H with n ) 1) and gradient hetero-
nuclear multiple-bond correlation (gHMBC; nJC-H with n ) 2,
3) to observe the thiocarbonyl carbon peaks with respect to the
proton peaks in the spectra and thereby confirmed the structures
of intermediates I and II. The gHSQC spectrum is used to
determine the C-H connectivity involving direct C-H coupling.
These structures of intermediates I and II suggested from
the NMR experiments are generated by the nucleophilic attack
9
J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010 181

A R T I C L E S
Jung et al.
Figure 3. 2D NMR spectra from gHMBC experiments for (a) intermediate I and (b) intermediate II.
Scheme 1. Proposed Thermal Decomposition Mechanism of Zn(DMTC)2 in the Presence of Octylamine To Form ZnS Nanoparticles in Three
Steps: Intermediate Formation (Step 1), Decomposition To Release Thiourea (Step 2), and Condensation To Make the Zn-S-Zn Linkage
(Step 4); Step 3 Shows the Amine Exchange Reaction
of the lone-pair electrons of nitrogen (from the coordinated
octylamine) on the most electron-deficient thiocarbonyl carbon
of the Me2N-CS2 ligand in Zn(DMTC)2; this process is
described in Scheme 1. Zn(DMTC)2 has been known to form a
tetrahedral structure with D2d symmetry in the gas phase30 and
a dimeric structure in the crystal having a pseudotetrahedral
coordination sphere around the zinc atoms, with each zinc atom
bound to one chelating DMTC ligand and two bridging DMTC
ligands.31 In solutions containing coordinating ligands such as
amines and carboxylate ions, pentacoordinated complexes are
generated, as reported in the literature.24,29,32 The results of the
-
(31) Klug, H. P. Acta Crystallogr. 1966, 536–546.
(30) Hagen, K.; Holwill, C. J.; Rice, D. A. Inorg. Chem. 1989, 28, 3239–
(32) McCleverty, J. A.; Spencer, N.; Bailey, N. A.; Shackleton, S. L.
J. Chem. Soc., Dalton Trans. 1980, 1939–1944.
3242.
9
182 J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010

Thermal Decomposition Mechanism of Single-Molecule Precursors
A R T I C L E S
Figure 4. (a) TEM image, (b) XRD pattern, and (c) UV-vis absorption spectrum (in hexane) of the solid precipitate obtained by thermal decomposition
of Zn(DMTC)2 in the presence of octylamine. (d) TEM image of the solid precipitate generated when oleic acid was added as a capping ligand.
above-mentioned NMR analyses suggest that 2 equiv of
octylmaine binds to Zn(DMTC)2 to generate penta- and hexa-
coordinated complexes in solution. With an increase in tem-
perature, the coordinated octylamines migrate to the most
electron-deficient thiocarbonyl carbons of the dialkyldithiocar-
bamate ligands to generate the asymmetric intermediate I at 80
°C and the symmetric intermediate II at a higher temperature
of ∼100 °C (step 1 in Scheme 1). Through proton migration,
intermediate II is in equilibrium with a neutral form, which
generates HS-ZnLn by releasing N-octyl-N′,N′-dimethylthiourea
(step 2 in Scheme 1). Upon condensation of HS-ZnLn,
Zn-S-Zn linkages are formed, and H2S is released. Hence,
HS-ZnLn can be regarded as a building block of ZnS nano-
particles. ZnS nanoparticles are produced by the consecutive
decompositions of the amine-inserted dithiocarbamate ligands
accompanied by the condensation of HS-ZnLn (step 4 in
Scheme 1).
The colorless solution of Zn(DMTC)2 and 2 equiv of
octylamine turned yellow upon heating. This was accompanied
by gas evolution and the formation of a pale-yellow precipitate.
The organic side product in the supernatant was confirmed to
be pure N-octyl-N′,N′-dimethylthiourea by NMR spectroscopy
and fast-atom-bombardment MS (FAB-MS) (Figure S1 in the
Supporting Information). The evolved gas was trapped using
an aqueous Pb2+ solution to form a black precipitate of PbS;
this confirmed that the gas was H2S (Figure S2 in the Supporting
Information). By TEM, XRD, and UV-vis spectroscopy, the
pale-yellow precipitate obtained in the thermal reaction of
Zn(DMTC)2 with 2 equiv of octylamine was confirmed to be
ZnS nanoparticles (Figure 4). The average size of these ZnS
nanoparticles was on the order of a few nanometers. The ZnS
nanoparticles had a broad size distribution because they were
synthesized in the absence of a sufficient amount of capping
ligand and appropriate size control (Figure 4a). The thermal
reaction of Zn(DMTC)2 with 2 equiv of octylamine in the
presence of oleic acid, a capping ligand, afforded uniformly
sized ZnS nanoparticles (Figure 4d).
It is also possible that the protons of intermediate II migrate
to the -N(CH3)2 group instead of the coordinated sulfur, thereby
resulting in an amine exchange reaction (step 3 in Scheme 1).
This reaction is similar to the transamidation reaction of
carboxyamides with amines, which proceeds only in the
presence of Lewis acid catalysts such as organometallic
complexes of Al(III) or Ti(IV).33-35 A small amount of
exchanged dimethylamine, HN(CH3)2, was detected at ∼2.3 ppm
in the VT-NMR spectrum of Zn(DMTC)2 ·2NH2Oct (Figure 1),
and the generation of HN(CH3)2 was clearly detected in the
spectra of the solid intermediates at 2.3 ppm (Figure 2b,c). The
rate of thermal decomposition of Zn(DMTC)2 in the presence
of excess octylamine increased, as reported for common
pyrolysis reactions in the literature, and N,N′-dioctylthiourea
(symmetric thiourea) was obtained instead of N-octyl-N′,N′-
dimethylthiourea (asymmetric thiourea) as the organic side
product in the supernatant. In the presence of excess octylamine,
transamidation (step 3 in Scheme 1) afforded Zn(S2CNHOct)2
and/or its dimeric form, which could be coordinated with the
excess octylamine to generate a new hexacoordinated structure,
as shown below:
(33) Eldred, S. E.; Stone, D. A.; Gellman, S. H.; Stahl, S. S. J. Am. Chem.
Soc. 2003, 125, 3422–3423.
(34) Hoerter, J. M.; Otte, K. M.; Gellman, S. H.; Stahl, S. S. J. Am. Chem.
Soc. 2006, 128, 5177–5183.
(35) Kissounko, D. A.; Hoerter, J. M.; Guzei, I. A.; Cui, Q.; Gellman, S. H.;
Stahl, S. S. J. Am. Chem. Soc. 2007, 129, 1776–1783.
9
J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010 183

A R T I C L E S
Jung et al.
DETC was used to prepare lead and cadmium complexes, and their
decomposition reactions with octylamine at 100 °C were monitored.
As expected, asymmetric N-octyl-N′,N′-diethylthiourea was de-
tected in the supernatant, and CdS (10-15 nm) and PbS (200-500
nm) nanoparticles were obtained (Figures S6 and S7 in the
Supporting Information). The size of the obtained nanoparticles
appeared to be related to the relative stability of the precursor metal
complexes. Since the average sizes of Pb2+, Cd2+, and Zn2+ are
133, 109, and 88 pm, respectively, in their hexacoordinated
complexes,36 the stabilities of these complexes are expected to be
in the following order: Pb(DETC)2 < Cd(DETC)2 < Zn(DMTC)2.
The rate of the decomposition reaction was well-consistent with
the aforementioned trend.
The critical role of the amine in the thermolysis method used
for the synthesis of metal sulfide nanoparticles could be extended
to the synthesis of metal oxide (MOx) nanoparticles as well. A
study of the thermolysis of cadmium oleate to afford CdO
nanoparticles revealed that the decomposition temperature in the
presence of octylamine was ∼50 °C lower than that in the absence
of octylamine. The organic side product isolated from the super-
natant was found to be octyl oleylamide. Detailed studies must be
carried to confirm the decomposition mechanism. However, on the
basis of our proposed decomposition mechanism and in view of
the major role of amines in decomposition, we speculate that
thermal decomposition could be extended to many interesting
applications, such as the synthesis of nanomateials within a selected
area for patterning and generation of alloy nanoparticles through
consecutive thermolysis.
Consecutive decomposition of the hexacoordinated structure
(similar to step 2 in Scheme 1) yielded symmetric N,N′-
dioctylthiourea.
The average size of the ZnS nanoparticles generated from
the different reaction conditions using various amount of
octylamine clearly showed the effect of thermal decomposi-
tion kinetics. Even though the size distribution and the shape
regularity were not ideal for comparison with those for the
nanoparticles generated in the presence of the capping ligand
(oleic acid), the size of the generated ZnS nanoparticles
tended to decrease as the amount of octylamine increased.
The average sizes of the ZnS nanoparticles from the reactions
with 2 and 5 equiv of octylamine were measured as 6 ( 2
and 3 ( 1 nm, respectively (Figure S3 in the Supporting
Information). When 10 equiv of octylamine was used, the
shape of generated ZnS was changed to elongated sphere and
nanorod forms (average diameter of nanorod: 3 ( 1 nm,
length: 10 ( 4 nm). The absorption spectra of these ZnS
nanomaterials also showed that the absorption maxima shifted
to higher energy when a large amount of octylamine was
used to accelerate the decomposition reaction (Figure S4 in
the Supporting Information). These results could be explained
by the fact that the rate of thermal decomposition of
Zn(DMTC)2 was accelerated in the presence of the excess
amount of octylamine to generate a larger number of seeds
and produce smaller-sized nanoparticles, and the shape of
ZnS nanomaterials were not well-controlled in the absence
of a surface-stabilizing (capping) ligand such as oleic acid.
Thiourea, H2S, and ZnS nanoparticles were not obtained when
a secondary amine such as dioctylamine (HNOct2) was used instead
of the primary amine octylamine in the decomposition of
Zn(DMTC)2. In this case, only HN(CH3)2 was obtained,
and Zn(DMTC)2 was converted to Zn(S2CNOct2)2 through the
previously mentioned amine exchange reaction. This was probably
due to the steric hindrance around the thiocarbonyl carbon. This
steric hindrance could cause an unfavorable steric interaction when
a proton is in close proximity to the lone pair of the sulfur atom
(to prevent the decomposition represented by step 2 in Scheme 1)
while causing a favorable proton migration to the -N(CH3)2 group
(to enable the amine exchange reaction shown in step 3 in Scheme
1). More precise calculations and experiments are required to
confirm the proposed decomposition mechanism. However, the
results obtained from another control experiment supported this
hypothesis. A Zn complex having a primary amine unit on the
dithiocarbamate ligand, Zn(S2CNHOct)2, could react with dioctyl-
amine (a secondary amine) and decompose into asymmetric
N-octyl-N′,N′-dioctylthiourea, H2S, and ZnS nanoparticles (Figure
S5 in the Supporting Information).
Conclusions
We have investigated the mechanism underlying amine-
promoted thermal decomposition of single-molecule precursors,
namely, zinc alkyldithiocarbamate complexes, in order to under-
stand the conversion of the precursor molecules into intermediates,
their decomposition into nanoparticles, and the critical role of the
amine in the process. Nucleophilic attack of the coordinated amine
on the electron-deficient thiocarbonyl carbon of the alkyldithio-
carbamate ligand is the first step in the decomposition of zinc
alkyldithiocarbamate complexes. NMR and MS studies of the
organic side products generated from different alkylamines at
various concentrations have revealed that steric hindrance in the
alkylamine is very critical to the rate of decomposition of the zinc
alkyldithiocarbamate complexes and the competitive reaction
pathways involving the intermediates.
The proposed mechanism of the amine-promoted thermal
decomposition of zinc dithiocarbamate precursor complexes
could be extended to the synthesis of metal sulfide and metal
oxide nanoparticles from other metal precursor complexes;
further research must be carried out to confirm this mechanism.
Acknowledgment. This work was supported by a Korea Science
and Engineering Foundation (KOSEF) grant funded by the Korean
Government (MEST, R01-2008-000-20302-0). Y.K.J. and J.I.K.
gratefully acknowledge the BK21 Fellowship.
Supporting Information Available: Detailed characterization
data of metal sulfide nanoparticles and related side products
from the thermal decomposition reactions by NMR, MS, TEM,
and XRD. This material is available free of charge via the
Internet at http://pubs.acs.org.
The proposed mechanism of the amine-promoted thermal
decomposition reactions of zinc dithiocarbamate precursor com-
plexes can be extended to other metals such as lead and cadmium.
JA905353A
(36) Shannon, R. D. Acta Crystallogr. 1976, A32, 751–767.
9
184 J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010