Controlled morphogenesis of organic polyhedral nanocrystals from cubes, cubooctahedrons, to octahedrons by manipulating the growth kinetics

Article abstract of DOI:10.1021/ja108730u

Morphological control of organic nanocrystals (ONCs) is important in the fields ranging from specialty chemicals to molecular semiconductors. Although the thermodynamic shape can be readily predicted, most growth morphologies of ONCs are actually determin

Full text of DOI:10.1021/ja108730u

ARTICLE
pubs.acs.org/JACS
Controlled Morphogenesis of Organic Polyhedral Nanocrystals
from Cubes, Cubooctahedrons, to Octahedrons by Manipulating
the Growth Kinetics
Longtian Kang,†,‡ Hongbing Fu,*,† Xinqiang Cao,†,‡ Qiang Shi,† and Jiannian Yao*,†
†Beijing National Laboratory for Molecular Sciences (BNLMS), Institute of Chemistry, Chinese Academy of Sciences,
Beijing 100190, People’s Republic of China
‡Graduate University of Chinese Academy of Sciences, Beijing 100039, People’s Republic of China
S Supporting Information
b
ABSTRACT: Morphological control of organic nanocrystals (ONCs) is important
in the fields ranging from specialty chemicals to molecular semiconductors. Although
the thermodynamic shape can be readily predicted, most growth morphologies of
ONCs are actually determined by kinetic factors and remain poorly understood. On
the basis of the reduction of zinc tetraphenylporphyrin perchlorate (ZnTPPþClO4
)
-
with sodium nitrite (NaþNO2-), we synthesized two series of ONCs of aquozinc
tetraphenylporphyrin (ZnTPP H2O), in the presence of either cetyltrimethylammo-
3
nium bromide (CTAB) or poly(vinyl pyrrolidone) (PVP) as the capping ligands. As
the cationic precursors of ZnTPPþ are separated in the solution phase, smoothly
controlled release of ZnTPP H2O building blocks via the reduction reaction facilitates the separation between the nucleation and growth
3
stages during the formation of ONCs and provides a high and tunable supersaturation unavailable by employing conventional crystallization
techniques. We found that CTAB mainly serve as the colloidal stabilizer, while selective adhesion of PVP on the {020}s facet alters the crystal
habits significantly. In both cases, manipulation of the growth kinetics had been achieved by adjusting the concentration of ZnTPP H2O
3
growth units, and consequently, the supersaturation for the crystallization, thus yielding ONCs with well-controlled sizes and shapes.
Remarkably, thermodynamically stable octahedrons have been obtained at high supersaturation in both CTAB and PVP cases.
’ INTRODUCTION
however, remains a great challenge.15 This is because most shapes of
ONCs are actually controlled by the kinetics of the molecular
growth process through which assembly occurs.16 Arresting kineti-
cally stable topology by controlling the nucleation and subsequent
growth kinetics has been applied as a successful strategy for
generating inorganic nanocrystals of various shapes.12b In particular,
dot-, rod-, spindle-, and tetrapod-shaped semiconductor quantum
dots have been synthesized by adjusting the monomer concentra-
tion using the sol method.17 And cube, cubooctahedra, and
octahedra of nobel metals have been fabricated by changing the
precursor ratio and introducing surfactants as the shape controller in
the polyol process.18 Nonetheless, the effects of kinetic control on
the growth morphologies of ONCs remain largely unexplored,
partially due to the limited tuning range available for conventional
solution crystallization process.
Controlled synthesis of inorganic semiconductor and/or metal
nanocrystals with well-defined sizes, shapes, and compositions had
provided a powerful tool for tailoring their properties1 and paved the
way for far-reaching applications ranging from optoelectronics,2
catalysis,3 plasmonics,4 to medical diagnostics.5 Organic molecular
crystals are fundamentally different from inorganic ones, because of
weak van der Waals intermolecular interactions.6 The growth
morphology of organic nanocrystals (ONCs) coupled with its size
are of great importance for industrial separation, purification, and
storage processes;7 moreover, they play a crucial role in developing
novel organic semiconductor technologies,8 such as single-crystal
nanolasers,9 transistor arrays,10 sensors, and photoconductors.11
However, although tailor-made molecules can generally be obtained
by organic synthesis, the controlled synthesis of ONCs lags far
behind their inorganic counterparts.1,12
The synthesis of organic molecules relies on the strength of
covalent bond connecting between atoms, whereas organic molec-
ular crystals are held together by a multitude of weak intermolecular
interactions, thus often referred to as “supermolecule(s) par excel-
lence”.13 The internal structure and symmetry of organic crystals
might be engineered by focusing on the molecular recognition
events between constituent molecular building blocks.13,14 Ensuring
that molecular components aggregate in a specific way and thus
generate ONCs with a desirable size, shape, and therefore function,
Herein, by employing the colloid chemical reaction method,19
two series of ONCs of aquozinc tetraphenylporphyrin (ZnTPP
3
H2O) with well-controlled sizes and shapes had been synthesized on
the basis of the reduction of zinc tetraphenylporphyrin perchlorate
(ZnTPPþClO4-) with sodium nitrite (NaþNO2-), in the pre-
sence of either cetyltrimethylammonium bromide (CTAB) and
poly(vinyl pyrrolidone) (PVP) as the capping ligand. As the cationic
Received: September 28, 2010
Published: January 18, 2011
r
2011 American Chemical Society
1895
dx.doi.org/10.1021/ja108730u J. Am. Chem. Soc. 2011, 133, 1895–1901
|

Journal of the American Chemical Society
ARTICLE
precursors are separated from each other in the solution phase,
smoothly controlled release of neutral molecular building blocks via
the reduction reaction facilitates the separation between the nuclea-
tion and growth stages during the formation of ONCs and provides
a high and tunable supersaturation unavailable by employing
conventional crystallization techniques. We found that the capping
ligands of CTAB mainly serve as a colloidal stabilizer, while selective
adhesion of PVP on the {020}s facet alters the crystal habits
significantly. In both cases, manipulation of the growth kinetics
had been achieved by adjusting the concentration of ZnTPP H2O
3
growth units, yielding ONCs with well-controlled sizes and shapes.
Remarkably, thermodynamically favorable octahedrons have been
obtained at high supersaturation in both CTAB and PVP cases.
’ RESULTS
In a typical synthesis, precursors of ZnTPPþClO4- were first
synthesized via reaction 1:20
2ZnTPP þ 2AgþClO4- þ I2 f
2ZnTPPþClO4- þ 2AgI ðsÞ
ð1Þ
-
Next, 1 mL of 1 mM ZnTPPþClO4 solution in anhydrous
acetonitrile was rapidly injected into different volumes (VNaNO2
)
Figure 1. Scanning electron microscopy (SEM) images of nanocubes
with an edge length of (A) 100 ( 5 nm prepared at CZnTPP = 0.143 mM
and (B) 203 ( 9 nm prepared at CZnTPP = 0.167 mM, in the presence of
CTAB surfactant. (C) Transmission electron microscopy (TEM) image
of 203 nm nanocubes. (E) SAED pattern taken from a single cube shown
in (D). In (E), the circled and triangle sets of spots with similar d spacing
values of 9.5 Å are due to {001}s and {201}s Bragg reflections, while
squared and pentagon sets of spots with similar d spacing values of 6.7 Å
are due to {200}s and {202}s Bragg reflections. (F) Schematic model for
of 10 mM NaþNO2- aqueous solution containing predissolved
CTAB or PVP as the capping ligands. Following the reduction of
ZnTPPþ by NO2- via reaction 2:
2ZnTPPþClO4- þ NaþNO2- þ 3H2O f
2ZnTPP H2OðsÞ þ NaþNO3- þ 2HþClO4-
ð2Þ
3
newly generated neutral ZnTPP H2O molecules undergo nu-
3
the cube. (G) Molecular packing arrangement of ZnTPP H2O in a cube,
3
cleation and growth, giving rise to ONCs. In our experiments, the
viewed almost perpendicular to the {202}s crystal plane.
reductant of NaþNO2 was stoichiometrically excessive, while
-
the amount of ZnTPPþClO4- was fixed at 1.0 mL of 1.0 mM.
pattern shows the single crystal structure of the cube (Figure 1D).
The circled and triangle sets of spots with similar d spacing values of
9.5 Å are due to {001}s (including {001} and {001}) and {201}s
(including {201} and {201}) Bragg reflections (d{001}s = 9.47 Å,
Therefore, the amount of ZnTPP H2O molecules produced via
3
reaction 2 is actually the same in all samples. Consequently, the
monomer concentration of ZnTPP H2O molecules (CZnTPP
)
3
can be easily adjusted as a function of VNaNO2, according to
d
{201}s = 9.45 Å, and —{001}/{201} = 90ꢀ); thereby, squared and
CZnTPP = 1.0 ÷ (1.0 þ VNaNO2) mM. We found that the sizes and
pentagon sets of spots with similar lattice spacing values of 6.7 Å are
shapes of ZnTPP H2O ONCs were sensitive to the value of
3
assigned to reflections from {200}s (including {200} and {200})
CZnTPP as well as the capping ligands of either CTAB or PVP
and {202}s (including {202} and {202}) crystal planes (d{200}s
=
presented in the system.
d{202}s = 6.7 Å and —{200}/{202} = 90ꢀ). Correlation of Bragg
CTAB as the Capping Ligand. Figures 1-4 depict the
reflections identified in Figure 1E with the orientation of the
cube shown in Figure 1D makes it clear that the cube is bound by
{200}s and {202}s facets on the side surfaces and by {020}s
(including {020} and {020}) facets on the top and bottom
surfaces (Figure 1F). Figure 1G presents the packing arrange-
morphology evolution pathway of ZnTPP H2O ONCs as a
3
function of CZnTPP, in the presence of CTAB (5 mM) surfactant. If
CZnTPP < 0.120 mM, no particles are detectable. At CZnTPP
=
0.120-0.125 mM, it reaches the nucleation threshold, generating
amorphous nanoparticles of 51 ( 4 nm in diameter (Figure S2). In
the range of CZnTPP = 0.143-0.167 mM, transition from amorphous
to crystalline structures leads to the formation of nanocubes with a
side-length controllable from 100 ( 5 nm at CZnTPP = 0.143 mM
(Figure 1A) to 203 ( 9 nm at CZnTPP = 0.167 mM (Figure 1B
and C). Because of the uniformity of their size and shape, these
nanocubes are able to self-assemble into ordered three-dimensional
arrays on the supporting substrate (Figures 1A-C). Figure 1E dis-
plays the selected area electron diffraction (SAED) pattern recorded
by directing the electron beam perpendicular to the flat square
ment of ZnTPP H2O molecules in the cube, viewed almost per-
3
pendicular to the {202}s crystal plane. As one can see, ZnTPP H2O
3
molecules arrange themselves with the molecular plane parallel to
{020}s crystal facet and perpendicular to {200}s and {202}s facets.
[Note that the molecular arrangements of ZnTPP H2O within
3
{200}s and {202}s crystal planes are almost the same (Figure S3).]
The fact that SAED patterns measured for different single cubes are
quite similar indicates that the {020}s faces of nanocubes are parallel
to the supporting substrate. As evidenced by X-ray diffraction (XRD)
measurement (Figure 2), this preferential orientation of nanocubes
on the substrate enhances the relative intensity of {020}s diffrac-
tion peak. The ratio between the intensities of the {020}sand{100}s
(or {111}s) diffraction peaks is 2.8 in the nanocube spectrum
surface of a single cube (Figure 1D). Monoclinic ZnTPP H2O crys-
3
tal (CCDC no.: ZNTPOR03) belongs to the space group of C2, with
cell parameters of a = 18.903(4) Å, b = 9.672(2) Å, c = 13.379(3) Å,
R = γ =90ꢀ, β = 134.922(3)ꢀ. The nearly square symmetry of SAED
1896
dx.doi.org/10.1021/ja108730u |J. Am. Chem. Soc. 2011, 133, 1895–1901

Journal of the American Chemical Society
ARTICLE
Figure 2. XRD patterns of cubes, truncated cubes, cubooctahedrons,
and octahedrons, prepared in the presence of CTAB surfactant. The top
line shows the standard powder spectrum simulated on the basis of the
single crystal data by using the DIAMOND software.
Figure 4. SEM images of octahedrons with controllable sizes prepared
at CZnTPP = (A) 0.250, (B) 0.500, and (C) 0.667 mM, respectively. The right
bottom insets in (A) and (B) are SAED patterns taken from single octahe-
drons shown in the right top insets. The circled and squared sets of spots are
due to {001}sand{201}s Bragg reflections. The inset in (C) shows the TEM
image of a single octahedron. (D) Schematic model for octahedron. (E)
Figure 3. SEM images of (A) truncated cubes (width, 195 ( 12 nm and
length, 195 ( 12 nm) prepared at CZnTPP = 0.182 mM, and (B) cubooc-
tahedrons (width, 135 ( 7 nm and length, 218 ( 18 nm) prepared at
Molecular packing arrangement of ZnTPP H2O in an octahedron.
3
diffraction peak is enhanced in the spectra of truncated cubes and
cubooctahedrons. By increasing CZnTPP further to 0.25 mM,
single crystal octahedrons with an edge-length of 246 ( 15 nm
are formed (Figure 4A). Further growth from the six corners of
an octahedron takes place as shown in Figure 4B at CZnTPP = 0.500
mM (Figure 4B), finally leading to larger octahedrons with an edge-
length of 417 ( 15 nm at CZnTPP = 0.667 mM (Figure 4C).
Combining with the results of cubooctahedrons, Figure 4D
shows a schematic representation for an octahedron. Note that
the dihedral angles measured for a octahedron in the inset of
Figure 4C are consistent with the crystallographic data, that is,
—{110}/{110} (or —{111}/{111})=71.8ꢀ, and —{110}/{110}
(or —{111}/{111})=108.2ꢀ (Figure 4E).
CZnTPP = 0.200 mM. (C) TEM image of cubooctahedrons. The right bottom
insets in (A) and (C) are SAED patterns taken from the single ONCs shown
in the right top insets. (D) Schematic model for the cuboctahedron.
(the bottom line in Figure 2), much higher than the standard value of
1.0 (the top line in Figure 2).
The evolution of nanocubes into truncated ones with a width
195 ( 12 nm and a length of 246 ( 15 nm is observed at
CZnTPP g 0.182 mM (Figure 3A). According to the SAED
pattern (the bottom inset of Figure 3A), the short and long axes
of truncated tube are along the [020]s and [200]s (or [202]s)
directions (the top inset of Figure 3A), respectively. The rapid
growth along the [020]s direction finally eliminates itself, result-
ing in single crystal cuboctahedron at CZnTPP = 0.200 mM, which
has a width of 135 ( 7 nm and a length of 218 ( 18 nm
(Figure 3B and C). The SAED patterns of truncated cube and
cuboctahedron, shown in the bottom insets of Figure 3A and C,
respectively, confirm that the eight truncated crystal facets are
{110}s and {111}s series of facets (Figure 3D). To keep the
balance and lower the position of center-of-gravity, both trun-
cated cubes and cubooctahedrons sit on the supporting substrate
with their {200}s or {202}s faces. Indeed, it can be seen from
Figure 2 that the relative intensity of the {200}s (or {202}s)
PVP as the Capping Ligand. As shown in Figure 5, we observed
a different morphology evolution pathway of ZnTPP H2O crystals
3
if PVP (fixed at 0.1 M in repeating units in all the reaction mixtures)
is selected as the capping ligand. Primary nanoparticles with a
diameter of 36 ( 4 nm are not formed until CZnTPP reaches
0.140-145 mM (Figure 5A), indicating a higher nucleation thresh-
old in PVP than that in CTAB (0.120-125 mM). At CZnTPP
=
0.167 mM, square nanoflakes are formed with a side-length of 342 (
15 nm and a thickness of 51 ( 5 nm (Figure 5B). Also confirmed by
TEM image (inset of Figure 5B), these square nanoflakes are able to
1897
dx.doi.org/10.1021/ja108730u |J. Am. Chem. Soc. 2011, 133, 1895–1901

Journal of the American Chemical Society
ARTICLE
Figure 6. XRD patterns of square flakes, planar-tetrapods, hexapods,
and octahedrons, prepared in the presence of PVP as the capping ligand.
The top line shows the standard powder spectrum simulated on the basis
of the single crystal data by using the DIAMOND software.
and planar-tetrapod present a semispherical core, as indicated by
TEM images in the bottom insets of Figure 5C and D. This
confirms that they are developed from particles. The side-length
and thickness of planar-tetrapods (Figure 5C) are similar to
those of square flakes (Figure 5D), whereas the crosses become
larger with a diagonal-length of 1994 ( 89 nm and a thickness of
398 ( 35 nm (Figure 5E and S6). The serious restriction of the
growth along the [020]s direction in flake, planar-tetrapod, and
cross shrinks those surfaces bounded by {hkl} facets with k = 0,
such as {200}s, {202}s, {001}s, and {201}s. Therefore, it can be
seen from Figure 6 that the {200}s (or {202}s) and {001}s (or
{201}s) XRD peaks, which are clearly resolved in the spectrum of
cubes (Figure 2), are actually absent from the spectra of flake,
planar-tetrapod, and cross, indicating these faces are less abun-
dant on the surfaces.
The top {020}s faces of flakes and planar-tetrapods appear
quite smooth, but grooves are present along diagonals of the cross
(Figures 5E and S6). By increasing CZnTPP to 0.333 mM (Figure 5F),
pairs of triangular sheets rise up in a steep wall from rough diagonal
grooves of top and bottom sides of the cross, probably directed by the
growth along [020]s directions as shown by the white circle in the top
inset of Figure 5E. The overall shape resembles a hexapod that
factually provides a backbone for an octahedron (top inset of
Figure 5F and Figure S7). Indeed, single crystal octahedrons are
formed at CZnTPP = 0.667 mM (Figure 5H) based on the hexapod via
a fill-in mechanism, as suggested by the quasi-octahedron obtained at
CZnTPP = 0.500 mM (Figures 5G and S8). Factually, the sizes of
hexapods, quasi-octahedrons, and octahedrons are similar with an
edge-length of 2420 ( 180 nm.
Figure 5. SEM images of ZnTPP H2O ONCs prepared at CZnTPP
=
3
(A) 0.143 (particles), (B) 0.167 (square flakes), (C) 0.200 (square
flakes), (D) 0.222 (planar tetrapods), (E) 0.250 (crosses), (F) 0.333
(hexapods), (G) 0.500 (growing octahedrons), and (H) 0.667 mM
(octahedrons), respectively, in the presence of PVP (fixed at 0.1 M in
repeating units in all the reaction mixtures) as the capping ligand. The
inset in (B) is the TEM image of 1D self-assembly of square nanoflakes.
The right top and bottom insets in (C)-(G) are the corresponding
high-magnification SEM and TEM images, respectively.
self-assemble into ordered 1D array rather than 3D self-assembly
observed for cubes. The SAED pattern reveals that square nanoflake
(Figure S4) is similar to nanocubes, also bounded by {020}s,
{200}s, and {202}s series of facets. However, the aspect ratio of the
former (6.7) is much higher than that of the latter (1.0), suggesting
that the growth along the [020]s direction is seriously restricted
in nanoflakes. The aspect ratio of square flakes adds up to 13 at
’ DISCUSSION
In our experiments, we changed the total volume of the reaction
mixture to adjust the monomer concentration of CZnTPP, and,
consequently, the supersaturation (σ), according to σ = CZnTPP
/
CZnTPP = 0.200 mM (Figure 5C); meanwhile, these nanoflakes with
Ceq, where CZnTPP and Ceq represent the actual concentration and
a side-length of 1131 ( 74 nm and a thickness of 87 ( 7 nm lose the
ability to 1D self-assemble (Figure 5C). These results demonstrate
the equilibrium concentration of ZnTPP H2O molecules in the
3
coexistence phase.21 By assuming Ceq(CTAB) = 0.120 mM and
Ceq(PVP) = 0.140 mM, Table 1 summarizes the morphology
clearly that the self-assembling behavior of ZnTPP H2O ONCs
3
strongly depends on its shape and size.
variations of ZnTPP H2O ONCs as a function of the value of σ
3
Under conditions of CZnTPP > 0.200 mM, preferential growth
along the [001]s and [201]s directions makes the shape of
in both CTAB and PVP cases.
Predicted Equilibrium and Growth Morphologies. To
gain further insight into the mechanistic basis, the equilibrium
ZnTPP H2O single crystals like a planar-tetrapod at CZnTPP
=
=
3
0.222 mM (Figures 5D and S5) and even like a cross at CZnTPP
shape of ZnTPP H2O crystal for minimum total surface energy
3
0.250 mM (Figures 5E and S6). More than 30% of square flakes
has been calculated by using the software of the Material Studio
1898
dx.doi.org/10.1021/ja108730u |J. Am. Chem. Soc. 2011, 133, 1895–1901

Journal of the American Chemical Society
ARTICLE
Table 1. Summary of Shapes of ZnTPP H2O ONCs as a Function of the Value of σ, Prepared in the Presence of Either CTAB or
3
PVP in the System
CTAB as the capping ligand
PVP as the caping ligand
regionb
VNaNO2 (mL)
CZnTPP (mM)
σa
regionb
shape
σa
shape
7.0
6.0
5.0
4.5
4.0
3.5
3.0
2.0
1.0
0.5
0.125
0.143
0.167
0.182
0.200
0.222
0.250
0.333
0.500
0.667
1.04
1.19
1.39
1.52
1.67
I
particle
II
cube
1.02
1.19
I
particle
II
cube
II
square plates
III
III
truncated cube
cuboctahedron
1.43
1.59
1.78
2.38
3.57
4.76
II
square plates
planar-tetrapod
cross
III
III
IV
IV
IV
2.08
2.78
III
III
octahedron
quasi-octahedron
hexapod
quasi-octahedron
5.56
III
octahedron
octahedron
a The supersaturation σ is calculated according to σ = CZnTPP/Ceq, and Ceq = 0.120, 0.40 mM for CTAB, PVP, respectively. b The roman numerals label
the region of the supersaturation σ, in which ONCs with a specific shape are generated.
attach
{hkl}
Table 2. Surface Free (γ{hkl}) and Attachment (E
)
Energies of Various Crystal Facets {hkl} Calculated by Using
the Material Studio Package
attach
{hkl}
{hkl}
d{hkl} (Å)
γ{hkl} (kcal/mol)
E
(kcal/mol)
{111}s
{110}s
{201}s
{001}s
{020}s
{200}s
{202}s
7.84
7.84
9.45
9.47
4.84
6.69
6.69
33.5
33.8
45.4
48.7
50.5
75.5
75.5
-72.1
-72.8
-86.8
-93.1
-95.5
-126.9
-127.0
package.21,22 The calculated surface energies (γ{hkl}) of the
various crystal faces {hkl} follow the order: γ{200}s ≈ γ{202}s
>
γ
{020}s > γ{001}s ≈ γ{201}s > γ{110}s ≈ γ{111}s (Table 2), yielding
an octahedron bound by low-energy {100}s and {111}s faces and
truncated on all the six corners (Figure 7A). This suggests that
octahedron-shaped morphology obtained at high supersaturation in
both CTAB and PVP cases is the thermodynamically favorable one.
The growth morphology, however, depends on the relative growth
rel
{hkl}
rates R
of the crystal faces {hkl}.16a,21 The widely adopted
attachment energy model relates the energy (Ea{thtkacl}h) released when
rel
{hkl}
two layers of the crystal structure {hkl} are brought together to R
according to R
,
Figure 7. (A) The equilibrium shape of ZnTPP H2O crystal for
rel
attach
3
= A ꢀ E
, where A is a proportional con-
{hkl}
{hkl}
minimum total surface energy and (B) the predicted growth morphol-
ogy based on the attachment energies, calculated by using the software of
Material Studio package.
stant.23 Figure 7B illustrates that the predicted growth morphology
based on the attachment energies (Table 2) is also an analogy of
octahedron. As compared to the equilibrium shape (Figure 7A),
{020}s faces are eliminated in the growth morphology besides the
truncation on {001}s and {201}s facets (Figure 7B). Among the
interplanar spacings d{hkl} in Table 2, d{020}s = 4.84 Å is the shortest.
Therefore, elimination of {020}s faces in the growth morphology
suggests a preferential growth along the [020]s directions. In any
event, the predicted growth morphology shown in Figure 7B is
according to Δμ = kBT ln(σ), where Δμ is the difference in
chemical potential between the growth units of ZnTPP H2O in
3
the crystal and the liquid phase, kB is Boltzmann’s constant, and T
is the absolute temperature (298 K).21 It must be noted that the
rel
{hkl}
treatment of R
based on the attachment energy model,
attach
{hkl}
according to Rr{ehlkl} = A ꢀ E
(A is a proportional constant),
does not include the driving force of Δμ/kBT.23 This might be
different from those shapes of ZnTPP H2O ONCs obtained at low
3
supersaturation in both CTAB (Figures 1-3) and PVP (Figure 5)
cases, which are actually bound by high-energy faces, such as {200}s,
{202}s, and {020}s faces.
responsible for the inconsistency of the predicted growth mor-
phology (Figure 7B) with the observed shapes of ZnTPP H2O
3
ONCs at low supersaturation in both CTAB and PVP cases.
As above-mentioned, we controlled the growth shapes of
Kinetic Control on the Growth of ZnTPP H2O ONCs. As
3
ZnTPP H2O ONCs by adjusting the monomer concentration
the crystallization is intrinsically a nonequilibrium process, the
crystal shape can be dominated by the kinetics of the growth
process.14,16 The most important mechanisms, in which the
3
of CZnTPP, therefore, the supersaturation σ (Table 1), which
represents the driving force for the crystallization (Δμ/kBT),
1899
dx.doi.org/10.1021/ja108730u |J. Am. Chem. Soc. 2011, 133, 1895–1901

Journal of the American Chemical Society
ARTICLE
supersaturation has been taken into account, are two-dimen-
nanoflakes, rather than nanocubes in the case of CTAB, are
formed in the region II of 1.40-1.50 g σ g 1.20 in the case of
PVP (Table 1), due to the restriction of the growth of {020}s faces.
sional (2D) nucleation and spiral growth models.16a,23b,24 For
rel
{hkl}
these two models, the relative growth rate R
of the crystal face
{hkl} follows:
This selective adhesion of PVP to {020}s faces even lowers γ{020}s
<
.
γ{001}s ≈ γ{201}s, thus raising ΔGq{020}s > ΔG{q 001}s ≈ ΔGq{201}s
rel
R
ꢁ C ꢀ expð- ΔGqfhklg=kBTÞ
ð3Þ
fhklg
Therefore, under conditions of region III of σ > 1.40-1.50 (see the
PVP part in Table 1), ΔGq{020}s > Δμ > ΔG{q 001}s ≈ ΔG{q 201}s results
in the rapid growth along the [001]s and [201]s directions. It is
where C is the concentration of the growth units (CZnTPP in our
system), and ΔGq{hkl} represents the activation free energy for the
rel
rel
rel
rel
growth of the crystal face {hkl}.16a,24 It can be seen that the ratio
R
{001}s
≈ R
> R
≈ R
that shapes ZnTPP H2O
{201}s
{200}s
{202}s
3
rel
{hkl}
ONCs like a planar-tetrapod or a cross. In the region IV of σ >
2.0-2.30 (see the PVP part in Table 1), the growth barrier of
{020}s is eventually overcome. The explosive growth along the
[020]s directions generates a hexapod, which provides a backbone
between the relative growth rates of R
depends on the kinetic
barriers ΔGq{hkl} for the growth of the crystals faces {hkl}. Generally,
the higher is the surface energy of γ{hkl}, the lower is the kinetic
barrier of ΔGq{hkl}. Therefore, the high-energy faces grow faster than
the low-energy faces in most cases.16a,23b,25 According to the cal-
culated order of γ{hkl} (Table 2), the value of ΔGq{hkl} might follow
for filling-in ZnTPP H2O units to form the thermodynamically
3
favorable octahedron.
the order: ΔGq{200}s ≈ ΔGq{202}s ≈ ΔGq{020}s ≈ ΔG{q001}s
≈
ΔG{q 201}s ≈ ΔGq{110}s ≈ ΔG{q 111}s
.
’ CONCLUSIONS
In our wet chemical method, the cationic precursors of
By employing the colloid chemical reaction method, two series
ZnTPPþ are separated in the solution phase. The growth units
of ONCs of ZnTPP H2O with well-controlled sizes and shapes had
3
-
been synthesized on the basis of the reduction of ZnTPPþClO4
of neutral ZnTPP H2O molecules are gradually released via the
3
with NaþNO2-, in the presence of either CTAB or PVP as the
capping ligand. As the cationic precursors of ZnTPPþ are separated
from each other in the solution phase, smoothly controlled release of
reduction reaction 2. Once σ g 1.0 (Δμ/kBT g 0) reaches the
nucleation threshold, a nucleation burst generates primary
particles.26 More and more growth units are then fed by the
reduction of the remaining precursors. As long as the consump-
tion of feedstock by the growing ONCs is not exceeded by the
release rate of neutral growth units via the reduction reaction, no
new nuclei form. Therefore, our colloidal chemical method
facilitates the separation between the nucleation and growth
ZnTPP H2O building blocks via the reduction reaction facilitates
3
the separation between the nucleation and growth stages of ONCs
and provides a high and tunable supersaturation unavailable in
conventional crystallization techniques. We found the capping ligand
of CTAB mainly serves as a colloidal stabilizer, while selective
adhesion of PVP on the {020} facet significantly alters the crystal
habits. In both cases, manipulation of the growth kinetics had been
stages during the formation of ZnTPP H2O ONCs.19 Moreover,
3
convenient adjustment of the precursor concentration provides a
tunable supersaturation for manipulating the nucleation and
growth kinetics. We indentified a critical value of σ ≈ 1.20 in
both CTAB and PVP cases (Table 1). If σ < 1.20 (region I for
both CTAB and PVP parts in Table 1), particles are always
obtained, but amorphous in nature probably because the chemi-
cal potential of the system (Δμ) is too low to overcome any
of ΔGq{hkl}. If σ g 1.20 (region II for both CTAB and PVP parts in
Table 1), the growth of {200}s and {202}s faces takes place as a
achieved by adjusting the concentration of ZnTPP H2O growth
3
units, and, consequently, the supersaturation for the crystallization,
thus yielding ONCs with well-controlled sizes and shapes. None-
theless, thermodynamically favored octahedrons have been obtained
at high supersaturation in both CTAB and PVP cases.
’ ASSOCIATED CONTENT
result of Δμ g ΔGq{200}s ≈ ΔG{q 202}s
.
S
Supporting Information. Precursor preparation, experi-
b
In the case of CTAB (see the CTAB part in Table 1),27 cube-
shaped ONCs are formed in the region II of 1.40-1.50 g σ g
1.20. However, in the region III of σ > 1.40-1.50 (Table 1), the
chemical potential of the system (Δμ) is high enough to over-
mental procedures, characterization techniques, Material Studio
calculations, and additional data and results. This material is
available free of charge via the Internet at http://pubs.acs.org.
come the barrier of ΔGq{020}s, making Rr{e0l20}s > Rr{e2l00}s ≈ Rrel
.
’ AUTHOR INFORMATION
{202}s
Accordingly, the rapid growth of {020}s faces results in truncated
cubes and cubooctahedrons. Once {020}s faces grow out of
existence, remained growth of {200}s and {202}s faces at high
supersaturation finally eliminates itself, leaving octahedrons
bounded by low-energy {100}s and {111}s facets (see the CTAB
part in Table 1).
Corresponding Author
hongbing.fu@iccas.ac.cn; jnyao@iccas.ac.cn
’ ACKNOWLEDGMENT
This work was supported by the National Natural Science
Foundation of China (nos. 20873163, 20373077, 20925309),
the Chinese Academy of Sciences (“100 Talents” program),
and the National Basic Research Program of China (973)
2011CB808402.
The situation is quite different in the case of PVP, because PVP
molecules can interact strongly with ZnTPP molecules through
hydrogen bonding.28 It is the solubilization effect induced by
PVP that results in a higher nucleation threshold concentration
in the case of PVP (Ceq = 0.140 mM) than that in the case of
CTAB (Ceq = 0.120 mM) (Table 1). Furthermore, PVP mole-
cules might selectively adhere to {020}s faces through hydrogen
bonding between pyrrolidone groups of PVP and apical H2O
’ REFERENCES
(1) Burda, C.; Chen, X. B.; Narayanan, R.; El-Sayed, M. A. Chem.
Rev. 2005, 105, 1025.
(2) (a) Huynh, W. U.; Dittmer, J. J.; Alivisatos, A. P. Science 2002,
295, 245. (b) Sun, Q.; Wang, Y. A.; Li, L. S.; Wang, D. Y.; Zhu, T.; Xu, J.;
Yang, C. H.; Li, Y. F. Nat. Photon. 2007, 1, 717.
moieties of ZnTPP H2O exposed on {020}s faces. The selective
3
adhesion of PVP to {020}s faces can lower the γ{020}s, thus
{020}s
raisingΔG{q020}s, leading toadecrease ofRrel .29 Indeed, square
1900
dx.doi.org/10.1021/ja108730u |J. Am. Chem. Soc. 2011, 133, 1895–1901

Journal of the American Chemical Society
ARTICLE
(3) (a) Bell, A. T. Science 2003, 299, 1688. (b) Niu, Y. H.; Yeung,
L. K.; Crooks, R. M. J. Am. Chem. Soc. 2001, 123, 6840.
(4) (a) Kelly, K. L.; Coronado, E.; Zhao, L. L.; Schatz, G. C. J. Phys.
Chem. B 2003, 107, 668. (b) Jin, R. C.; Cao, Y. W.; Mirkin, C. A.; Kelly,
K. L.; Schatz, G. C.; Zheng, J. G. Science 2001, 294, 1901.
(5) (a) Huang, X. H.; El-Sayed, I. H.; Qian, W.; El-Sayed, M. A.
J. Am. Chem. Soc. 2006, 128, 2115. (b) Jiang, W.; Kim, B. Y. S.; Rutka,
J. T.; Chan, W. C. W. Nat. Nanotechnol. 2008, 3, 145.
(6) (a) Silinsh, E. A. Organic Molecular Crystals: Their Electronic
States; Springer-Verlag: Berlin, 1980. (b) Pope, M.; Swenberg, C. E.
Electronic Processes in Organic Crystals and Polymers; Oxford University
Press: Oxford, 1999.
(21) Sunagawa, I. Crystals: Growth, Morphology and Perfection; Cam-
bridge University Press: Cambridge, 2005.
(22) Mann, S. Angew. Chem., Int. Ed. 2000, 39, 3392.
(23) (a) Bennema, P.; van der Eerden, J. P. Morphology of Crystals;
Terra Scientific (TERRPUB): Tokyo, 1987; pp 1-75. (b) Boerrigter,
S. X. M.; Cuppen, H. M.; Ristic, R. I.; Sherwood, J. N.; Bennema, P.;
Meekes, H. Cryst. Growth Des. 2002, 2, 357.
(24) Mullin, J. W. Crystallization, 3rd ed.; Butterworth Heinemann:
Oxford, 1992; Chapter 6, pp 202-260.
(25) Lee, S. M.; Cho, S. N.; Cheon, J. W. Adv. Funct. Mater. 2003, 15,
441.
(26) Lamer, V. K.; Dinegar, R. H. J. Am. Chem. Soc. 1950, 72, 4847.
(27) It has been established that as-formed particles can be stabilized
by the formation of a protective layer of CTAB through hydrophobic
interactions (ref 19), which are non-selective with regard to different
crystal faces. Moreover, CTAB surfactant molecules that adhere to
ONCs’ surface are dynamically exchangeable, allowing the growth units
on (growth) and off (dissolution) the growing ONCs. See the reference:
Lei, Y. L.; Liao, Q.; Fu, H. B.; Yao, J. N. J. Am. Chem. Soc. 2010, 132, 1742.
(28) Viseu, T. M. R.; Hungerford, G.; Coelho, A. F.; Ferreira, M. I. C.
J. Phys. Chem. B 2003, 107, 13300.
(7) (a) Rodriguez-Hornedo, N.; Murphy, D. J. Pharm. Sci. 1999, 88,
651. (b) Docherty, R.; Clydesdale, G.; Roberts, K. J.; Bennema, P. J. Phys.
D: Appl. Phys. 1991, 24, 89.
(8) (a) Sundar, V. C.; Zaumseil, J.; Podzorov, V.; Menard, E.; Willett,
R. L.; Someya, T.; Gershenson, M. E.; Rogers, J. A. Science 2004, 303,
1644. (b) Liu, S. H.; Wang, W. C. M.; Briseno, A. L.; Mannsfeld, S. C. B.;
Bao, Z. N. Adv. Mater. 2009, 21, 1217. (c) Nakanishi, H.; Oikawa, H. In
Single Organic Nanoparticles; Masuhara, H., Nakanishi, H.; Sasaki, K.,
Eds.; Springer-Verlag: New York, 2003; Chapter 2, pp 17-31. (d) Zang,
L.; Che, Y. K.; Moore, J. S. Acc. Chem. Res. 2008, 41, 1596. (e) Zhao,
Y. S.; Fu, H. B.; Peng, A. D.; Ma, Y.; Liao, Q.; Yao, J. N. Acc. Chem. Res.
2010, 43, 109. (f) Li, I. J.; Hu, W. P.; Liu, Y. Q.; Zhu, D. B. Acc. Chem. Res.
2010, 43, 529. (g) Zhang, X. J.; Dong, C.; Zapien, J. A.; Ismathullakhan,
S.; Kang, Z. H.; Jie, J. S.; Zhang, X. H.; Chang, J. C.; Lee, C. S.; Lee, S. T.
Angew. Chem., Int. Ed. 2009, 48, 9121.
(29) Sun, Y. G.; Xia, Y. N. Science 2002, 298, 2176.
(9) (a) Samuel, I. D. W.; Turnbull, G. A. Chem. Rev. 2007, 107, 1272.
(b) Zhao, Y. S.; Peng, A. D.; Fu, H. B.; Ma, Y.; Yao, J. N. Adv. Mater.
2008, 20, 1661. (c) O’Carroll, D.; Lieberwirth, I.; Redmond, G. Nat.
Nanotechnol. 2007, 2, 180.
(10) (a) Briseno, A. L.; Mannsfeld, S. C. B.; Lin, M. M.; Liu, S. H.;
Tseng, R. J.; Reese, C.; Roberts, M. E.; Yang, Y.; Wudl, F.; Bao, Z. N.
Nature 2006, 444, 913. (b) Nelson, S. F.; Lin, Y. Y.; Gundlach, D. J.;
Jackson, T. N. Appl. Phys. Lett. 1998, 72, 1854. (c) Luo, L.; Liu, G.;
Huang, L. W.; Cao, X. Q.; Liu, M.; Fu, H. B.; Yao, J. N. Appl. Phys. Lett.
2009, 95, 263312.
(11) (a) Che, Y. K.; Yang, X. M.; Lin, G. L.; Yu, C.; Ji, H. W.; Zuo,
J. M.; Zhao, J. C.; Zang, L. J. Am. Chem. Soc. 2010, 132, 5743. (b) Zhao,
Y. S.; Wu, J. S.; Huang, J. X. J. Am. Chem. Soc. 2009, 131, 3158.
(12) (a) Murray, C. B.; Kagan, C. R.; Bawendi, M. G. Annu. Rev.
Mater. Sci. 2000, 30, 545. (b) Yin, Y. D.; Alivisatos, A. P. Nature 2005,
437, 664.
(13) (a) Moulton, B.; Zaworotko, M. J. Chem. Rev. 2001, 101, 1629.
(b) Weissbuch, I.; Addadi, L.; Lahav, M.; Leiserowitz, L. Science 1991,
253, 637. (c) Zaworotko, M. J. Cryst. Growth Des. 2007, 7, 4.
(14) (a) Hollingsworth, M. D. Science 2002, 295, 2410. (b) Desiraju,
G. R. Nat. Mater. 2002, 1, 77. (c) Braga, D.; Brammer, L.; Champness,
N. R. CrystEngComm 2005, 7, 1.
(15) (a) Jonkheijm, P.; van der Schoot, P.; Schenning, A. P. H. J.;
Meijer, E. W. Science 2006, 313, 80. (b) Wang, Y. B.; Fu, H. B.; Peng,
A. D.; Zhao, Y. S.; Ma, J. S.; Ma, Y.; Yao, J. N. Chem. Commun. 2007,
1623.
(16) (a) Liu, X. Y.; Boek, E. S.; Briels, W. J.; Bennema, P. Nature
1995, 374, 342. (b) Huang, L. W.; Liao, Q.; Shi, Q.; Fu, H. B.; Ma, J. S.;
Yao, J. N. J. Mater. Chem. 2010, 20, 159. (c) Lei, Y. L.; Liao, Q.; Fu, H. B.;
Yao, J. N. J. Phys. Chem. C 2009, 113, 10038.
(17) (a) Peng, Z. A.; Peng, X. G. J. Am. Chem. Soc. 2001, 123, 1389.
(b) Peng, Z. A.; Peng, X. G. J. Am. Chem. Soc. 2002, 124, 3343. (c)
Manna, L.; Scher, E. C.; Alivisatos, A. P. J. Am. Chem. Soc. 2000, 122,
12700.
(18) (a) Tao, A.; Sinsermsuksakul, P.; Yang, P. D. Adv. Mater. 2006,
45, 4597. (b) Tsung, C. K.; Kuhn, J. N.; Huang, W. Y.; Aliaga, C.; Huang,
L. I.; Somorjai, G. A.; Yang, P. D. J. Am. Chem. Soc. 2009, 131, 5816.
(19) Kang, L. T.; Wang, Z. C.; Cao, Z. W.; Ma, Y.; Fu, H. B.; Yao,
J. N. J. Am. Chem. Soc. 2007, 129, 7305.
(20) Shine, H. J.; Padilla, A. G.; Wu, S. M. J. Org. Chem. 1979,
44, 4069.
1901
dx.doi.org/10.1021/ja108730u |J. Am. Chem. Soc. 2011, 133, 1895–1901