Sonolytic control of rate of gold(III) reduction and size of formed gold nanoparticles: Relation between reduction rates and sizes of formed nanoparticles

Article abstract of DOI:10.1246/bcsj.75.2289

Gold nanoparticles have been applied to various materials, e.g., catalysts, analytical probes, sensors, and nonlinear optics materials. The sonochemical reduction of gold(III) in the presence of a small amount of 1-propanol to minimize the effects of organic molecules on the nucleation and growth processes of Au particles was investigated. The rates of gold(III) reduction were strongly dependent on the atmosphere, bulk solution temperature, intensity of the ultrasound, and distance of the reaction vessel from the oscillator. The rates of reduction under several atmospheres were in the order of CH₄ = CO₂ < N₂ < Ne < He < Ar < Kr, where no reduction occurred under the CH₄ and CO₂ atmospheres. Reduction rates were influenced by the cavitation phenomenon. Upon irradiation, colloidal gold particles with a surface plasmon absorption were formed. The average size of the formed Au particles increased from 30 to 120 nm by selecting the irradiation parameters. The size of Au particles was closely correlated to the initial rate of gold(III) reduction; the higher the reduction rate, the smaller the particles.

Full text of DOI:10.1246/bcsj.75.2289

c. Jpn.
© 2002 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 75, 2289–2296 (2002) 2289
e Chemical
ciety of Ja-
n
e Chemical
ciety of Ja-
n
M
Sonolytic Control of Rate of Gold(Ⅲ) Reduction and Size of Formed
Gold Nanoparticles: Relation between Reduction Rates and Sizes of
Formed Nanoparticles
Sonolytic Control of Rate of Gold(Ⅲ)
K. Okitsu et al.
02
89
96
Kenji Okitsu,* Akihiko Yue,† Shuji Tanabe,† Hiroshige Matsumoto,† Yoshihiro Yobiko,†† and
Youngeok Yoo†††
SJA8
09-2673
100
Japan Science and Technology Corporation, 4-1-8 Hon-cho, Kawaguchi, Saitama 332-0012
†Faculty of Engineering, Nagasaki University, 1-14 Bunkyo-machi, Nagasaki 852-8521
††Technology Research Institute of Osaka Prefecture, 2-7-1 Ayumino, Izumi, Osaka 594-1157
†††Department of Environmental Education, Daegu University, 15 Naeri, Jinryang, Kyungsan, Kyungbuk 712-714, Korea
02
(Received April 22, 2002)
02
.3
The sonochemical reduction of gold(Ⅲ) [tetrachloroaurate(Ⅲ)] was carried out under various irradiation conditions
in an aqueous solution containing only a small amount of 1-propanol. Identical starting materials and the same initial
concentrations were used in each preparation. The rates of gold(Ⅲ) reduction were strongly dependent on the atmo-
sphere, the temperature of the bulk solution, the intensity of the ultrasound, and the distance of the reaction vessel from
the oscillator. For example, the rates of reduction under several atmospheres were in the order of CH4 = CO2 < N2 < Ne
< He < Ar < Kr, where no reduction proceeded under the CH4 and CO2 atmospheres. It was clearly found that the rates
of reduction were influenced by the cavitation phenomenon. Upon irradiation, colloidal gold particles having a surface
plasmon absorption were formed, although in the absence of any stabilizers for the gold particles. It was found by TEM
observations that the average size of the formed gold particles changed from 30 to 120 nm by selecting the irradiation pa-
rameters. The size of the gold particles was closely correlated to the initial rate of gold(Ⅲ) reduction; the higher the rate
of reduction, the smaller the particles. This relationship suggested that the rate of reduction would affect the initial gold
nucleation processes. In a 200 kHz standing wave system used in the present study, the strength of the generated shock
waves as well as the mechanical effects could be regarded as relatively weak.
Gold nanoparticles have been widely applied to various ma-
terials, such as analytical probes,1 biological markers,2 color-
ings,3 catalysts,4 sensors,5 and nonlinear optics materials.6
Such versatile applications are mainly based on the interesting
physicochemical properties arising from the size effects of the
gold nanoparticles. In fact, due to the growing need to develop
a simple method for preparing such functional metal nanopar-
ticles, a number of techniques have been reported during the
last decade. Most of the studies in colloid science are focused
on the control of the size, shape, and structure of metal nano-
particles from a practical point of view.7 In addition, a con-
trolled-chemical reduction system in an appropriate solution is
suggested to be one of the most promising methods in recent
years.7 The mechanism of formation of noble metal nanoparti-
cles including gold, however, has not yet been clarified, partic-
ularly in a solution system using the reduction process of noble
metal ions. The reasons that the mechanism could not be clar-
ified in detail are presumably attributed to the following rea-
sons: There have been difficulties in (Ⅰ) the quantitative analy-
sis of the concentration of unreduced metal ions coexisting
with the formed colloidal particles and their stabilizer in a so-
lution, (Ⅱ) the precise determination of the rate of reduction of
metal ions because the rate was often too fast to trace the
course of the reduction, and (Ⅲ) the control of the rate of re-
duction in conventional thermal reduction systems. Especially,
it is very difficult to control the rate of reduction using identi-
cal starting materials and the same initial concentrations. Thus
far, most of the studies for producing size-controlled particles
have been based on the change in the amounts, types and com-
binations of the starting materials such as metal ions, reducing
agents and stabilizers.7 Based on these reasons, there are very
few reports that provide a clear relationship between the rate of
reduction of noble metal ions and the size of the formed metal
particles in the solution systems. If both the mechanisms of
the reduction of metal ions and of the formation of the colloi-
dal metal could be clarified in more detail, advances in the field
of colloid science as well as advances in nano-technology
should be achieved.
This manuscript describes the sonochemical formation of
gold nanoparticles without any stabilizers and the rates of
gold(Ⅲ) reduction are quantitatively estimated under various
irradiation conditions using a colorimetric method. Locally
generated multi-bubbles with an extremely high temperature
and high pressure are used for the sonochemical reactions.
Nagata et al. and one of the authors8 previously reported the ef-
fect of various organic additives on the rate of gold(Ⅲ) reduc-

2290 Bull. Chem. Soc. Jpn., 75, No. 10 (2002)
Sonolytic Control of Rate of Gold(ꢀ)
tion under ultrasonic irradiation. Here it was clearly indicated
that organic additives with the appropriate hydrophobic prop-
erty enhanced the rates of reduction. For example, alcohols,
ketones, surfactants and water-soluble polymers acted as ac-
celerators for the reduction of noble metals ions under ultra-
sonic irradiation.8,9 Grieser and co-workers10 also reported the
effect of alcohol additives on the reduction of gold(Ⅲ). They
insisted that the rate of the sonochemical reduction of gold(Ⅲ)
was related to the Gibbes surface excess concentration of the
alcohol additives. The average sizes of the sonochemically
formed gold particles were less than 10 nm in the presence of
dodecyl sodium sulfate.8 Although ultrasonic cavitation has
been used in the preparation of various types of transition met-
al nanoparticles during the last decade,11 to our knowledge,
there are only a few reports related to the rate-control of metal
formation and the size-control of the formed particles. In a
previous communication, we have preliminary reported that
the size of the formed gold particles changed by selecting the
parameters involving the irradiation conditions and proce-
dures.12 In this study, we performed a further detailed investi-
gation on various parameters influencing the rates of the
gold(Ⅲ) reduction and the size of the formed gold particles, in
which identical starting materials were used in each prepara-
tion to clarify the net effect of the rate of gold(Ⅲ) reduction on
the size of the formed gold particles. The effects of ultrasonic
cavitation on the rate of reduction and on the behavior of the
formed colloidal particles are also discussed.
Fig. 1. Standing-wave type irradiation system.
tion was measured by a calorimetric method, as described else-
where.13 In addition, the rate of H2O2 formation in the sonolysis
of Ar-saturated water was estimated by a method similar to Fricke
dosimetry:8 A 65 mL of Ar-saturated aqueous solution containing
50 mM FeSO4 and 0.8 N H2SO4 was irradiated and then the
amount of Fe(Ⅲ) ions (ε = 2194 M−1 cm−1, λ = 304 nm) formed
from H2O2 and ·OH oxidation of Fe(Ⅱ) was spectrophotometrical-
ly measured.
Experimental
Colorimetric Analysis. The absorption spectra of the irradi-
ated solutions were measured by an UV-vis spectrophotometer
(Shimadzu UV-2100). The concentration of gold(Ⅲ) in the irradi-
ated solution was determined by a colorimetric method using
NaBr,8,14 i.e., the absorption peak of [AuBr4]−, which was formed
in a ligand exchange reaction between Cl− and Br−, was clearly
observed at 382 nm by the addition of saturated NaBr solution to
the sample solution containing the unreduced [AuCl4]−. At the
same time, the aggregation of the colloidal gold particles rapidly
occurred. The colorimetric analysis was, therefore, carried out for
the solution filtered through a membrane filter with a pore-size of
0.2 µm to avoid any spectrophotometric interference of the gold
aggregates formed after the addition of the NaBr solution.
Characterization of the Formed Gold Nanoparticles. The
sonochemically formed gold particles were observed using a
transmission electron microscope (TEM, JEOL JEM-100S) at an
acceleration voltage of 100 kV. Specimens for the TEM observa-
tions were prepared as follows: The irradiated solution (irradia-
tion time, 20 min) was immediately mixed with an aqueous sus-
pension of alumina powders to immobilize the formed gold parti-
cles on the surface of the alumina. This procedure was carried out
to avoid aggregation of the formed gold particles during the subse-
quent drying processes. The sonochemically formed gold parti-
cles were rapidly immobilized on the surface of the alumina pow-
ders so that the color of the solution became transparent. On the
other hand, the white alumina powders were tinged reddish pur-
ple. The suspension composed of the gold particles immobilized
on the alumina powders was filtered and washed with distilled wa-
ter. The aqueous suspension of such powders was then dropped
onto a Cu grid coated with a collodion film and then dried in a
vacuum.
Reagents. Reagent-grade Na[AuCl4]·2H2O, 1-propanol,
FeSO4·7H2O, H2SO4, NaBr, and α-Al2O3 (a nominal particle size
of 0.5 µm) were purchased from Wako and used without further
purification. All the water used in the present experiment was
treated by a conventional ion-exchange and distillation procedure.
Helium, neon, argon, krypton, nitrogen, carbon dioxide and meth-
ane of more than 99.99% purity were obtained from Japan Air
Liquid, Nippon Sanso and Sumitomo Seika.
Procedure of Ultrasonic Irradiation. Ultrasonic irradiation
was carried out using an ultrasonic generator (Kaijo 4021type;
Lot. No. Y232, frequency; 200kHz) and a 65 mm φ barium titan-
ate oscillator (Kaijo; Lot. No. Y300), which was operated at ca. 20
to 200 W in a water bath maintained at a constant temperature.
The irradiation set-up and the characteristics of the reaction vessel
used are illustrated in Fig. 1. The glass vessel was cylindrical and
it had a port covered by a silicon rubber septum for gas bubbling
or sampling without exposing the sample to air. The bottom of the
vessel was planar, 1 mm thick and 55 mm φ in diameter. An aque-
ous solution of Na[AuCl4] (1.0 mM, 65 mL) was added to the ves-
sel and then it was purged with the appropriate gas. As an acceler-
ator for the gold(Ⅲ) reduction, 1-propanol was injected into the
gas-purged solution, corresponding to a 20 mM concentration. In
order to minimize the effects of the 1-propanol molecules on the
nucleation or growth of the gold particles, a very small amount of
1-propanol was added in the present study. Various irradiation pa-
rameters were selected, such as the atmosphere, the temperature of
the solution, the intensity of the ultrasound, and the distance be-
tween the reaction vessel and the oscillator. During the course of
the irradiation, the vessel was closed from the air atmosphere. On
the other hand, the ultrasonic power injected into the sample solu-

K. Okitsu et al.
Bull. Chem. Soc. Jpn., 75, No. 10 (2002) 2291
Results and Discussion
Sonochemical Reduction of Gold(Ⅲ). The reduction of
gold(Ⅲ) smoothly proceeded upon irradiation, which could be
attributed to the formation of reducing species from the 1-pro-
panol and water sonolysis. The chemical reactions induced by
the ultrasonic irradiation of a liquid are known to result from a
cavitation phenomenon and the sites of the sonochemical reac-
tions are characterized by three different regions in an aqueous
solution system: (1) The inside of the collapsing cavitation
bubbles where several thousands of degrees and more than
hundreds of atmospheres are produced.15,16 (2) The interfacial
region between the cavitation bubbles and the bulk solution
where the temperature is lower than the inside of the cavitation
bubbles but still high enough for thermal decomposition of the
solutes to occur. (3) The bulk solution at ambient temperature
where the reactions of the solute molecules with various radi-
cals and unstable species, which escaped from the interfacial
region, take place.
Fig. 3. Rate of Au(Ⅲ) reduction (ꢀ) in the presence of
20 mM 1-propanol and rate of H2O2 formation (ꢅ) in pure
water under various intensities. Condition: Ar atmosphere,
temperature; 20 °C, distance; 4.0 mm.
According to many previous reports about sonochemistry, it
is predictable that the chemical reactions induced by the ultra-
sonic cavitation should be sensitively influenced by the irradia-
tion conditions and procedures. In this study, various ultrason-
ic parameters influencing the rates of the gold(Ⅲ) reduction
were investigated. The dependence of the ultrasonic intensities
on the rates of reduction were first considered. Figure 2 shows
the changes in the concentration of gold(Ⅲ) for several ultra-
sound intensities, in which the intensities were evaluated by
the calorimetric method. It was clearly observed that the con-
centration of gold(Ⅲ) rapidly decreased with the irradiation
time and the reduction completely was finished within the
sured from the water sonolysis in the absence of gold(Ⅲ) and
1-propanol. Both the rates of gold(Ⅲ) reduction and of H2O2
formation were also found to increase with the increasing in-
tensities. In addition, the threshold value of the gold(Ⅲ) re-
duction seemed to be smaller than that of the H2O2 formation.
Nagata et al. and one of the authors previously reported that
the sonochemical reduction of gold(Ⅲ) readily occurred in an
aqueous solution due to the addition of organic compounds,
such as alcohols, ketones, surfactants and water-soluble poly-
mers.8 It was found that organic additives are a source of re-
ducing species under sonication and that the rate of gold(Ⅲ)
reduction strongly depended on the types and amounts of such
organic additives. The reduction pathways via several reduc-
ing species have also been suggested in the presence of an al-
cohol, RHOH, as follows. Equations 1–4 show the common
basal reactions in aqueous sonochemistry,
20min irradiation of 1.20 and 1.43 W cm−2
. The rates of
gold(Ⅲ) reduction changed by the ultrasound intensities. The
initial rates of the gold(Ⅲ) reduction are summarized in Fig. 3.
To discuss the relation between the rates of reduction and the
cavitation phenomenon, the rates of H2O2 formation are also
plotted in Fig. 3, in which the rates of H2O2 formation are mea-
)))
H2O → OH + ·H
2·OH → H2O2
2·H → H2
(1)
(2)
(3)
(4)
·
·OH + ·H → H2O
where the symbol of ))) indicates the action of ultrasonic irra-
diation. The reducing species of H atoms and H2 molecules
are formed during the sonolysis of water. The rates of gold(Ⅲ)
reduction by H atoms and H2 molecules, however, might be ac-
tually slow, although the rate constants in both cases are ex-
tremely high.8,17 The RHOH molecules that accumulated on
the interface of the cavitation bubbles also rapidly react with
active radicals as follows:
Fig. 2. Changes in the concentration of Au(Ⅲ) during ultra-
sonic irradiation under various intensities. Condition: 1-
propanol; 20 mM, Ar atmosphere, temperature; 20 °C, dis-
tance; 4.0 mm. (ꢀ) 0.15 W cm−2, (ꢁ) 0.53 W cm−2, (ꢂ)
·
RHOH + ·OH(·H) → ROH + H2O(H2)
(5)
·
0.92 W cm−2, (ꢃ) 1.20 W cm−2, (ꢄ) 1.43 W cm−2
.
where secondary reducing radicals, ROH, are formed by hy-

2292 Bull. Chem. Soc. Jpn., 75, No. 10 (2002)
Sonolytic Control of Rate of Gold(ꢀ)
drogen abstraction from RHOH with OH radicals and H at-
oms. Strong oxidants such as OH radicals are advantageously
quenched and H2 molecules are also generated in this reaction.
Furthermore, the RHOH molecules are pyrolyzed at the inter-
face and the inside of the bubbles,18
RHOH → Pyrolysis products
(6)
where a part of the pyrolysis products also correspond to re-
ducing species.8 Several types of reducing species subsequent-
ly react with gold(Ⅲ) to form gold metal particles.
)))
Au3+ + H2O + RHOH →→→ Au0 + 3H+ + products (7)
nAu0 → (Au0)n
(8)
Fig. 4. Changes in the concentration of Au(Ⅲ) during ultra-
sonic irradiation under various atmospheres. Condition:
Intensity; 1.43 W cm−2, 1-propanol; 20 mM, temperature;
20 °C, distance; 4.0 mm. (ꢅ) Kr, (ꢆ) Ar, (ꢇ) Ne, (ꢈ) He,
(ꢀ) N2, (ꢁ) CO2, (ꢂ) CH4.
Gold(Ⅱ) ions might be formed in the reduction processes, but
they are very unstable so that the following disproportionation
reaction would also occur.19
2Au2+ → Au+ + Au3+
(9)
gases such as the ratio of specific heats, γ = Cp/Cv, the thermal
conductivity, and the solubility. The γ value is one of the most
important factors influencing the attained temperature inside
the collapsing bubbles, according to the hot spots theory.20
As seen in Fig. 3, it was found that the rates of gold(Ⅲ) re-
duction were ca. 8–24 times faster than those of the H2O2 for-
mation for each ultrasound intensity. In particular, the rate of
H2O2 formation was extremely slow when the intensity of the
ultrasound was weak, while the rate of gold(Ⅲ) reduction
seemed to be relatively fast. This result implies that the reduc-
tion pathways, as mentioned above, might be different among
the different ultrasound intensities. The results in Fig. 3
roughly suggest that the reduction of gold(Ⅲ) proceeds via py-
rolysis radical species (formed by Eq. 6) as the main reduction
pathway and the reduction via secondary reducing radicals
(formed by Eq. 5) could be regarded as a minor reaction path-
way, particularly at low intensity around the threshold value of
the water sonolysis. Riesz and co-workers previously reported
that various pyrolysis radicals were formed in the sonolysis of
alcohol-aqueous solutions, which were determined by a spin
trapping ESR method.18 It is reasonable that pyrolysis radicals
are also formed in the present irradiation system because the
pyrolysis of water molecules easily occurs. As seen in Figure 3,
the difference in the threshold values between the gold(Ⅲ) re-
duction and the H2O2 formation might be attributed to the ease
of pyrolysis between the 1-propanol and water molecules. On
the other hand, it was also recognized that no reduction oc-
curred using a conventional ultrasonic cleaner (Honda Electric
Co., W-113, 28kHz, 100W, Bath-volume: ca. 2 L). These re-
sults mean that the gold(Ⅲ) reduction requires sufficiently hot
bubbles which cause pyrolysis of the water and 1-propanol
molecules. The temperatures and/or number of cavitation bub-
bles should increase with the increasing intensity in the present
irradiation system.
The maximum temperature in the collapsing bubble, Tmax
,
could be roughly determined by the following equation:
Tmax = T0Pmax(γ −1)/P
(10)
where T0 is the temperature of the sample solution, Pmax is the
maximal pressure in the liquid at the moment of transient col-
lapse, and P is the sum of the vapor pressure of the solvent and
the atmospheric gas pressure. As seen in Fig. 4, the rates of re-
duction under the noble gas atmospheres were faster than those
under the polyatomic molecule gas atmospheres. Under the
CO2 and CH4 atmospheres, no reduction proceeded at all. This
result was in good agreement with the order of the γ values,
which increase in the order of CH4 (γ ;1.3) = CO2 (1.3) < N2
(1.4) < noble gases (ca. 1.6–1.7). Furthermore, different rates
were also observed among the noble gas atmospheres. The
rates were in the order of He < Ne < Ar < Kr. This order was
consistent with the attained temperature inside the collapsing
bubbles as previously reported.21 The maximum temperature
inside the collapsing bubbles would decrease with the increas-
ing thermal conductivity of the gases, where the conductivity
(10−4 W m−1 K−1) was in the order of He (1499) > Ne (493)
> Ar (177.2) > Kr (94.2) at 1atm and 300 K.22 Furthermore, it
has also been suggested that the solubility of the gases might
contribute to the number of cavitation bubbles:16 the solubility
in water (cm3 kg−1) increases in the order of He (9.375) < Ne
(11.15) < Ar (36.60) < Kr (66.98) at 1atm and 293.15 K.23
By using such atmospheric parameters, the rates of the
sonochemical reduction of gold(Ⅲ), as well as of noble metal
ions, should be controlled.
Figure 4 shows the effect of the atmospheric gases on the
sonochemical reduction of gold(Ⅲ). It was observed that the
rates of reduction were strongly dependent on the irradiated at-
mospheres in spite of the ultrasound intensity being the same
in all cases. In general, the cavitation effects are well-known
to be influenced by the physical properties of the atmospheric
Figures 5 and 6 show the initial rates of gold(Ⅲ) reduction
obtained under additional irradiation parameters. Figure 5 in-
dicates the effects of the distance from the oscillator to the bot-
tom of the reaction vessel. The maximum rate of reduction

K. Okitsu et al.
Bull. Chem. Soc. Jpn., 75, No. 10 (2002) 2293
lows: When the temperature of the bulk solution (T0) increas-
es, Tmax becomes lower because the value of P steeply increas-
es with the increasing amount of water and 1-propanol vapor
inside the bubble. In addition, the value of γ also decreases be-
cause the γ value of water and 1-propanol is substantially low-
er than that of argon gas (e.g., γ = 1.67 for argon, ca. 1.3 for
water and ca. 1.1 for 1-propanol). Consequently, the maxi-
mum temperature in the cavitation bubbles drops with increas-
ing the amount of water and 1-propanol vapor inside the bub-
ble. Thus, the amounts of reducing species decrease for a
higher temperature solution. This speculation was also sup-
ported by other reports25,26 in which the relation between the
solvent and/or solute vapor in the collapsing bubbles and its
decomposition efficiency was clearly indicated.
Fig. 5. Rate of Au(Ⅲ) reduction as a function of the distance
between the reaction vessel and the oscillater. Condition:
Intensity; 1.43W cm−2, Aratmosphere, temperature;20 °C.
Formation of Gold Nanoparticles. The color of the solu-
tion rapidly changed from an initial pale yellow to reddish pur-
ple due to the ultrasonic irradiation. UV-vis spectral analyses
indicated that a surface plasmon absorption attributed to the
colloidal gold particles had emerged. This result suggests the
progress of the gold(Ⅲ) reduction and the formation of gold
nanoparticles, as seen in Eqs. 7 and 8. The sonochemically
formed gold metals were characterized by TEM spectroscopy.
To avoid aggregation of the formed gold particles during the
preparation of the TEM specimens, we added submicron-sized
alumina powders immediately after the sonochemical reduc-
tion.
Figure 7 shows representative TEM photographs in which
the ultrasonic intensity is the only changing irradiation param-
eter. A number of small gold particles immobilized on the alu-
mina surface were clearly observed and the formation of
spherical gold particles was recognized in each preparation.
For example, the average size of the gold particles was 53.8
nm at 0.92 W cm−2, 50.8 nm at 1.20 W cm−2, and 35.5 nm at
1.43 W cm−2, respectively. Since no stabilizer was used in the
present preparation, the particle sizes were slightly larger com-
pared to those prepared in the presence of suitable stabilizers.8
The gold particles prepared by the other irradiation parame-
ters were also characterized. Table 1 shows the initial rates of
the gold(Ⅲ) reduction, the average size of the formed gold par-
ticles and the reduction ratio under various irradiation condi-
tions. The standard deviation (SD) of the size of the formed
gold particles is also indicated in Table 1. The sizes of the
formed gold particles were found to sensitively change with
the irradiation parameters. To clarify the relation between the
rate of gold(Ⅲ) reduction and the size of the formed gold parti-
cles, we plotted the experimental data in Fig. 8. Despite the
different irradiation conditions, it was clearly recognized that
the size of the gold particles decreased with an increase in the
rate of gold(Ⅲ) reduction. If shock waves generated by ultra-
sonic irradiation affected the growth of the gold particles, the
size of the particles might become larger with stronger shock
waves. According to the results of Fig. 3, one can notice that
the rate of gold(Ⅲ) reduction is closely related to the absorbed
ultrasonic power, which should correspond to the strength of
the cavitation. Therefore, the strength of shock waves must
become greater when the rate of gold(Ⅲ) reduction is higher,
assuming a simple dependence of the cavitation strength vs
the absorbed ultrasonic power. To allow us to consider the ef-
fect of the ultrasonic intensities on the particle sizes, the sizes
Fig. 6. Rate of Au(Ⅲ) reduction as a function of the temper-
ature of the bulk solution. Condition: Intensity; 1.43
W cm−2, Ar atmosphere, distance; 4.0 mm.
was seen at the distance of ca. 3.8 mm, which was approxi-
mately consistent with the half-wavelength of the ultrasound
(3.71 mm) used in the present study. This result could be ex-
plained as follows: The sound is effectively transmitted into
the reaction vessel when the node plane formed in the standing
wave system is located just at the bottom of the vessel.24
Figure 6 shows the influence of the bulk solution tempera-
ture. It was recognized that the initial rates of reduction sensi-
tively changed in spite of a small change in the temperature
from 2 to 40 °C. The rate increased by increasing the tempera-
ture up to 20 °C, but it decreased at higher temperatures. This
result could be elucidated by the following cavitation phenom-
enon: In the range of 2 to 20 °C, as the temperature of the bulk
solution increases, the water and alcohol molecules are more
easily vaporized in the cavitation bubbles, resulting in the ef-
fective formation of reducing species which promote the re-
duction. On the other hand, when the temperature of the solu-
tion further increased, the temperature generated in the col-
lapsing bubble does not elevate due to the cushion effect which
was caused by the incorporation of an excess amount of water
and alcohol molecules into the bubble. The cushion effect
could be theoretically explained on the basis of Eq. 10, as fol-

2294 Bull. Chem. Soc. Jpn., 75, No. 10 (2002)
Sonolytic Control of Rate of Gold(ꢀ)
Fig. 7. Effect of the ultrasonic intensity on the size of the formed Au nanoparticles. Condition: 1-propanol; 20 mM, Ar atmosphere,
temperature; 20 °C, distance; 4.0 mm, irradiation time; 20 min. (a) 0.92 W cm−2, (b) 1.20 W cm−2, (c) 1.43 W cm−2
.
Table 1. Initial Rate of Au(Ⅲ) Reduction, Average Size of Sonochemically Formed Au Particles and Reduction Ratio under Various
Irradiation Conditions
Parameters
Initial rate of
Intensity of
Distancea)
/mm
Run
reduction
Average size SD/nm
Reduction ratiob)/%
Temperature of
bulk solution/°C
/µM min−1
ultrasound
/W cm−2
0.53
1
2
3
4
5
6
7
8
9
20
20
20
20
20
20
40
30
2
4.0
4.0
4.0
4.0
3.5
4.5
4.0
4.0
4.0
4.0
41.1
65.1
71.3
87.5
75.8
49.2
37.0
49.5
50.2
60.4
125.7 52.7c)
53.8 16.6
50.8 17.1
55
78
100
100
98
84
63
74
79
0.92
1.20
1.43
1.43
1.43
1.43
1.43
1.43
35.5
33.3
8.0
8.5
71.8 16.4
71.5 19.0
58.3 16.1
40.3 11.4
41.6 10.1
10
1.43
10
100
a) Distance between reaction vessel and oscillator, b) Reduction ratio at 20 min irradiation. c) The number of measurments are not
enough.
obtained from the intensity-dependent experiments are indicat-
ed as the closed-circle symbol in Fig. 8. It was confirmed that
the rate of reduction became faster as the ultrasonic intensity
increased. It seems that the size of the gold particles is little
affected by the strength of the shock waves because the smaller
particles were formed at the stronger ultrasound intensity. The
behavior of the gold particles under sonication is also dis-
cussed in a later section.
The relation between the size of the formed gold particles
and the rate of gold(Ⅲ) reduction was analyzed on the basis of
the material balance. Assuming the formation of monodis-
persed gold nanoparticles with a spherical shape, the material
balance is shown as Eq. 11:
(4/3) π (D/2)3ρ N = CM
(11)
Fig. 8. Relation between the rate of Au(Ⅲ) reduction and the
average size of the formed gold particles. Each error bar
corresponds to the standard deviation of the size of the
gold paritcles. Closed circles correspond to the depen-
dence of the ultrasonic intensities, in which the rate of re-
duction increases with increasing the intensity. (a) 0.53 W
where D, ρ, N, C and M are the diameter of the formed particle
(m), the density (kg m−3), the number of particles (m−3), the
initial concentration of the metal ions (mol m−3) and the mo-
lecular weight (kg mol−1), respectively. Thus, D can be ex-
pressed as follows:
cm−2, (b) 0.92 W cm−2, (c) 1.20 W cm−2, (d) 1.43 W cm−2
.

K. Okitsu et al.
D = (6CM/πρN)1/3
Bull. Chem. Soc. Jpn., 75, No. 10 (2002) 2295
(12)
particles, the size of the gold particles should not consistently
increase with the decreasing rate of gold(Ⅲ) reduction for the
The total number of formed gold particles (N) should be corre-
lated to the number of formed nuclei. In addition, the nuclei
number should also correspond to the initial rate of gold(Ⅲ) re-
duction (dC/dt). Thus, the following equation could be as-
sumed:
different experimental parameters.
Accordingly, if the
sonochemical nucleation and/or growth process of metal clus-
ters and nanoparticles was hardly affected by the ultrasounds,
the present results might also be applied to the interpretations
of the formation process in a conventional reduction system.
Conclusion
N ∝ (dC/dt)
(13)
In the present study, the sonochemical formation of gold
particles was carried out in the presence of a small amount of
1-propanol to minimize the effects of organic molecules on the
nucleation or growth processes of the gold particles. The rates
of sonochemical reduction and the size of the formed particles
could be changed by selecting the irradiation parameters, in
spite of using the same starting materials and the same initial
concentrations. The rate of gold(Ⅲ) reduction strongly affect-
ed the sizes of the formed gold nanoparticles. When a high
frequency was used, physical effects due to the cavitation on
the nucleation in the nano-scale might be quite small. In addi-
tion, the effects of the ultrasonic cavitation on colloidal disper-
sion systems would be somewhat different in the size of the
colloidal particles between the nano- and micro-meter ranges.
In the near future, by selecting stabilizers and combining or-
ganic accelerators as well as by applying the irradiation param-
eters clarified in the present study, transition metal nanoparti-
cles having various particle sizes and distributions could be
formed.
Therefore, Eq. 12 could be roughly expressed as follows:
D ∝ (6CM/πρ)1/3·(dC/dt)−1/3
(14)
Equation 14 can adequately account for the present experimen-
tal results such that the size of the formed gold particles de-
creases with the increasing rate of gold(Ⅲ) reduction. The bro-
ken line in Fig. 8 is drawn as a visual aid.
On the other hand, if strong shock waves were generated, it
would be expected that the size of the gold particles should be-
come larger at the higher intensities so that a clear relationship
between the rate of reduction and the particle size could not be
obtained as seen in Fig. 8. The observed clear relation be-
tween them indicates that shock waves generated in the present
irradiation system might be extremely weak. Consequently,
our experimental results imply that the initial rate of gold(Ⅲ)
reduction strongly affects the nucleation process of the gold
particles: The size of the gold particles depends on the number
of the gold nuclei which correspond to the rate of reduction.
The cavitation bubbles with an extremely high temperature
and high pressure were locally generated by the ultrasonic irra-
diation in solution. Their size and lifetime were calculated to
be about several micrometers and less than sub-microseconds,
respectively, in the 200 kHz irradiation system.27 Thus, the
radical reactions that proceeded in the bulk solution could be
regarded as a mild condition at around room temperature.
Analogous to this, the nucleation and growth of the gold parti-
cles is also suggested to occur in the bulk solution at such mild
temperatures.
References
1
a) L. G. Olson, Y. S. Lo, T. P. Jr. Beebe, and J. M. Harris,
Anal. Chem., 73, 4268 (2001). b) B. Neiman, E. Grushka, and O.
Lev, Anal. Chem., 73, 5220 (2001).
2
a) A. Kusumi, Y. Sako, and M. Yamamoto, Biophys. J., 65,
2021 (1993). b) J. W. Slot, and H. J. Geuze, J. Cell. Biol., 90, 533
(1981).
3
4
P. Mulvaney, MRS Bull., 26, 1009 (2001).
a) M. Haruta, and M. Date, Appl. Catal., A, 222, 427
(2001). b) M. Valden, X. Lai, and D. W. Goodman, Science, 281,
1647 (1998).
Ultrasound is generally used to disperse or agglomerate
some particles or powders in solution by use of its mechanical
effects, such as micro-jet impacts and shock waves, but the be-
havior of colloidal nanoparticles under sonication has not yet
been clarified. It has been reported that the transition metal
particles in the microns range are subjected to shock waves in-
duced by the irradiation of a 20 kHz ultrasonic generator, re-
sulting in the occurrence of interparticle collisions via a melt-
ing process.28 On the other hand, the application of high-fre-
quency ultrasound is suitable for cleaning a silicone wafer
without any damage.29 It appears that the mechanical effects
due to the cavitation are attributed to the size of the collapsing
bubbles: When the sizes of the collapsing bubbles are larger,
the stronger mechanical effects would emerge. Therefore, me-
chanical effects would be relatively weak in the present 200
kHz sonication system in comparison with a conventional low-
frequency sonication system.30 In practice, various types of
nanoparticles could be successfully formed without any aggre-
gation by ultrasonic irradiation in the presence of stabiliz-
ers.8,9,31 If any mechanical effects affected the growth of the
5
a) F. Meriaudeau, T. Downey, A. Wig, A. Passian, M.
Buncick, and T. L. Ferrell, Sens. Actuators, B, 54, 106 (1999). b)
Y. Jin, X. Kang, Y. Song, B. Zhang, G. Chng, and S. Dong, Anal.
Chem., 73, 2843 (2001).
6
I. Tanahashi, Y. Manabe, T. Tohda, S. Sasaki, and A.
Nakamura, J. Appl. Phys., 79, 1244 (1996).
a) M. A. EL-Sayed, Acc. Chem. Res., 34, 257 (2001). b) N.
7
R. Jana, L. Gearheart, and C. J. Murphy, Langmuir, 17, 6782
(2001). c) R. M. Crooks, M. Zhao, L. Sun, V. Chechik, and L. K.
Yeung, Acc. Chem. Res., 34, 181 (2001). d) D. V. Leff, P. C.
Ohara, J. R. Heath, and W. M. Gelbart, J. Phys. Chem., 99, 7036
(1995). e) C. H. Walker, J. V. St. John, and P. Wisian-Neilson, J.
Am. Chem. Soc., 123, 3846 (2001). f) B. V. Enustun, and J.
Turkevich, J. Am Chem. Soc., 85, 3317 (1963). g) H. Hirai, Y.
Nakao, and N. Toshima, Chem. Lett., 1978, 545. h) G. Schmid, R.
Pugin, T. Sawitowski, U. Simon, and B. Marler, Chem. Commun.,
1999, 1303. i) T. Yonezawa, S. Onoue, and N. Kimizuka,
Langmuir, 17, 2291 (2001). j) K. Esumi, A. Suzuki, A. Yamahira,
and K. Torigoe, Langmuir, 16, 2604 (2000). k) S. Chen, and K.

2296 Bull. Chem. Soc. Jpn., 75, No. 10 (2002)
Sonolytic Control of Rate of Gold(ꢀ)
24 A. Henglein, Ultrasonics, 25, 6 (1987).
25 J. Buttner, M. Gutierrez, and A. Henglein, J. Phys. Chem.,
95, 1528 (1991).
Kimura, Langmuir, 15, 1075 (1999). l) J. H. Hodak, A. Henglein,
and G. V. Hartland, J. Phys. Chem. B, 104, 9954 (2000). m) M. Y.
Han, and C. H. Quek, Langmuir, 16, 362 (2000).
8
Y. Nagata, Y. Mizukoshi, K. Okitsu, and Y. Maeda, Radiat.
26 Y. Mizukoshi, H. Nakamura, H. Bandow, Y. Maeda, and Y.
Nagata, Ultrason. Sonochem., 6, 203 (1999).
Res., 146, 333 (1996).
9
a) K. Okitsu, H. Bandow, Y. Maeda, and Y. Nagata, Chem.
27 The resonance radius (R0) of the bubbles and the collapsing
time (τ) was calculated to be ca. 13 µm and ca. 0.67 µs,
respectively, on the basis of Neppiras’ equations [Ref. 20] with the
parameters of 1.256 × 105 sec−1 for ωa and 1000 kg/m3 for ρ, and
the assumed parameters of 1.01325 × 105 N m−2 for Ph and
2.0265 × 105 N m−2 for Pa. In addition, the radius of the hot
bubbles could be calculated to be ca. 4 µm according to the
equation of adiabatic compression, in which the temperature of
the resonance bubbles and the hot bubbles, and γ value were 293
K, 3500 K and 1.67, respectively.
Mater., 8, 315 (1996). b) K. Okitsu, Y. Mizukoshi, H. Bandow, Y.
Maeda, T. Yamamoto, and Y. Nagata, Ultrason. Sonochem., 3, 249
(1996).
10 F. Grieser, R. Hobson, J. Sostaric, and P. Mulvaney,
Ultrasonics, 34, 547 (1996).
11 a) A. Gedanken, R. Reisfeld, E. Sominski, O. Palchik, Y.
Koltypin, G. Panczer, M. Gaft, and H. Minti, J. Phys. Chem. B,
104, 7057 (2000). b) G. Dantsin, and K. S. Suslick, J. Am. Chem.
Soc., 122, 5214 (2000). c) N. A. Dhas, A. Ekhtiarzadeh, and K. S.
Suslick, J. Am. Chem. Soc., 123, 8310 (2001). d) W. Chen, W.
Cai, Z. Zhang, and L. Zhang, Chem. Lett., 2001, 152.
12 K. Okitsu, A. Yue, S. Tanabe, H. Matsumoto, and Y.
Yobiko, Langmuir, 17, 7717 (2001).
13 T. Kimura, T. Sakamoto, J-M. Leveque, H. Sohmiya, M.
Fujita, S. Ikeda, and T. Ando, Ultrason. Sonochem., 3, 157 (1996).
14 Y. Mizukoshi, K. Okitsu, H. Bandow, Y. Nagata, and Y.
Maeda, Bunseki Kagaku, 45, 327 (1996).
15 Y. T. Didenko, W. B. McNamara, and K. S. Suslick, J. Am.
Chem. Soc., 121, 5817 (1999).
16 Y. Nagata, H. Okuno, Y. Mizukoshi, and Y. Maeda, Chem.
Lett., 2001, 142.
17 A. S. Ghosh-Mazumadar, and E. J. Hart, Adv. Chem. Ser.,
81, 231 (1968).
18 V. Misik, and P. Riesz, Ultrason. Sonochem., 3, 173 (1996).
19 K. Kurihara, J. Kizling, P Stenius, and J. H. Fendler, J. Am.
Chem. Soc., 105, 2574 (1983).
28 S. J. Doktycz, and K. S. Suslick, Science, 247, 1067
(1990).
29 “Chou-onpa Binran,” Maruzen, Japan (1999), pp. 669–677.
30 To discuss the cavitation strength related to mechanical
effects (shock waves and microjet impacts, etc), the effects of
cavitation on an immersed aluminum foil were investigated. The
result showed that no pit attributed to cavitation damage was
observed, suggesting that mechanical effects would be weak in the
present 200 kHz irradiation system. On the other hand, a number
of pits and holes on the foil were observed by irradiation using a
conventional 28 kHz ultrasonic cleaner.
31 a) Y. Nagata, Y. Watanabe, S. Fujita, T. Dohmaru, and S.
Taniguchi, J. Chem. Soc., Chem. Commun., 1992, 1620. b) Y.
Mizukoshi, K. Okitsu, T. Yamamoto, R. Oshima, Y. Nagata, and Y.
Maeda, J. Phys. Chem. B, 101, 5470 (1997). c) Y. Mizukoshi, R.
Oshima, Y. Maeda, and Y. Nagata, Langmuir, 15, 2733 (1999). d)
K. Okitsu, A. Yue, S. Tanabe, and H. Matsumoto, Chem. Mater.,
12, 3006 (2000). e) K. Okitsu, M. Murakami, S. Tanabe, and H.
Matsumoto, Chem. Lett., 2000, 1336. f) Y. Mizukoshi, T.
Fujimoto, Y. Nagata, R. Oshima, and Y. Maeda, J. Phys. Chem. B,
104, 6028 (2000). g) Y. Mizukoshi, E. Takagi, H. Okuno, R.
Oshima, Y. Maeda, and Y. Nagata, Ultrason. Sonochem., 8, 1
(2001). h) K. Okitsu, A. Yue, S. Tanabe, and H. Matsumoto, Bull.
Chem. Soc. Jpn., 75, 449 (2002).
20 E. A. Neppiras, Phys. Rep., 61, 159 (1980).
21 a) C. Krishna, T. Kondo, and P. Riesz, J. Chem. Phys., 93,
5166 (1989). b) Y. T. Didenko, and S. P. Pugach, Ultrason.
Sonochem., 1, 9 (1994). c) F. R. Young, J. Acoust. Soc. Am., 60,
100 (1976). d) W. B. McNamara Ⅲ, Y. T. Didenko, and K. S.
Suslick, Nature, 401, 772 (1999).
22 “Kagaku Binran,” Maruzen, Japan (1984), Vol. Ⅱ, p. 71.
23 “Kagaku Binran,” Maruzen, Japan (1984), Vol. Ⅱ, p. 159.