Optimal sample preparation for nanoparticle metrology (statistical size measurements) using atomic force microscopy

RESEARCH PAPER

Optimal sample preparation for nanoparticle metrology(statistical size measurements) using atomic forcemicroscopy

Christopher M. Hoo Æ Trang Doan ÆNatasha Starostin Æ Paul E. West ÆMartha L. Mecartney

Received: 27 January 2009 / Accepted: 22 April 2009 / Published online: 18 May 2009

� Springer Science+Business Media B.V. 2009

Abstract Optimal deposition procedures are deter-

mined for nanoparticle size characterization by

atomic force microscopy (AFM). Accurate nanopar-

ticle size distribution analysis with AFM requires

non-agglomerated nanoparticles on a flat substrate.

The deposition of polystyrene (100 nm), silica (300

and 100 nm), gold (100 nm), and CdSe quantum dot

(2–5 nm) nanoparticles by spin coating was opti-

mized for size distribution measurements by AFM.

Factors influencing deposition include spin speed,

concentration, solvent, and pH. A comparison using

spin coating, static evaporation, and a new fluid cell

deposition method for depositing nanoparticles is also

made. The fluid cell allows for a more uniform and

higher density deposition of nanoparticles on a

substrate at laminar flow rates, making nanoparticle

size analysis via AFM more efficient and also offers

the potential for nanoparticle analysis in liquid

environments.

Keywords Atomic force microscope �Instrumentation � Nanoparticles �Nanoscale metrology � Size characterization

Introduction

Nanoparticles play a critical role in a wide range of

applications including nano-optical devices (Taminiau

et al. 2007), data storage devices (Thomson et al.

2004), nanoscale sensors (Brust and Kiely 2002), and

pharmaceuticals (Juillerat et al. 2005). A number of

methods exist for the deposition of nanoparticles on

substrates for the study of nanoparticle properties.

There is considerable interest in using nanoparticles

to build complex structures using ‘‘bottom-up’’ self-

assembly techniques, ‘‘top-down’’ lithographic tech-

niques or combinations of both. For example, gold

nanoparticles can be deposited in specific arrays by

functionalizing a silicon wafer and immersing the

substrate in a nanoparticle solution (Zheng et al.

2000). Nanowires and nanoparticle arrays can be

made by depositing nanoparticles using applied AC

voltages on patterned substrates (Xiong et al. 2006,

2007). Dip coating of colloidal suspensions (Ghosh

et al. 2007) and extreme-ultraviolet interference

lithography (EUV-IL) (Juillerat et al. 2005) has also

C. M. Hoo � T. Doan � M. L. Mecartney (&)

Department of Chemical Engineering and Materials

Science, University of California, Irvine, 916 Engineering

Tower, Irvine, CA 92697-2575, USA

e-mail: [email protected]

N. Starostin � P. E. West

Technology Center, Pacific Nanotechnology, Inc., 18468

Ward St., Fountain Valley, CA 92708, USA

Present Address:N. Starostin

Rosemount Analytical Inc., Emerson Process

Management, 2400 Barranca Parkway, Irvine,

CA 92602, USA

123

J Nanopart Res (2010) 12:939–949

DOI 10.1007/s11051-009-9644-8

been used to produce complex arrays of nanoparti-

cles. The spin coating technique (Bornside et al.

1987) has also been utilized to deposit nanoparticles

onto substrates. When used in conjunction with

substrates patterned using lithography, complex

structures and geometries can be fabricated (Xia

and Brueck 2004, Xia et al. 2004). Spin coating of

nanoparticles has also been used to create binary

colloidal crystals (Wang and Mohwald 2004), uni-

form thin films (Leite et al. 2006; Ogi et al. 2007),

and monodisperse nanoparticle depositions by con-

trolling nanoparticle concentration (Hong et al.

2002).

The development of uniform arrays of complex

nanostructures by the techniques described above

relies on a uniform and controlled size for studying

nanoparticles. The exact nanoparticle size is also

important since a number of properties such as the

melting temperature (Shi 1994) and diffusion rates

(Nakade et al. 2003) are influenced by the size of

nanoparticles. Thus, it is critical to be able to

characterize accurately the size of individual nano-

particles and to determine the size distribution.

Atomic force microscopy (AFM) is one technique

that has the required resolution to easily characterize

individual nanoparticles. Transmission electron

microscopy also can be used, but requires a high

vacuum and only gives a 2D silhouette. Dynamic

light scattering measures aggregate value with lim-

ited success for wide size distributions or anisotropic

particles (Hoo et al. 2008). One major challenge in

the AFM is that agglomeration of particles prevents

an accurate analysis. The aim of our study was to

develop methods to prevent nanoparticle agglomer-

ation and to provide a highly uniform and concen-

trated deposition of nanoparticles on atomically flat

substrates for AFM analysis.

In this article, it is demonstrated how spin coating

parameters (such as nanoparticle solvent, spin speed,

pH, concentration, and choice of substrate) influence

nanoparticle deposition for a wide rang of materials

(gold, polystyrene, silica, and CdSe quantum dots).

Flow deposition of nanoparticles deposited using a

fluid cell in the AFM was also evaluated. Optimal

fluid flow rates and deposition times for the fluid cell

were determined, and the results were compared to

optimal spin coating conditions and static evapora-

tion to evaluate the relative effectiveness of these

deposition techniques.

Experimental procedures

Materials

Aqueous suspensions of 300 and 100 nm silica

nanoparticles and 100 nm NIST traceable polysty-

rene nanoparticles were obtained from Duke Scien-

tific, Inc. An aqueous suspension of 100 nm gold

nanoparticles was obtained from Ted Pella, Inc., and

2–5 nm CdSe quantum dots were obtained from

American Dye, Inc. and Evident Technologies. V1

grade mica disks and poly-L-lysine were obtained

from Ted Pella, Inc., poly-D-lysine was obtained from

BD Bioscience, Inc., and intrinsic (111) silicon

wafers were obtained from Nitride Silicon Wafers.

Silicon wafers were cleaned in Piranha solution (7:3

H2SO4:H2O2) for 10 min and washed with de-ionized

water prior to nanoparticle deposition.

Substrate modification

Silicon wafers were modified by heat treating the

wafers at 300 �C for a duration of 1 h to grow an

oxide layer and increase the surface roughness of the

silicon wafer. Mica disks were cleaved to expose a

clean layer of mica and the surface was functional-

ized with poly-L-lysine or poly-D-lysine.

Spin coating procedure

Substrates were mounted to a 15-mm AFM disk prior

to spin coating in a Headway Research, Inc., EC 101

spin coater. Substrates were spin coated at speeds

ranging from 1,000 to 5,000 rpm in 500 rpm incre-

ments for durations of 30 s. Micropipettes were used

to deposit the solutions. Substrates were then ana-

lyzed using a Pacific Nanotechnology, Inc. Nano

RpTM

atomic force microscope in tapping mode and a

Philips XL-30 scanning electron microscope (SEM).

Environmental fluid cell

One hundred nanometer of NIST traceable polystyrene

nanoparticles were diluted in water to a concentration

of 3 9 10-3 v/v. Substrates were placed in the fluid

cell (Fig. 1), and the diluted solution was pumped into

the cell at various flow rates ranging from 0.06 to

0.12 mL/s. Solution remained in the environmental

cell for a duration of 30 s, 1 min, and 5 min before

940 J Nanopart Res (2010) 12:939–949

123

flushing the cell with de-ionized water. The water was

removed and the substrate was allowed to dry in air in

the AFM before taking AFM scans.

Results and discussion

Nanoparticle evaporation onto a substrate

One of the simplest methods of depositing nanopar-

ticles is to use a static substrate, deposit the solution

and allow for air evaporation. While the evaporation

technique can create nanoparticle thin films (Lin et al.

2001), this technique is not optimal for individual

nanoparticle analysis as it can lead to heavy agglom-

eration. In order to illustrate this, a 100 nm polysty-

rene solution diluted to 3 9 10-3 v/v with water was

dried on a mica substrate functionalized with poly-L-

lysine (Fig. 2).

The drop evaporation technique deposits nanopar-

ticles in large agglomerate clusters due to the particle

interactions during drying (Fujita and Yamaguchi

2006), and individual nanoparticle analysis is difficult

due to agglomeration.

The effect of solvent choice on nanoparticle

deposition

In order to evaluate the effect of solvent choice on the

deposition of silica nanoparticles via spin coating,

two nanoparticle solutions with a concentration of

3 9 10-3 v/v were made with 300 nm silica nano-

particles. The as-received colloidal suspension was

diluted with de-ionized water to make the first

suspension. The second suspension was diluted with

isopropanol. Since isopropanol evaporates quickly,

capillary forces may influence nanoparticle deposition

and increase nanoparticle density on the silicon sub-

strate compared to the water solution. Diluted colloidal

silica solutions were deposited on a research grade

silicon wafer and spun coat at 2,000 rpm for 30 s.

Fig. 1 Experimental

schematic of fluid cell

Fig. 2 AFM image of 100 nm polystyrene nanoparticles

diluted to 3 9 10-3 v/v with water dried on a mica substrate

functionalized with poly-L-lysine

J Nanopart Res (2010) 12:939–949 941

123

Following the spin coat process, the sample was allowed

to air dry before analysis was conducted.

The nanoparticle solution with water (Fig. 3)

shows a non-uniform dispersion of nanoparticles

over the substrate, while the nanoparticle solution

with isopropanol (Fig. 4) shows a tendency to form

small clusters of nanoparticles. This is attributed to

capillary forces induced on the particles during rapid

solution evaporation and spin coating.

While isopropanol increases the density of silica

nanoparticles deposited on the substrate, individual

nanoparticle size analysis is difficult due to

agglomeration.

Solutions of polystyrene and gold nanoparticles

diluted with isopropanol were spin coated at

2,000 rpm for 30 s on silicon wafers, but both

polystyrene and gold nanoparticles did not adhere

to the silicon wafer if isopropanol was used as the

dilution solvent.

The effect of spin speed on nanoparticle

deposition

In order to evaluate the effect of spin speed on the

deposition of silica nanoparticles, a 3 9 10-3 v/v

dilution of 300 nm silica nanoparticles in isopropanol

with a pH of 7 was used. This diluted colloidal

solution was deposited on silicon wafers at spin

speeds from 1,500 to 5,000 rpm in increments of 500

for 30 s (Figs. 5, 6, 7, 8). Following the spin coat

process, the samples were allowed to air dry before

SEM analysis.

At 1,500 rpm (Fig. 7), individual nanoparticles

and agglomerated nanoparticles were observed over

the entire substrate surface. At 2,000 rpm (Fig. 8),

the concentration of agglomerated nanoparticles

decreased while the concentration of individual

nanoparticles increased. At 2,500 rpm (Fig. 9), the

concentration of agglomerated nanoparticles

increases and as the spin speed increases to

5,000 rpm (Fig. 10), clusters of agglomerated nano-

particles are all that remain on the substrate surface.

At low spin speeds, the substrate rotational speed

is insufficient to disperse the nanoparticles over the

substrate surface prior to evaporation resulting in

agglomeration. At fast spin speeds, uniform spread-

ing is not achieved, and nanoparticles form large

agglomerates. At the optimal spin speed of

2,000 rpm, the substrate rotates at sufficient speed

Fig. 3 SEM image of 300 nm silica nanoparticles diluted

with de-ionized water to 3 9 10-3 v/v and spin coated onto a

silicon wafer at 2,000 rpm

Fig. 4 SEM image of 300 nm silica nanoparticles diluted

with isopropanol to 3 9 10-3 v/v and spin coated onto a

silicon wafer at 2,000 rpmFig. 5 SEM image of 300 nm silica nanoparticles spin coated

at 1,500 rpm on a silicon wafer for 30 s

942 J Nanopart Res (2010) 12:939–949

123

to disperse individual nanoparticles over the substrate

surface while limiting the formation of large nano-

particle agglomerates.

The effect of concentration on nanoparticle

deposition

In order to evaluate the effect of concentration on the

deposition of silica nanoparticles via spin coating,

3 9 10-3 v/v (Fig. 9) and 3 9 10-2 v/v (Fig. 10)

dilutions of 300 nm silica nanoparticles concentra-

tions were made with isopropanol, yielding a pH of 7.

The diluted colloidal solutions were deposited on a

research grade silicon wafer and spun coat at

2,000 rpm for 30 s. Following the spin coat process,

the sample was allowed to air dry before analysis was

conducted.

Comparison between the dilute colloidal suspen-

sion (Fig. 9) and the more concentrated colloidal

suspension (Fig. 10) shows that as the concentration

of nanoparticles in solution is increased, the density

of nanoparticles over the surface increases.

Fig. 6 SEM image of 300 nm silica nanoparticles spin coated

at 2,000 rpm on a silicon wafer for 30 s

Fig. 7 SEM image of 300 nm silica nanoparticles spin coated

at 2,500 rpm on a silicon wafer for 30 s

Fig. 8 SEM image of 300 nm silica nanoparticles spin coated

at 5,000 rpm on a silicon wafer for 30 s

Fig. 9 SEM image of 300 nm silica nanoparticles diluted to

3 9 10-3 v/v in isopropanol with a pH of 7

Fig. 10 SEM image of 300 nm silica nanoparticles diluted to

3 9 10-2 v/v in isopropanol with a pH of 7

J Nanopart Res (2010) 12:939–949 943

123

The effect of colloidal suspension pH

on nanoparticle deposition

In order to evaluate the effect of pH on the deposition

of 300 nm silica nanoparticles via spin coating, a

solution with a concentration of 3 9 10-3 v/v was

made with water. The pH of the solution was changed

by adding either sodium hydroxide or nitric acid.

Three solutions were made with pH values of 3, 9,

and 14. Upon reaching the correct pH, dilutions were

sonicated for 5 min to ensure particle dispersion in

solution. The diluted colloidal solutions were then

deposited on a research grade silicon wafers and spin

coated at 2,000 rpm for 30 s. Following the spin

coating process, samples were allowed to air dry

before analysis was conducted.

The solution with a pH of 3 (Fig. 11) shows

agglomerated silica particles that have deposited to

form long nanoparticle strands. The strands were

oriented outward indicting that the agglomerated

particles were aligned in this manner due to centrif-

ugal forces induced on the particles during spin

coating. The solution with a pH of 9 (Fig. 12) shows

that silica particles do not agglomerate to each other

significantly but also shows that the substrate surface

lacks a dense deposition of particles implying that

nanoparticle adhesion to the surface is inhibited.

Compared to the solution with a pH of 9, the solution

with a pH of 14 (Fig. 13) shows an increase in

nanoparticle agglomeration as well as an increase in

nanoparticle density over the substrate surface.

Although making the solution extremely basic

increased the overall nanoparticle density over the

substrate, nanoparticle agglomeration also increased.

Adjusting the pH failed to improve nanoparticle

deposition most likely because the as-received nano-

particles were stable in solution at a pH of 7, and by

adjusting the pH, the attractive and repulsive forces

between nanoparticles were altered making an unsta-

ble colloidal dispersion (Schmidt 2000).

Experiments with polystyrene and gold nanopar-

ticles showed that adjusting the pH of the neutral

colloidal suspension caused nanoparticles to spin off

the substrate rather than deposit on the silicon wafer,

resulting in no deposition. Again, the optimal pH was

that of the original colloidal solution as provided by

the manufacturer, which has been optimized to

prevent flocculation.

Fig. 11 300 nm silica nanoparticle solution with a pH 3

deposited by spin coating at 2,000 rpm for 30 s on a silicon

wafer

Fig. 12 300 nm silica nanoparticle solution with a pH 9

deposited by spin coating at 2,000 rpm for 30 s on a silicon

wafer

Fig. 13 300 nm silica nanoparticle solution with a pH 14

deposited by spin coating at 2,000 rpm for 30 s on a silicon

wafer

944 J Nanopart Res (2010) 12:939–949

123

The effect of substrate choice on nanoparticle

deposition

In order to investigate how substrate selection can

influence nanoparticle deposition, mica substrates

were selected for comparison with silicon due to the

hydrophilic nature of freshly cleaved mica. On a

hydrophilic substrate, nanoparticles in solution are

expected to disperse evenly due to a low wetting

angle. A 3 9 10-3 v/v dilution of 300 nm silica in

water was spin coated at 2,000 rpm for 30 s and

allowed to dry in air before AFM analysis.

On mica (Fig. 14), silica nanoparticle agglomera-

tion is minimal with nanoparticles spread out evenly

over the surface. In addition to non-agglomerated

deposition, cleaved mica yields a contamination free

surface that requires no chemical cleaning and thus

reduced the possibility of AFM artifacts from tip

contamination. The ionic nature of the silica surface

would also promote adhesion on mica. AFM charac-

terization is essential as the surface of mica tends to

charge preventing easy SEM analysis.

Polystyrene and gold nanoparticles were also spin

coated on cleaved mica substrates but nanoparticles

failed to deposit on the substrate. This result indicates

the need to functionalize the substrate to deposit

polymeric and metallic nanoparticles for analysis.

Quantum dots were diluted in water to a concen-

tration of 2.5 9 10-4 v/v and spin coated onto a

cleaved mica substrate (Fig. 15) and a cleaned silicon

wafer (Fig. 16). While the quantum dots deposited on

both the mica wafer and silicon, quantum dots

deposited on the mica substrate were more uniformly

distributed in a more dense packing.

The effect of substrate alteration on nanoparticle

deposition

Oxidized silicon wafers

In order to investigate the effect of altering the

surface of a silicon wafer surface by changing surface

roughness, a silicon wafer was heat treated at 300 �CFig. 14 AFM image of 300 nm silica nanoparticles deposited

on mica

Fig. 15 AFM image of quantum dots diluted in water to

2.5 9 10-4 v/v deposited on a cleaved mica substrate

Fig. 16 AFM image of quantum dots diluted in water to

2.5 9 10-4 v/v deposited on a silicon wafer

J Nanopart Res (2010) 12:939–949 945

123

for 1 h to oxidize the surface. Following heat

treatment, the silicon wafer was allowed to air cool

to room temperature before deposition of silica

nanoparticles. A 3 9 10-3 v/v 300 nm silica nano-

particle dilution with a pH of 7 in water was then

spun coat at 2,000 rpm for 30 s onto the silicon

wafer.

SEM micrograph of the silicon substrates heat

treated for 1 h at 300 �C is shown in Fig. 17. AFM

image of the silica nanoparticles is shown in Fig. 18.

From SEM and AFM images of deposited silica

nanoparticles on heat treated silicon wafers (Fig. 17),

nanoparticles deposit evenly over the surface in a less

agglomerated fashion compared to the agglomeration

of silica nanoparticles on non heat treated silicon

(Fig. 6). The even dispersion of silica nanoparticles is

due to the growth of an oxide layer on the silicon

wafer that increases the surface roughness of the

substrate. The surface roughness before and after

annealing was 0.10 and 0.25 nm, respectively. In

contrast, polystyrene and gold nanoparticles depos-

ited on oxidized silicon wafers formed agglomerated

clusters across the surface.

Quantum dots were not deposited onto heat treated

silicon wafers because the size of the quantum dots is

small enough such that the oxide layer roughness on

the same scale would influence the size distribution

analysis.

Mica functionalized with poly-L-lysine

Mica wafers were functionalized using poly-L-lysine,

and polystyrene and gold nanoparticles were spin

coated onto the functionalized substrate at 2,000 rpm

for 30 s. Nanoparticles dispersed evenly over the

substrate in higher densities (Fig. 19) compared to

non-functionalized substrates (Fig. 20).

Quantum dots deposited onto atomically smooth,

cleaved mica substrates functionalized with poly-L-

lysine did not adhere on the functionalized mica

substrate. Poly-L-lysine inhibited quantum dot depo-

sition, most likely due to the charge associated with

the poly-L-lysine.Fig. 17 SEM image of 300 nm silica nanoparticles deposited

on a silicon wafer oxidized at 300 �C for 1 h

Fig. 18 AFM image of 300 nm silica nanoparticles deposited

on a silicon wafer oxidized at 300 �C for 1 h

Fig. 19 AFM image of polystyrene nanoparticles diluted in

water to 3 9 10-3 v/v deposited on a mica substrate function-

alized with poly-L-lysine

946 J Nanopart Res (2010) 12:939–949

123

Fluid cell nanoparticle deposition

A fluid cell was used to demonstrate the dispersion of

NIST traceable 100 nm polystyrene nanoparticles

deposited poly-D-lysine functionalized glass sub-

strates over three time periods: 30 s, 1 min, and

5 min.

From Fig. 21, nanoparticle density was found to be

12.1 particles/lm2, 8.3 particles/lm2, and 12.4 parti-

cles/lm2 for the 30 s, 1 min, and 5 min samples,

respectively. The 30 s sample and the 5 min sample

have very similar particle densities, however, higher

agglomeration was observed in the 30 s deposition

sample. The 1 min and 5 min deposition showed

little agglomeration, allowing easy AFM character-

ization of the size and density of particles.

In order to evaluate the effect of flow rate on

nanoparticle deposition, polystyrene nanoparticles

were deposited on poly-D-lysine glass functionalized

substrates at fluid flow rates of 0.06 and 0.12 mL/s

(Fig. 22a, b).

Particle densities were found to be 12.7 and

9.2 particles/lm2 for samples deposited at flow rates

of 0.06 and 0.12 mL/s, respectively. This result

showed that higher particle density is produced at a

slower flow rate due to the lack of a turbulent flow

regime. Compared to Fig. 21, particles also distribute

over the sample surface much more evenly. When

particles were deposited in still solutions, agglomer-

ation increases because particles settle onto the

surface. Flowing fluid over the substrate swept the

particles over the surface during deposition. Hence,

deposition in flowing fluid results in a more even

particle distribution and slightly higher particle

density than deposition in still fluids.

Figure 23 shows 100 nm polystyrene particles

deposited on a mica substrate functionalized with

poly-L-lysine via spin coating at 2,000 rpm for 30 s.

This can be compared with polystyrene particles

deposited by fluid cell deposition in Fig. 22. The

initial polystyrene nanoparticle concentration in both

spin coating and fluid cell experiments is the same.

Particle density on the spin-coated sample was found

to be 4.2 particles/lm2 (Fig. 21) compared to particle

densities of 12.7 and 9.2 particles/lm2 (Fig. 20a, b)

for samples deposited at flow rates of 0.06 and

Fig. 20 AFM image of polystyrene nanoparticles diluted in

water to 3 9 10-3 v/v deposited on a mica substrate

Fig. 21 100 nm NIST traceable polystyrene nanoparticles deposited with flow cell on glass functionalized with poly-D-lysine in still

solution for (a) 30 s, (b) 1 min, (c) 5 min

J Nanopart Res (2010) 12:939–949 947

123

0.12 mL/s, respectively. It can be concluded that

deposition with the flow cell in flowing fluid can

provide higher particle density without agglomeration

when compared to deposition via spin coating.

Conclusions

Optimal deposition parameters for spin coating

nanoparticles for AFM analysis are listed in Table 1.

Nanoparticles colloidal dispersions obtained from

the various vendors were stable in solution, and

changing the pH of the nanoparticle solutions did not

enhance nanoparticle deposition onto substrates, most

likely due to destabilization of the sol. For the silica,

gold and polystyrene nanoparticles, all of which were

a similar size, the optimal dilution concentration is

similar in all three cases. The optimal quantum dot

concentration was an order of magnitude lower due to

the fact that the size is two orders of magnitude

smaller. An equivalent volume concentration of

2–5 nm particles would contain a much higher

density of particles compared to the same v/v solution

of 100 nm particles.

There are advantages to using a fluid cell to

deposit nanoparticles on a substrate compared to spin

coating and static drying. A fluid cell allows for a

more uniform and higher density deposition of

nanoparticles over the substrate at laminar flow rates

Fig. 22 100 nm NIST

traceable polystyrene

nanoparticles deposited

with flow cell on glass

substrates functionalized

with poly-D-lysine in

flowing fluid at flow rates

(a) 0.06 mL/s and (b)

0.12 mL/s

Fig. 23 100 nm polystyrene nanoparticles in a 1:100 dilution

with water spin coated at 2,000 rpm on a mica substrate

functionalized with poly-L-lysine

Table 1 Optimal deposition conditions for spin coating nanoparticles

Nanoparticle Dilution solvent pH Optimal concentration Optimal substrate Optimal spin

speed

Silica (100–300 nm)

Fig. 18

Water 7 3 9 10-3 v/v Silicon wafer heat treated at 300 �C for 1 h

or cleaved mica

2,000 rpm for

30 s

Gold (100 nm)

Fig. 24

Water 7 5 9 10-3 v/v Cleaved mica functionalized with

poly-L-lysine

2,000 rpm for

30 s

Polystyrene (100 nm)

Fig. 23

Water 7 3 9 10-3 v/v Cleaved mica functionalized with

poly-L-lysine

2,000 rpm for

30 s

CdSe Quantum Dots

(2–5 nm) Fig. 15

Toluene 7 2.5 9 10-4 v/v Cleaved mica 2,000 rpm for

30 s

948 J Nanopart Res (2010) 12:939–949

123

and also can allow for scanning of nanoparticles in

liquid environments. However, compared to the spin

coating method, the fluid cell requires a larger

amount of the solution in order to fill the fluid cell,

while spin coating requires only a small amount of

solution to wet the substrate.

Acknowledgments This work was supported by the U.S.

Navy under contract # N00244-06-P-2341 and N00244-05-P-

2456. Additional support from Pacific Nanotechnology Inc. is

gratefully acknowledged.

References

Bornside DE, Macosko CW, Scriven LE (1987) On the mod-

eling of spin coating. J Imag Technol 13(4):122–130

Brust M, Kiely CJ (2002) Some recent advances in nano-

structure preparation from gold and silver particles: a

short topical review. Colloids Surf A 202:175–186. doi:

10.1016/S0927-7757(01)01087-1

Fujita M, Yamaguchi Y (2006) Development of three-dimen-

sional structure formation simulator of colloidal nano-

particles during drying. J Chem Eng Jpn 39(1):83–89. doi:

10.1252/jcej.39.83

Ghosh M, Fan F, Stebe KJ (2007) Spontaneous pattern for-

mation by dip coating of colloidal suspensions on

homogeneous surfaces. Langmuir 23:2180–2183. doi:

10.1021/la062150e

Hong YK, Kim H, Lee G, Kim W, Park JL, Cheon J, Koo JY

(2002) Controlled two-dimensional distribution of nano-

particles by spin-coating method. Appl Phys Lett 80:844–

846. doi:10.1063/1.1445811

Hoo CM, Starostin N, West P, Mecartney ML (2008) A

comparison of atomic force microscopy (AFM) and

dynamic light scattering (DLS) methods to characterize

nanoparticle size distributions. J Nanopart Res 10:89–96.

doi:10.1007/s11051-008-9435-7

Juillerat F, Solak H, Bowen P, Hoffmann H (2005) Fabrication

of large-area ordered arrays of nanoparticles on patterned

substrates. Nanotechnology 16:1311–1316. doi:10.1088/

0957-4484/16/8/055

Leite ED, Lee EJH, Ribeiro C, Longo E (2006) Controlled

thickness deposition of ultrathin ceramic films by spin

coating. J Am Ceram Soc 89(6):2016–2020. doi:10.1111/

j.1551-2916.2006.00992.x

Lin XM, Jaeger HM, Sorensen CM, Klabunde KJ (2001)

Formation of long-range-ordered nanocrystal superlattices

on silicon nitride substrates. J Phys Chem B 105(17):

3353–3357. doi:10.1021/jp0102062

Nakade S, Saito Y, Kubo W, Kitamura T, Wada Y, Yanagida S

(2003) Influence of TiO2 nanoparticle size on electron

diffusion and recombination in dye-sensitized TiO2 solar

cells. J Phys Chem B 107:8607–8611. doi:10.1021/

jp034773w

Ogi T, Modesto-Lopez LB, Iskandar F, Okuyama K (2007)

Fabrication of a large area monolayer of silica particles on

a sapphire substrate by a spin coating method. Colloid

Surf A 297:71–78

Schmidt HK (2000) Nanoparticles for ceramic and nanocom-

posite processing. Mol Cryst Liquid Cryst 353:165–179.

doi:10.1080/10587250008025657

Shi FG (1994) Size dependent thermal vibrations and melting

in nanocrystals. J Mater Res 9(5):1307–1313. doi:

10.1557/JMR.1994.1307

Taminiau TH, Segerink FB, Moerland RJ, Kuipers L, Van

Hulst NF (2007) Near-field driving of a optical monopole

antenna. J Opt A 9:S315–S321. doi:10.1088/1464-

4258/9/9/S06

Thomson T, Toney MF, Raoux S, Lee SL, Sun S, Murray CB,

Terris BD (2004) Structural and magnetic model of self-

assembled FePt nanoparticle arrays. J Appl Phys

96(2):1197–1201. doi:10.1063/1.1759393

Wang D, Mohwald H (2004) Rapid fabrication of binary col-

loidal crystals by stepwise spin-coating. Adv Mater

16(3):244–247. doi:10.1002/adma.200305565

Xia D, Brueck SRJ (2004) A facile approach to directed

assembly of patterns of nanoparticles using interference

lithography and spin coating. Nano Lett 4(7):1295–1299.

doi:10.1021/nl049355x

Xia D, Biswas A, Li D, Bruek SRJ (2004) Directed self-

assembly of silica nanoparticles into nanometer-scale

patterned surfaces using spin coating. Adv Mater

16:1427–1432. doi:10.1002/adma.200400095

Xiong X, Makaram P, Busnaina A, Bakhtari K, Somu S,

McGruer N, Park J (2006) Large scale directed assembly

of nanoparticles using nanotrench templates. Appl Phys

Lett 89:193108. doi:10.1063/1.2385067

Xiong X, Busnaina A, Selvarasah S, Somu S, Wei M, Mead J,

Chen C, Aceros J, Makaram P, Dokmeci MR (2007)

Directed assembly of gold nanoparticle nanowires and

networks for nanodevices. Appl Phys Lett 91:063101. doi:

10.1063/1.2763967

Zheng J, Zhu Z, Chen H, Liu Z (2000) Nanopatterned

assembling of colloidal gold nanoparticles on silicon.

Langmuir 16:4409–4412. doi:10.1021/la991332o

Fig. 24 Optimal deposition of 100 nm gold with conditions

listed in Table 1

J Nanopart Res (2010) 12:939–949 949

123