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 nanosph
Trang 1Chapter 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
Trang 2substrate 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 co-assembly.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 self-organization of colloidal PS beads and PS line patterns generated via NIL technique These templates were employed to direct the assembly of semiconductor
Trang 3nanoparticles 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
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
Trang 4Table 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) A V (nm2) r ss(nm) A S (nm2) L (nm) A C (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
Diameter of each PS bead = D
Radius of each PS bead, r ps = D/2
Height of green triangle, H = 2 2
ps r
D −
Area of void between three PS beads, A V =
2
1H D – 3(
2
1r ps 2 θ) (θ =
3 rad.)
Radius of small sphere,r ss = (2/3) H - r ps
Area of small sphere, AS = π( r ss)2
Width of cube, L = 2r ss sin 45°
Area of cube, A C = L2
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
(A)
L
2 r ss
45°
L
r ps
(B)
H
θ
D
r ps
r ss
3
2
H
(C)
Trang 56.1.2 Assembly of Ag 2 S 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
bly of Ag2S on silicon substrates thod, (b) dipping & interface method
, regular PS beads pattern can be easily obtained pon solvent evaporation, convective mass crospheres to assemble at the
air-solvent-ed hexagonal close-packair-solvent-ed monolayer
b
a
Figure 6.2 SEM images showing direct assem
through: (a) solvent evaporation me
As reported by Kitaev9 and Kim7
through convective vertical evaporation U
flow and capillary forces cause the PS mi
substrate interface to form a well-order
ramanian16 also found that on slowly evaporating the water, the monodispers
Trang 6PS particles will organize themselves in an ordered pattern due to a gradual increase
in
bly of PS beads (1.053 µm) at different ethod: (a) 0.26%, (b) 0.65%
icrosphere presents two types of cavities interstitial sites between three adjoining tices When this PS pre-assembled pattern
S 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
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
Figure 6.3 SEM images showing the assem
concentrations through solvent evaporation m
The surface of the close-packed PS m
suitable for the arrangement of nanoparticles:
spheres and channels bridging these inters
was dipped into Ag2S dispersion, the Ag2
b
a
Trang 7bly.13, 19 Capillary
s 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
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 assem
force
d
c
Trang 8multilayers and aggregations of Ag2
most area
S particles were covering the PS beads pattern in
Figure 6.5 SEM images showing assembly of Ag S nanoparticles on PS beads
te
30 µL, (e) 50 µL, and (f) 100 µL
2 mplate with different amount of nanoparticles: (a) 10 µL, (b) 15 µL, (c) 20 µL, (d)
SEM images in Figure 6.5 showed the assembly patterns using varying amount of
f
d
b
a
c
e
Trang 9Ag2S 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
arked with rectangular box in Figure 6.6), the
ident However, some patterns arked with circle), which indicated further
2S pattern: sitting in the interstices of PS
moval of PS beads from the template
It is known that the shape and morphology of nanoparticles have great effect on their assembly behavior Non-spherical nanoparticles show different types of
self-some regions of the assembly (m
closed-packed pattern of the PS beads was clearly ev
were removed together with PS beads (m
trial should be done to obtain optimized Ag
pattern
Figure 6.6 Ag2S pattern after the re
b
a
Trang 10assembly.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
e diameter of the disks is
e (Table 6.1) of the ~1 µm PS beads, we
mble into the cavity by sitting on their
Figure 6.7 It is clear that although
rstices, most of them were randomly
PS pattern was removed, there is no regular patterned CuxS nanoparticles can be found Thus, no further investigations were done
on its assembly
bly of CuxS nanodisks on different trate, and (b) 30 µL on PS pattern
bled into the interstitial and channel cavities of PS close-packed pattern while only partial Cu S nanodisks can
be better than that of Ag2S nanoparticles Since the averag
slightly larger than the estimated cavity siz
would expect these nanodisks can only asse
sides, which proved by SEM images shown in
some CuxS nanoparticles can enter into the inte
dispersed on the top of PS pattern After
Figure 6.7 SEM images showing the assem
substrate: (a) 30 µL on bare Si subs
In conclusion, Ag2S nanoparticles can be assem
Trang 11Th he nanoparticles shape The p owever, 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.
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
treatment for 5 hours Although the depth of the channel may also be changed with
dimension would not affect the assembly behavior of our nanodisks in this study
inal PS line-pattern with 250 nm ) pattern after ATRP treatment giving channel spacing of 200 nm
e arrangement of nanoparticles depends on space geometry as well as on t
re-assembled PS beads template, h
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
in Figure 6.8, the channel spacing changed from 250 nm to 200 nm after ATRP
the ATRP process because the residual layer in the channel was not removed, this
Figure 6.8 SEM images showing (a) the orig
channel spacing prepared by NIL, and (b
b
a
Trang 12The detailed assembly procedure was di
CuxS nanodisks was varied and a su
for the best assembly results All our attemp
nanodisks) step and washing (by solvent) step, Cu
inside the channels of the PS line-patte
assembly is believed to be similar to a sedim
scussed in Section 2.7.5 The amount of itable concentration ~ 5 ×10-4 mol/L was chosen
ts showed that after participating (of CuxS
xS nanodisks would accumulate
rn, rather than on top of the line pattern The entation process combining with physical
be influenced by physical or spatial constraint and surface property of PS channels
na
tion of PS channels and the arrangement of nanodisks inside the channels
First, we investigated the effect of channel spacing to the assembly of CuxS
nodisks 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)
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,
b
a