Template assisted synthesis and assembly of nanoparticles 6

Chapter 6

Chapter 6

Template-assisted

assembly

of

Ag2S/CuXS (x = 1.75) nanoparticles

As introduced in Chapter 1, assembly as one of the most efficient methods is used
to order small particles on surfaces. Further growth of these ordered structures into
2D or 3D well-defined and sufficiently large colloidal structures have potential
application in photonics. The use of physical template to assemble colloidal particles
(e.g. SiO2, ZrO2) into aggregates with long-range order has proven to be a versatile
approach for the fabrication of more efficient light sources, detectors etc.1, 2 Generally,
this approach is called template-assisted assembly.
In template-assisted assembly process, a topographically patterned (formed by
assembly

of

polymer

beads/copolymer3,

photolithography,

electron

beam

lithography4 etc.) or chemically patterned surface (produced by flexible aliphatic
molecules as linking groups)5 is normally used as template. However, templates of
patterned topography offer more accurate positioning of particles compared with a
chemically patterned surface. They are used to create a well-defined spatial
distribution of forces that direct the motion of particles towards specific areas of the
substrate.
The use of wet colloidal self-assemblies as template to define the structure for
further nanoparticles assembly has been demonstrated.6,

7

Wang et al have co-

crystallized Au@SiO2 nanospheres together with PS latex spheres on quartz slides.8
They also investigated the relocalization of silica colloidal spheres using 2D patterned

193


Chapter 6
substrate as the templates through stepwise spin-coating technique.6 Kitaev et al
reported the formation of well-ordered self-assembled binary colloidal crystal (silica
& PS spheres) films in the scale of a few square centimeters, using microspheres with
a large disparity of sedimentation rates through accelerated evaporation induced coassembly.9
Another commonly used technique to produce templates is lithography such as
nanoimprint lithographic technique (NIL)10. Xia have demonstrated the capability of
template-assisted assembly in producing a rich variety of polygonal, polyhedral,

spiral11, and hybrid aggregate of spherical PS spheres or silica colloids on physical
template made by conventional microlithographic techniques.3, 12 The structure of the
assemblies could be conveniently controlled by simply changing the shape and
dimensions of the template.
Template-assisted assembly method typically combines physical templating and
capillary forces to assemble colloidal particles into uniform aggregates and structures.
The assembly of colloidal particles relies on the interaction between particle and/or
particles and surfaces to drive the formation of ordered arrangements. Depending on
the nature of the interaction between the particles themselves and the template surface,
adequate driving forces such as gravitational sedimentation by solvent evaporation13,
fluid flow14, electric field, or centrifugal force due to spinning15 are employed to
facilitate the assembly process.
Two templates were investigated in this chapter, namely the spontaneous selforganization of colloidal PS beads and PS line patterns generated via NIL technique.
These templates were employed to direct the assembly of semiconductor

194


Chapter 6
nanoparticles. In this work, we have studied two specific nanoparticles, i.e. faceted
Ag2S nanoparticles and CuxS nanodisks. Assembled Ag2S nanoparticles could be
used as optical filter, emitter while regular assembled nanodisks could give rise to
technologically useful properties, such as anisotropic electrical transport and optical
properties.

6.1 Assembly of Ag2S nanoparticles on pre-assembled
polystyrene beads
6.1.1 Estimation of the size of PS beads needed
Before using polystyrene (PS) beads as templates, calculations were carried out to
determine the size of PS beads needed for the assembly of specific sizes of

nanoparticles. Figure 6.1 illustrated the calculations and Table 6.1 gave the estimated
sizes of the cavities for certain diameter of the PS beads used in the template
Ag2S nanoparticles were prepared using our reported procedures detailed in
Sections 2.3.4 and 2.3.5. Faceted nanoparticles were prepared with an average size of
about 40-50 nm. Based on Table 6.1, the minimum size of PS beads that could be
used as template is 400 nm. Taking into consideration the lower estimation and also
availability of commercial PS beads in the laboratory, beads with average diameter of
1.053 µm were used as template in this study. Details about the assembly of PS
colloidal solution were described in Section 2.7.4. Some preliminary trials and
comparisons were made to decide which method and concentrations would be
optimum for the pre-assembled template of PS beads.

195


Chapter 6

rps

L

rss

D

rps

2
H
3

45°

H

L

2rss

(A)

(B)

Diameter of each PS bead = D
Radius of each PS bead, rps = D/2
Height of green triangle, H = D2 rps 2
Area of void between three PS beads, AV = 1 HD – 3( 1 rps2
2

2

( =

3

rad.)

Radius of small sphere, rss = (2/3) H - rps
Area of small sphere, AS = (rss)2
Width of cube, L = 2rss sin 45°

Area of cube, AC = L2
(C)
Figure 6.1. (A) Diagram illustrating the void in between three PS beads and a small
sphere which is in grey that can fit into the void; (B) Size of a cube that can fit into
the void can be estimated based on the size of the small sphere; (C) Calculations
steps to estimate the area of the void; and the maximum size of the small sphere,
AS, and cube, AC.
Table 6.1. Estimation of maximum sizes of the small sphere and cube for certain
diameter of PS beads. (Refer to Figure 6.1 for symbols)
D (nm)
H (nm)
AV (nm2)
rss (nm)
AS (nm2)
L (nm) AC (nm2)
100
87
403
8
188
11
120
200
173
1613
15
752
22
479
300

260
3628
23
1692
33
1077
400
346
6450
31
3007
44
1915
500
433
10078
39
4699
55
2992
600
520
14513
46
6767
66
4308
700
606
19754

54
9210
77
5863
800
693
25801
62
12030
88
7658
900
779
32654
70
15225
98
9693
1000
866
40314
77
18796
109
11966
1053
912
44700
81
20842

115
13268

196


Chapter 6
6.1.2 Assembly of Ag2S nanoparticles on PS beads pre-assembled
patterns
First, to study the interactions between Ag2S nanoparticles, the self-assembly of
Ag2S on bare silicon wafer was investigated by solvent evaporation or dipping &
interface method as detailed in Sections 2.7.2 and 2.7.3 respectively. Analysis under
SEM showed that Ag2S nanoparticles formed clusters instead of monolayers when
assembled by these two methods (Figure 6.2). It thus seems that strong interactions
existed among the Ag2S nanoparticles and resulted in aggregation. In the following,
we attempted to influence the assembly using template-assisted method, i.e. using
pre-assembled PS beads to define the location for nanoparticles aggregation.

a

b

Figure 6.2. SEM images showing direct assembly of Ag2S on silicon substrates
through: (a) solvent evaporation method, (b) dipping & interface method.
As reported by Kitaev9 and Kim7, regular PS beads pattern can be easily obtained
through convective vertical evaporation. Upon solvent evaporation, convective mass
flow and capillary forces cause the PS microspheres to assemble at the air-solventsubstrate interface to form a well-ordered hexagonal close-packed monolayer.
Subramanian16 also found that on slowly evaporating the water, the monodispersed

197

Chapter 6
PS particles will organize themselves in an ordered pattern due to a gradual increase
in their concentration. SEM images in Figure 6.3 showed that large area of regular
self-assembled PS beads pattern can be achieved through slow water evaporation
when the tilt angle of the Si substrate was set at 30°. The simplicity and the
reproducibility of this method were proven by many trials of assemblies. This regular
PS close-packed pattern has been used as template for the assembly of
nanoparticles.17, 18

a

b

Figure 6.3. SEM images showing the assembly of PS beads (1.053 µm) at different
concentrations through solvent evaporation method: (a) 0.26%, (b) 0.65%.

The surface of the close-packed PS microsphere presents two types of cavities
suitable for the arrangement of nanoparticles: interstitial sites between three adjoining
spheres and channels bridging these interstices. When this PS pre-assembled pattern
was dipped into Ag2S dispersion, the Ag2S nanoparticles supposed to pack into these
cavities. SEM images in Figure 6.4 showed clearly that the Ag2S nanoparticles had
aggregated into these cavities of the PS templates. Dipping & interface method seems
to give better assembly of Ag2S into these cavities although the area of assembly into
the cavities was not uniform.

198

Chapter 6

a

b

c

d

Figure 6.4. SEM images showing the assembly of Ag2S onto pre-assembled template
of PS beads by using (a, b) dipping & interface method, and (c, d) solvent evaporation
method.

Capillary forces due to surface tension between the nanoparticles and solvent have
been shown to play a major role for the ordering during assembly.13,

19

Capillary

forces allow the nanoparticles assembled together and accumulate into the cavities of
PS pattern. In the solvent evaporation method, the surface tension and convective
mass flow would act to pull the nanoparticles together. In the dipping & interface
method, the controlled movement of the template against the solvent provided the
pulling and capillary interaction. Nevertheless, it seems that the strength of the
surface tension between Ag2S nanoparticles and solvent was not strong enough to
overcome van der Waals interactions between the Ag2S nanoparticles, thus

199

Chapter 6
multilayers and aggregations of Ag2S particles were covering the PS beads pattern in
most area.

a

b

c

d

e

f

Figure 6.5. SEM images showing assembly of Ag2S nanoparticles on PS beads
template with different amount of nanoparticles: (a) 10 µL, (b) 15 µL, (c) 20 µL, (d)
30 µL, (e) 50 µL, and (f) 100 µL.

SEM images in Figure 6.5 showed the assembly patterns using varying amount of

200


Chapter 6
Ag2S nanoparticles. While larger areas of assembly would be expected using larger
amount of nanoparticles, the distribution of coverage is not even and the surface of

the PS beads was almost completely covered in some cases.
We have also attempted to remove the PS beads after assembly in order to expose
and examine the assembly of Ag2S nanoparticles clearly. Since PS beads were much
larger than the Ag2S nanoparticles, we achieved the removal by softly touching the
surface using a piece of adhesive tape. SEM images in Figure 6.6 confirmed that
most of the PS beads could be removed from the surface using this simple method. In
some regions of the assembly (marked with rectangular box in Figure 6.6), the
closed-packed pattern of the PS beads was clearly evident. However, some patterns
were removed together with PS beads (marked with circle), which indicated further
trial should be done to obtain optimized Ag2S pattern: sitting in the interstices of PS
pattern.

a

b

Figure 6.6. Ag2S pattern after the removal of PS beads from the template.

6.1.3 Assembly of CuxS (x = 1.75) nanodisks on the PS spheres pattern
It is known that the shape and morphology of nanoparticles have great effect on
their assembly behavior. Non-spherical nanoparticles show different types of self201


Chapter 6
assembly.20 For example, raft-like aggregates have been observed for nanorods
assembly.21-23 Thus, we have also investigated the assembly of copper sulfide (CuxS)
nanodisks on the PS beads pattern using the same method. CuxS nanoparticles with
regular disk shape (diameter ~ 100 nm; thickness ~ 15 nm) were prepared using hot
injection method developed in our laboratory (Section 2.3.5)24.
As shown in Figure 6.7(a), the size dispersity of the CuxS nanodisks was found to

be better than that of Ag2S nanoparticles. Since the average diameter of the disks is
slightly larger than the estimated cavity size (Table 6.1) of the ~1 µm PS beads, we
would expect these nanodisks can only assemble into the cavity by sitting on their
sides, which proved by SEM images shown in Figure 6.7. It is clear that although
some CuxS nanoparticles can enter into the interstices, most of them were randomly
dispersed on the top of PS pattern. After PS pattern was removed, there is no regular
patterned CuxS nanoparticles can be found. Thus, no further investigations were done
on its assembly.

a

b

Figure 6.7. SEM images showing the assembly of CuxS nanodisks on different
substrate: (a) 30 µL on bare Si substrate, and (b) 30 µL on PS pattern.

In conclusion, Ag2S nanoparticles can be assembled into the interstitial and
channel cavities of PS close-packed pattern while only partial CuxS nanodisks can.
202


Chapter 6
The arrangement of nanoparticles depends on space geometry as well as on the
nanoparticles shape. The pre-assembled PS beads template, however, was found to be
too soft and may be destroyed by the action of dipping or withdrawing from the
interface. In the following section, we prepare a harder template using nanoimprint
lithography (NIL) to further investigate the assembly of CuxS nanodisks.

6.2 Assembly of CuxS (x = 1.75) on PS line-pattern prepared
by Nanoimprint Lithography

PS line-patterns were fabricated through the combined use of NIL and ATRP as
presented in Section 2.7.5. The width of the line pattern, and hence the spacing
between channels, can be controlled by varying the time of ATRP reaction. As shown
in Figure 6.8, the channel spacing changed from 250 nm to 200 nm after ATRP
treatment for 5 hours. Although the depth of the channel may also be changed with
the ATRP process because the residual layer in the channel was not removed, this
dimension would not affect the assembly behavior of our nanodisks in this study.

a

b

Figure 6.8. SEM images showing (a) the original PS line-pattern with 250 nm
channel spacing prepared by NIL, and (b) pattern after ATRP treatment giving
channel spacing of 200 nm.

203


Chapter 6
The detailed assembly procedure was discussed in Section 2.7.5. The amount of
CuxS nanodisks was varied and a suitable concentration ~ 5 ×10-4 mol/L was chosen
for the best assembly results. All our attempts showed that after participating (of CuxS
nanodisks) step and washing (by solvent) step, CuxS nanodisks would accumulate
inside the channels of the PS line-pattern, rather than on top of the line pattern. The
assembly is believed to be similar to a sedimentation process combining with physical
restriction of PS channels and the arrangement of nanodisks inside the channels will
be influenced by physical or spatial constraint and surface property of PS channels.
First, we investigated the effect of channel spacing to the assembly of CuxS
nanodisks. SEM images in Figure 6.9 showed the assemblies of ~190 nm CuxS

nanodisks on PS line-pattern with 180 nm and 210 nm channel spacing. It is obvious
that when the size of the nanodisks is bigger than the channel spacing (Figure 6.9a),
the nanodisks were forced to stand on their sides. Whereas when the channel spacing
is bigger, the nanodisks sit on their faces inside the channels (Figure 6.9b).

a

b

Figure 6.9. Effect of channel spacing on the assembly of CuxS nanodisks (particle
size: ~ 190 nm): (a) channel spacing = 180 nm, and (b) channel spacing = 210 nm.

Next, we varied the size of CuxS nanodisks and investigated their assembly
behavior in the same channel spacing of 200 nm. From SEM images in Figure 6.10,
204


Chapter 6
it can be seen that while all the CuxS nanodisks can be assembled into the larger PS
channels, the assembled morphology become different depending on disk sizes. When
the diameter of the nanodisks is ~ 100 nm, which is much smaller than the channel
spacing, CuxS nanodisks were found to assemble randomly (Figure 6.10a). As the
diameter was increased to ~ 130 nm, a better assembled pattern was obtained with
most of the nanodisks sitting on their faces. When the diameter was increased to
comparable with the channel spacing ~ 200 nm, very regular assembly with nearly all
the CuxS nanodisks sitting on their faces was obtained.

a

b

c

Figure 6.10. SEM images showing assembly of CuxS nanodisks of different sizes into
PS line-pattern with 200 nm channel spacing: (a) 100 nm, (b) 130 nm, and (c) 200 nm.

These observations confirmed that the interplay between the sizes of nanodisks
and the channel spacing is a critical parameter to the formation of good assembly

205


Chapter 6
morphology. By carefully controlling this parameter, we believe it is possible to
regularly arrange CuxS nanodisks into PS channel in a well-ordered pattern.
In order to verify the surface effect of PS line-pattern to the assembly, PS linepattern without ATRP treatment was used as template for a trial assembly. The
corresponding SEM image in Figure 6.11 showed that scarce amount of CuxS
nanodisks were randomly dispersed in the channel. This suggested that indeed the
ATRP modification has assisted the assembly of CuxS nanodisks in the PS linepattern.

Figure 6.11. SEM image showing CuxS nanodisks assemble on PS line-pattern
without ATRP modification.

After precipitation, CuxS particles will locate both on the top and inside the PS
channels. The PS chains on the surface of PS channel are grown by ATRP. They are
much easier to swell into organic solvent (toluene) compared with cross-linked PS
chains. During washing step, the PS chains inside the channel will swell into solvent
(Gaussian chains) and wrap around CuxS nanodisks. These swollen chains block the
movement of CuxS nanodisks and thus hold these particles inside the channel. The PS
chains on top of the PS channels will push the CuxS nanodisks off the surface

accompanying with swelling into solvent. Thus CuxS nanodisks which located on the
206


Chapter 6
top of channels are washed away. As for the PS channels without ATRP treatment,
the cross-linked PS chains on the surface can only swollen a little and will not affect
the movement of CuxS nanodisks in solvent. In conclusion, ATRP treatment of PS
channels is prerequisite to pattern CuxS nanodisks.
Organic ligand is often added to assist or direct the assembly of nanoparticles into
regular assembled pattern. Thus, we attempted to add 20 µL DDT to 2 mL CuxS
dispersion before the assembly in the PS-line pattern. Because DDT is used as
capping agent in the preparation of CuxS nanodisks, thus precipitated CuxS were
washed thoroughly with ethanol before redispersed in toluene for assembly. SEM
images shown in Figure 6.12 presented a comparison between assemblies with and
without the addition of DDT. It thus seems that DDT will affect the density of CuxS
nanodisks inside PS channels, probably due to the interactions between DDT on CuxS
and swollen PS chains. Further investigations such as vary DDT to long chain amine
or dithiol molecules, are required to draw a conclusion if the morphology of the
assembly is also affected by organic ligands.

a

b

Figure 6.12. SEM images showing the effect of adding DDT to the assembly of CuxS
nanodisks (a) with DDT added, (b) without DDT.

207

Chapter 6

6.3 Conclusions
In summary, two kinds of templates have been employed to assist the assembly of
faceted Ag2S nanoparticles and CuxS nanodisks. The driving force for assembly is
capillary force and physical (spatial) constraint. Ag2S can be assembled into the
cavities of pre-assembled hexagonally closed-packed PS beads and the effect of
concentration on the assembled morphology was investigated and optimized. In the
second system, CuxS nanodisks can be assembled into the channels of PS line pattern.
The assembly of CuxS nanodisks was found to be mainly controlled by spatial
constraints (PS channels) and sizes of the nanodisks. CuxS nanodisks were found to
regularly pack inside the channel when the size of CuxS nanodisks fit well with the
channel spacing of the PS line pattern.

208


Chapter 6

References

1.

Xia, Y. N.; Gates, B.; Yin, Y. D.; Lu, Y. Advanced Materials 2000, 12, (10),
693-713.

2.

Arduini, M.; Rampazzo, E.; Mancin, F.; Tecilla, P.; Tonellato, U. Inorganica

Chimica Acta 2007, 360, (3), 721-727.

3.

Xia, Y. N.; Yin, Y. D.; Lu, Y.; McLellan, J. Advanced Functional Materials
2003, 13, (12), 907-918.

4.

Wang, D. Y.; Mohwald, H. Journal of Materials Chemistry 2004, 14, (4),
459-468.

5.

Chowdhury, D.; Maoz, R.; Sagiv, J. Nano Letters 2007, 7, (6), 1770-1778.

6.

Wang, D. Y.; Mohwald, H. Advanced Materials 2004, 16, (3), 244-247.

7.

Kim, M. H.; Im, S. H.; Park, O. O. Advanced Materials 2005, 17, (20), 25012505.

8.

Wang, D. Y.; Salgueirino-Maceira, V.; Liz-Marzan, L. W.; Caruso, F.
Advanced Materials 2002, 14, (12), 908-912.

9.

Kitaev, V.; Ozin, G. A. Advanced Materials 2003, 15, (1), 75-78.

10.

Low, H. Y. International Journal of Nanotechnology 2007, 4, (4), 389-403.

11.

Yin, Y. D.; Xia, Y. N. Journal of the American Chemical Society 2003, 125,
(8), 2048-2049.

12.

Yin, Y. D.; Lu, Y.; Gates, B.; Xia, Y. N. Journal of the American Chemical
Society 2001, 123, (36), 8718-8729.

209


Chapter 6
13.

Malaquin, L.; Kraus, T.; Schmid, H.; Delamarche, E.; Wolf, H. Langmuir
2007, 23, (23), 11513-11521.

14.

Yin, Y. D.; Lu, Y.; Xia, Y. N. Journal of the American Chemical Society 2001,
123, (4), 771-772.

15.

Varghese, B.; Cheong, F. C.; Sindhu, S.; Yu, T.; Lim, C. T.; Valiyaveettil, S.;
Sow, C. H. Langmuir 2006, 22, (19), 8248-8252.

16.

Subramanian, G.; Manoharan, V. N.; Thorne, J. D.; Pine, D. J. Advanced
Materials 1999, 11, (15), 1261-1265.

17.

Yin, Y. D.; Xia, Y. N. Advanced Materials 2001, 13, (4), 267-271.

18.

Zhong, Z. Y.; Yin, Y. D.; Gates, B.; Xia, Y. N. Advanced Materials 2000, 12,
(3), 206-209.

19.

Clark, T. D.; Ferrigno, R.; Tien, J.; Paul, K. E.; Whitesides, G. M. Journal of
the American Chemical Society 2002, 124, (19), 5419-5426.

20.

Jana, N. R. Angewandte Chemie-International Edition 2004, 43, (12), 15361540.

21.

Kim, F.; Kwan, S.; Akana, J.; Yang, P. D. Journal of the American Chemical
Society 2001, 123, (18), 4360-4361.

22.

Nikoobakht, B.; Wang, Z. L.; El-Sayed, M. A. Journal of Physical Chemistry
B 2000, 104, (36), 8635-8640.

23.

Carbone, L.; Nobile, C.; De Giorg, M.; Sala, F. D.; Morello, G.; Pompa, P.;
Hytch, M.; Snoeck, E.; Fiore, A.; Franchini, I. R.; Nadasan, M.; Silvestre, A.
F.; Chiodo, L.; Kudera, S.; Cingolani, R.; Krahne, R.; Manna, L. Nano Letters
2007, 7, (10), 2942-2950.

210


Chapter 6
24.

Lim, W. P.; Wong, C. T.; Ang, S. L.; Low, H. Y.; Chin, W. S. Chemistry of
Materials 2006, 18, (26), 6170-6177.

211