Applied Surface Science 368 (2016) 16–26

Contents lists available at ScienceDirect

Applied Surface Science

journal homepage: www.elsevier.com/locate/apsusc

Enhanced active aluminum content and thermal behaviour of

nano-aluminum particles passivated during synthesis using thermal
plasma route
Vikas L. Mathe a,∗ , Vijay Varma a , Suyog Raut a , Amiya Kumar Nandi b , Arti Pant b ,
Hima Prasanth b , R.K. Pandey b , Sudha V. Bhoraskar a , Asoka K. Das c
a
Department of Physics, Savitribai Phule Pune University, Pune 411007, Maharashtra, India
b
High Energy Materials Research Lab, Sutarwadi, Pune 411021, Maharashtra, India
c
Utkal University, VaniVihar, Bhubaneswar, Odisha 751004, India

a r t i c l e i n f o a b s t r a c t

Article history: Here, we report synthesis and in situ passivation of aluminum nanoparticles using thermal plasma reac-
Received 17 July 2015 tor. Both air and palmitc acid passivation was carried out during the synthesis in the thermal plasma
Received in revised form reactor. The passivated nanoparticles have been characterized for their structural and morphological
18 December 2015
properties using X-ray diffraction (XRD) and transmission electron microscopy (TEM) techniques. In
Accepted 25 January 2016
order to understand nature of passivation vibrational spectroscopic analysis have been carried out. The
Available online 28 January 2016
enhancement in active aluminum content and shelf life for a palmitic acid passivated nano-aluminum
particles in comparison to the air passivated samples and commercially available nano Al powder (ALEX)
Keywords:
Nano-aluminum
has been observed. Thermo-gravimetric analysis was used to estimate active aluminum content of all the
Passivation samples under investigation. In addition cerimetric back titration method was also used to estimate AAC
Thermal plasma and the shelf life of passivated aluminum particles. Structural, microstructural and thermogravomateric
Active aluminum content analysis of four year aged passivated sample also depicts effectiveness of palmitic acid passivation.
Thermo gravimetric analysis © 2016 Elsevier B.V. All rights reserved.

1. Introduction to stabilize them, therefore, the particle surface is subjected to

passivation. Passivation is an integral stage associated with the
During the last few decades, interest in nanophase materi- production of nano powder. It largely determines their chemical
als has been steadily increasing. One class of such materials is stability and activity in oxidation processes.
nanocrystalline powders, which, in many cases, exhibit significant Review of the literature indicates that in most of the cases, alu-
enhancement in a property when compared with micron-sized minum nano powders synthesized by electric-explosion of wire
particles or larger. For example, aluminum powder, typically method exhibits passivation of oxide films with an amorphous or
micron-sized, is commonly added to explosives, propellants, a crystalline structure. Oxide films on aluminum nano particles
and pyrotechnic compositions to improve their performance by (ANP) are formed during their slow oxidation in air in the course of
increasing their burn rates and enhancing their detonation prop- passivation. The films can have different thicknesses and chemical
erties. However, aluminum nanopowders have superior properties compositions, depending on the method of production and pas-
in this regard and show significant enhancement in the reactiv- sivation of powders. The thickness, composition, and structure of
ity [1]. Aluminum particles and their oxidation mechanisms have the passive layer on the particles play an important role during
become a focus of ongoing interest due to the huge heat of reaction its oxidation because these parameters determine the diffusion
associated with formation of aluminum oxide, namely, 31 kJ/g alu- rate of the gaseous oxidizer at low temperatures and, hence, the
minum burned. Usually, micro- and nano-sized aluminum powders ignition temperature. The oxide component is an energetic dead
immediately after synthesis in inert medium are highly pyrophoric; matter in combustion of powders being part of the compositions. It
has been observed that the weight content of the aluminum oxide
for most ANPs exceeds 10% and even it may extend up to more
∗ Corresponding author. than 50% for very small nano particles, whereas reported micro-
E-mail address: [email protected] (V.L. Mathe). sized powders oxide content is less than 5% [1–9]. For instance,

http://dx.doi.org/10.1016/j.apsusc.2016.01.246
0169-4332/© 2016 Elsevier B.V. All rights reserved.
 V.L. Mathe et al. / Applied Surface Science 368 (2016) 16–26 17

the weight content of metallic aluminum in ANPs stabilized by self 2. Experimental

passivating oxide/nitride layer, which will get oxidized and evolve
energy; is as high as 70–90% and rest of the part may remain un 2.1. Thermal plasma reactor set up
reacted or lost in self passivation. This seems to the drawbacks
of self passivation onto particles. Oxide layer significantly lowers The setup used for the synthesis and in situ passivation of nano-
the nano particle’s energy density, slows the nano particle com- aluminum particles is an indigenous DC transferred arc thermal
bustion rate, may prevent complete aluminum consumption. As a plasma reactor, which was developed and modified in the labo-
result of these factors, it is desirable to develop alternative meth- ratory and used to synthesize a variety of materials. Schematic
ods of passivation such as capping the aluminum particle with an and photograph of the intact thermal plasma reactor is shown in
organic layer [4,5,10–15]. Electrical explosion of wire, sonochemi- Fig. 1(a–b), respectively. The plasma reactor consists of a plasma
cal synthesis, laser ablation are known methods used to synthesis torch, with two gas inlets; one through plasma torch, emerging
aluminum nano particles. The moment these nano particles come out through the gap between a cathode and primary anode and
in contact with air/oxygen/nitrogen/water gets passivated at the the second to feed ambient gas (present sidewise). RF voltage is
cost of loss of metallic Al content. It should be noted that depending applied between the cathode and primary anode to achieve the
upon the methodology used to synthesis nano-aluminum particles, gaseous breakdown, leading to the formation of plasma. Plasma so
passivation strategy also changes and hence the metallic aluminum formed, eventually directed on a secondary anode which is sub-
content in a nano particle. jected to sufficiently high DC voltage. This leads to the formation
Several groups have used varieties of non-oxide passivating of the plasma plume having a length of about 5–10 cm and diam-
layer to stabilize metallic Al content and enhance the shelf life. eter of about 1 cm. The core temperature of plasma plume is on
Gromov et al. reported that inorganic passivating coatings for the order of 10,000 K whereas the temperature at the periphery
aluminum nano particles were demonstrated to be more advanta- is on the order of 1000 K. When such a high temperature plasma
geous compared to organic coatings, however, organic passivation plume impinged onto the secondary anode, which is aluminum
enhances shelf life significantly as compared to oxide passivated subjected on to water cooled graphite electrodes; starts evaporat-
n-Al [3,16]. Young-Soon Kwon et al. reported that the surface ing. The presence of thermal gradient between plasma zone and
passivation of aluminum nano powders by coatings of different surrounding ambient gas leads to the rapid quenching of the evap-
chemical origins leads to the enhanced powders properties suit- orated species, resulting in the formation of nano particles. ANP so
able for application in energetic systems [11]. It was noted that formed are then deposited on the inner wall of the thermal plasma
for all organic coated particles, the organic layer was penetrable reactor.
to atmospheric oxygen under the storage and resulted in inter-
nal oxide layer formation, whereas the most effective results are 2.2. System modification for in situ passivation of ANP
obtained for “dry”-passivated by boron, air and CO2 [11]. Gro-
mov et al. reported an important advantage of non-inert-coated Inside the plasma reactor a preventer plate was introduced as
aluminum nano particles is their expected higher combustion shown schematically in Fig. 2(a). Fig. 2(b) shows a photograph of
enthalpy compared to Al2 O3 -passivated Al particles [3]. Chung preventer plate taken from bottom side. The preventer plate facili-
et al. reported passivation of aluminum nano particles using alkyl- tates effective passivation of ANP. It was fixed in the top portion, at
substituted epoxides, where an alternative capping scheme for 10 cm from the centre of the plasma reactor, parallel to the base. On
producing stable aluminum nano particles with diameters on the the top of a preventer plate a metallic channel was kept, concentric
order of 20 nm is explained [13]. Park et al. investigated alu- with the torch having a radius of 10 cm. The preventer plate con-
minum nano particles with a thin carbon coating synthesized sists of holes carved in the inner circular area of metallic channel,
using either a laser induced plasma or a DC plasma-arc and concentric with the plasma torch. The position of preventer plate
found that carbon coating was thermally stable up to at least and hence the channel consisted of passivant in the solution form,
900 ◦ C, and that the particles were stable to oxidation below about is optimized one, from the view point of obtaining the passivant
800 ◦ C [17]. molecules without dissociation due to thermal effects of plasma
Thermal plasma, a versatile source of energy, an emerging tool inside the plasma reactor. The guided motion of ANP to the upward
which can be efficiently used for the synthesis of nano particles direction is achieved by suction through the top port of the plasma
of metals and metal oxides at bulk scale. Research efforts in this reactor. The passivation of ANP occur, once these are entered into
area are directed towards establishing control on phase, size, shape, the top portion of the reactor, where a metallic channel concentric
surface structure, purity of nano structurs, etc. A thermal plasma with plasma torch is kept. Exposure of ANP in air without organic
reactor comprised of a plasma torch, which can produce plasma passivating agent, results into air passivation of the ANP.
plume of about 5–10 cm in length and about 1 cm in diameter, and The setup described above, shown in Fig. 2 can be used for pas-
double walled water cooled chamber. A plasma torch enclosed in sivating polymeric material such as oleic acid, palmitic acid or any
double walled water cooled chamber can be used to tailor temper- other organic material onto the aluminum nano particles evapo-
ature profile and gas environment around the plasma plume and rated from plasma zone. Melting point of palmitic acid is 63 ◦ C,
hence it is possible to have deterministic control on the proper- which is suitable with the temperature gradient of the thermal
ties of final product as evidenced in our earlier reports [18–22]. plasma reactor at the location of the passivant channel (Fig. 2a). It is
To our knowledge, there are no reports on in situ passivation of optimized with the synthesis of ANP, i.e. when synthesis of ANP get
nano Al particles by non oxide, organic layer in transferred arc ther- started, palmitic acid starts to evaporate and fill in the passivating
mal plasma reactor. Hence, in the present studies, attempts have zone. ANP are guided in to the passivating zone. The passivation
been made in view of obtaining enhanced metallic Al content alu- process involves vapor phase condensation of the passivant on the
minum nano particle with extended shelf life by in situ passivation surface of ANP. The passivated particles deposit on the wall of the
of nAl particles by long chain fatty acid via vapor phase polymer- thermal plasma reactor and then collected manually from the wall.
ization process, inside high temperature transferred arc thermal
plasma reactor. Transmission electron microscopy, X-ray diffrac- 2.3. Synthesis of passivated ANP
tion, Fourier transform infrared spectroscopy and thermal analysis
techniques have been used to the investigation oxidation behaviour In order to obtain effective passivation of palmitic acid on to alu-
of nAl particles. minum particles, Operating pressure, power, flow rate of the gas,
18 V.L. Mathe et al. / Applied Surface Science 368 (2016) 16–26

Fig. 1. Indigenously designed and developed thermal plasma reactor. (a) schematic and (b) photograph of thermal plasma reactor.

position and diameter of annular channel were varied inside the Table 1
Synthesis parameters used to obtain in situ passivated ANP using DC transferred arc
plasma. It was observed that when the position of the channel was
thermal plasma reactor.
at the bottom, close to plasma plume, the plamitic acid got dis-
sociated due to high temperature of plasma plume. Therefore, the Sr. no. Parameters Value
channel was moved away from the plasma plume and underneath 1 Current 150 Amp
of channel a preventor plate was introduced. In addition, the evap- 2 Voltage 30 Volt
orated aluminum nano particles were guided by means of suction 3 Operating Pressure 375 Torr
4 Torch Gas (Flow Rate) Argon (@ 5 lit/min) + Nitrogen (@ 3 lit/min)
through the top ports with the help of rotary pump as mentioned
5 Ambient Gas (Pressure) Argon (250 Torr)
in the previous section. The optimized parameters are summarized 6 Run time 15 min
in Table 1.
Thermogravimetric analyzer (Model: Shimandzu TGA 60H)
was used to study oxidation behavior passivated ANP. All TGA microscope (model TECHNAI 200 G2 ) was used to record morphol-
plots were recorded at the heating rate of 10 ◦ C/min in oxygen ogy and extent of passivation onto ANP. BET surface analyzer (Smart
ambient with the gas flow rate of 50 ml/min. X-ray diffractome- Instruments, Mumbai, India, model Smart Sorb 91) was used to
ter (Bruker model Advance D-8) having Cu-K␣ X-ray radiation estimate the surface area. FTIR spectrometer (JASCO model FT/IR
was used to record diffraction patterns. Transmission electron 6100) was used to record the spectra in transmission mode to

Fig. 2. Setup for synthesis of in situ passivated ANP in the thermal plasma reactor.
 V.L. Mathe et al. / Applied Surface Science 368 (2016) 16–26 19

0 500 1000 1500 2000 2500 3000 3500 4000

110
100 ALEX
90
80
70 (c)
60
IPANP:PA
100

%T
80

60
(b)
40 IPANP:Air
100

80

60 (a)

Fig. 3. XRD patterns plots of in situ passivated ANP by palmitic acid (IP-n-Al:pa), 0 500 1000 1500 2000 2500 3000 3500 4000
in situ passivated ANP by air (IP-n-Al:air), Alex ANP and standard XRD lines of
aluminium and alumina for comparative studies. Wavenumber (cm-1)
Fig. 4. FTIR spectra of in situ passivated ANP by palmitic acid (IP-n-Al:pa), in situ
understand bonding aspect and nature of passivation layer of the passivated ANP by air (IP-n-Al:air) and Alex ANP.
nano prarticles under investigation. Cerimetric titration was used
to analyze metallic aluminum content in support of TGA data. In synthesis prevents formation of such amorphous oxide layer on the
order to investigate shelf life of passivated aluminum nano parti- surface of ANP, hence enhancement in AAC can be observed.
cles accelerated aging studies have been carried out. To carry out
accelerated aging studies, Al nano particles were subjected to 90% 3.2. FTIR analysis
humidity at 60 ◦ C. 50 mg of such sample was taken out after every
two hours and metallic aluminum was determined by cerimetric Vibrational spectra of passivated samples of nano-aluminium
back titration method. In situ passivation for selected batches with provide the required information about the bonding details of
high AAC was then repeated to confirm the reproducibility. Once the passivated surfaces. Fig. 4 compares the FTIR spectra of Air
reproducibility is confirmed, further structure and morphological passivated ANPs with that of palmitic acid passivated ANPs. FTIR
analysis were performed by XRD, TGA and TEM techniques. absorption spectrum of ALEX sample is used as a reference. Promi-
nent broad absorption band observed in all the samples around
3. Results 904 cm−1 and weak absorption peaks around 1270 and 1278 cm−1 ,
are attributed to [25,26] Al-O bond and indicate the surface oxida-
3.1. XRD analysis tion of aluminum nano particles. Apart from this, IP-n-Al:air sample
shows –OH stretching vibrations around 3596 cm−1 and –OH bend-
Fig. 3 shows XRD plots of in situ passivated ANP by air (IP-n- ing vibrations around 1697 cm−1 due to absorption of water on the
Al:air), in situ passivated ANP by palmitic acid (IP-n-Al:pa) and surface of air passivated samples.
commertially available Alex ANP. At the bottom of Fig. 3 standard In case of IP-n-Al:pa a prominent additional features are
XRD lines of aluminium and alumina are given for comparative observed at 2849 cm−1 and 2929 cm−1 corresponds to CH2 sym-
studies. Formation of n-Al phase was confirmed from XRD lines of metric and asymmetric stretching vibrations respectively. In
IP-n-Al:air and IP-n-Al:pa by comparison with the standard alu- addition, passivated sample also show weak, but prominent,
minum diffraction lines using JCPDS data (PDF# 85-1327). XRD absorption peak at 1469 cm−1 which correlates with the CH2
lines observed at 38.4◦ , 44.7◦ , 65.1◦ and 78.2◦ confirm formation and CH3 deformation vibrations of aluminum palmitate [26,27].
of n-Al phase. XRD data of Alex is similar to that of samples synthe- Another peak observed at 1588 cm−1 is originating from COO
sized in the present investigation. Sharp diffraction lines in the XRD asymmetric stretching vibrations in Al-Palmitate also confirm the
plot demonstrates highly crystalline nature of the samples. Average formation of such functional group on the surface of IP-n-Al:pa.
size of the crystallite was determined by using Scherer’s formula ALEX also indicates the presence of organic coating which is
[30], which was found to be around 27 nm for both the samples. In inferred from the broad absorption band around 1540 cm−1 arising
addition to the diffraction lines of metallic aluminum phase, weak from C O asymmetric stretching vibration in saturated aliphatic
diffraction lines assigned to Al2 O3 [JCPDS data (PDF# 77-0396) compounds. Table 2 gives a summary of the observed absorption
phase are observed. Percentage abundance of crystalline aluminum peaks and their possible assignments.
oxide was determined by quantitative phase analysis method
[23,24] using the expression % COXIDE = 100IAl /(IAl + 1.33IOxide ). The 3.3. BET surface area analysis
oxide content obtained from XRD analysis was about 3% in both
the samples, indicating presence of crystalline Al2 O3 phase. How- Surface area analysis was performed by using the BET surface
ever, air passivated ANP shows greater amount of amourphous area anlyzer. The surface area so obtained was further used for
oxide, which can be verified from TGA and cerimatric analysis. This average particle size determination. Surface area and particle size,
amorphous oxide layer is not protective and leads to futher diffu- of various ANP samples obtained from BET analysis are given in
sion of oxygen when exposed to air. Surface passivation during the Table 3.
20 V.L. Mathe et al. / Applied Surface Science 368 (2016) 16–26

Table 2 considering the oxidation reaction of aluminum

Absorption peaks observed in Fig. 4 and their possible assignments.
2Al + (3/2)O2 → Al2 O3 . Implying, AAC by TGA = m × [atomic
Name of the Observed Assignment Reference weight of 2Al/atomic weight of (3/2)O2 ] = m × 1.124.
sample absorption Careful observation of Fig. 5 reveals that oxidation behavior
(cm−1 )
of the nano particles under consideration is significantly differ-
IP-n-Al:pa. ∼(904, 1270, Al-O group stretching 25,26 ent. It is seen from Fig. 5 that the overall weight gain observed
1278) CH2 and CH3 group 26,27 for IP-n-Al:pa sample is highest when compared with that of IP-n-
1469 deformation vibrations 27
Al:air and ALEX sample. IP-n-Al:pa sample possesses 84% of weight
1588 in Al-Palmitate 27
2649 COO asymmetric gain which corresponds to 94% AAC. IP-n-Al:air possesses 69% of
2919 stretching in weight gain corresponding to 77.5% AAC whereas ALEX sample
Al-Palmitate show 60% weight gain corresponding to 67% AAC. High AAC of IP-
CH2 symmetric stretch
n-Al:pa is attributed to the effective passivation and demonstrate
in palmitic acid
CH2 asymmetric an increase in the non-oxidised metallic nano-aluminum content.
stretch in palmitic acid Careful analysis of TGA plots reveals various processes involved
IP-n-Al:air ∼ (904, 1270, Al-O group stretching 25,26 in the oxidation of nano-aluminum particles. Oxidation of nano Al
1278) OH bending and 25,26 particles is mainly governed by (i) chemical kinetics leading to the
∼(1697, 3596) stretching vibrations
formation of surface oxide shell and (ii) diffusion of oxygen leading
arising from water
adsorption to the oxidation of core aluminum which are very well discussed
ALEX ∼ (904, 1270, Al-O group stretching in the literature [25,26].
1278) OH bending and Trunov et al. [27] reported effect of variation of particle size of
∼(1697, 3596) stretching vibrations,
ANP on the oxidation and melting behaviour of nano-aluminum
1540 arising from water
adsorption
particles, concluding no correlation between the melting and oxi-
C O asymmetric dation process. Decrease in the size of the aluminum particles to
stretching vibration in nanometer sized particles result in to the transition of the oxida-
saturated aliphatics tion process from diffusion controlled to a kinetically controlled
process [8]. Enhancement in the surface to volume ratio of nano-
aluminum particles lowers the melting temperature than that for
200
micron sized aluminum particles. Rai et al. [28] studied the mech-
IP-n-Al:pa
Alex ANP anism of change in oxidation rates below and above the melting
180 IP-n-Al:air temperature of aluminum. Below the melting temperature of core
aluminum the oxidation process is governed by the slow diffusion
Weight Change (%)

160 of oxygen through the oxide shell whereas above melting point
enhanced oxidation rate was observed, which is attributed to dif-
fusion of both aluminum and oxygen through the oxide shell. Above
140
melting temperature molten core aluminum bursts out and get
oxidized which lead to formation of hollow alumina particles, as
120 reported earlier by our group [23].

100 3.5. Transmission electron microscopy (TEM) analysis

0 200 400 600 800 1000 1200 TEM analysis of ANP was performed by using “TECHNAI G2
Temperature ( C)
o
20U- TWIN (FEI, Netherlands)”, operated at 200 kV. Particle size
distribution, average particle size, % mass contribution, thickness
Fig. 5. TGA data of in situ passivated and Alex ANP. of passivating layer of as synthesized air passivated ANP and IP-n-
Al:pa were studied using TEM technique.
3.4. Thermo-gravimetric analysis
3.5.1. IP-n-Al:air samples
Fig. 5 show the TGA plots of IP-n-Al:air, IP-n-Al:pa and Alex ANP. TEM pictures of IP-n-Al:air sample are shown in Fig. 6(a and
TGA plots in Fig. 5 demonstrate a weight loss up to 440 ◦ C, which b) which depict spherical particles having wide size distribution.
is due to the loss of adsorbed gases, water vapors and passivant Both the images are of the same sample, which were used to deter-
from the surface of aluminum nano particles. Weight loss observed mine size distribution of the particles. Fig. 6(c) shows the TEM
in case of ALEX sample was less than that of nano Al powder image of single aluminium nano particle in which core shell fea-
synthesized using thermal plasma. Weight gain starts around ture of the particle is clearly seen. The shell thickness is found
500 ◦ C, which continues to 1044 ◦ C with an intermediate step to be 5 nm. Selected area electron diffraction pattern analysis for
around 620 ◦ C. More details are summarized in Table 4, along with single aluminum particle is depicted in Fig. 6(d). Comparison of
oxidized metallic aluminum content (i.e. active aluminum content, ‘d’ values with JCPDS data (JCPDS# 85-1327) of aluminum con-
AAC) determined from TGA plots [25]. To find out AAC from TGA firms the formation of aluminum. A diffused ring corresponding
data, total weight gain m of the sample was used, obtained by to (3 1 1) plane of aluminium oxide is also observed. The diffused

Table 3
BET surface area and particle size of synthesized ANP compared with ALEX ANP.

Sr. no. ANP sample Specific surface area, ABET (m2 /g) Particle size (nm), dBET = 6/(ABET × 2.7) [4]

1 IP-ANP:air 25 90
2 IP-ANP:pa 15 140
3 ALEX ANP 20 100
 V.L. Mathe et al. / Applied Surface Science 368 (2016) 16–26 21

Fig. 6. TEM analysis of air passivated ANP. (a and b) TEM images used for the determination of particle size distribution. (c) TEM image of air pasivated aluminum nano
particle which depicts presence of core shell structure with shell thickness of 5 nm. (d) Selected Area Electron Diffraction Pattern of IP-n-Al:air, marked planes are confirmed
from JCPDS # 85-1327. (e) Particle size distribution plotted, with the % mass contribution in the following table. (f) Line profile analysis for the thickness measurement of
oxide shell, which is evident from the sudden change in the intensity around 5 nm, implying oxide shell thickness = 5 nm.
22 V.L. Mathe et al. / Applied Surface Science 368 (2016) 16–26

Table 4
TGA data of aluminum nano particles.

Sr. no. TGA sample m (%) AAC (%) = m × 1.124 [25] Temperature (◦ C)

Start Saturation-I Saturation-II

1 IP-n-Al:pa 84 94 28 615 942

2 IP-n-Al:air 69 77.5 32 620 968
3 ALEX ANP 60 67 31 640 1044

nature of the ring indicates amorphous nature of alumina. Fig. 6(e) evident from the sudden change in the intensity count of Fig. 7(d).
show a particle size distribution for air passivated nano-aluminum The passivating layer is expected to reduce oxidation of aluminum
particles with average particle size of 46 nm. Air passivated nano- as palmitic acid prevents direct diffusion of oxygen into aluminum
aluminum particles samples demonstrated spherical morphology. and hence enhance the shelf life.
Thickness of the oxide layer was found to be 5 nm, as determined
from the line profile analysis of air passivated nano-aluminum par- 3.6. Cerimetric titration for the determination of AAC
ticles [8,23,28–31] samples described in Fig. 6(f), which shows a
good agreement with TEM image shown in Fig. 6(c). In cerimetric titration method aluminum is made to react with
It is seen from Fig. 6(f) that majority of the particles possesses cerric sulphate CeSO4 which is back titrated with Mohar salt
size less than 50 nm, however there are few particles of large size (i.e. ammonium ferrous sulphate solution: (NH4 )2 Fe(SO4 )·6H2 O) to
up to about 200 nm. As the active aluminum content depends upon obtain metallic Al content in a given powder. Here cerric sulphate
core mass of nano-aluminum particles to larger extent; percentage being strong oxidizer aluminum gets fully oxidized and hence it is
mass contribution of various particle size range was analyzed and possible to determine actual active aluminum content in a given
summarized in the table given in inset of Fig. 6. It is observed about powder provided there is no any passivatig layer which hinders
34% mass is possessed by the particles having size less than 100 nm full oxidation of aluminum nano particles [6,23,25,33,34]. Usually,
and about 22% mass is possesses by the particles having size in the cerimetric titration method is found to be suitable for air passi-
range of 100–150 nm. It is well documented in the literature that vated samples rather than organically passivated n-Al particles.
smaller the size of particles, greater the reactivity and hence more After optimization this method was adopted for the determination
oxidation of aluminum [6]. For example, aluminum nano particles of AAC of all synthesized samples.
having size of 10 nm are oxidized up to about 30% which leads to Fig. 8 compares the results of the AAC obtained by TGA and
less active aluminum content hence, IP-n-Al:air sample possesses cerimetric titrations methods. It is seen from Fig. 8(a) that values
low AAC as compared to IP-n-Al:pa. of AAC for IP-n-Al:air samples obtained using cerimetric titration
method are higher than that of TGA method even for four samples of
3.5.2. IP-n-Al:pa nano-aluminum prepared in different batches using same synthe-
Fig. 7 show TEM images of IP-n-Al:pa samples. Spherical mor- sis parameters. For IP-n-Al:air sample average AAC by TGA analysis
phology of IP-n-Al:pa samples is evident from the TEM images of is 78.75% whereas it is 84% obtained using cerimetric titration
Fig. 7(a and b) which are agglomerated due to surface passivation method. However, a reverse trend is observed in case of IP-n-Al:pa
of nano-aluminum particles by palmitic acid. Even though majority sample as shown in Fig. 8(b). For IP-n-Al:pa it is evident that aver-
of the particles possesses spherical morphology, one dimension- age AAC obtained using TGA method is 94%, whereas it is 79.75% for
ally grown structures are also seen prominently. Also majority of cerimetric titration method. The lower content of active aluminum
the particles possesses size less than 100 nm like that of IP-n-Al:air content in case of IP-n-Al:pa sample is attributed to incomplete oxi-
sample, the number of particles having size greater than 100 nm dation nano Al particles due to its effective passivation by palmitic
has increased as compared with IP-n-Al:air sample. Fig. 7(c) shows acid.
the TEM image of aluminum nano particle passivated by palmitic
acid. A core shell structure is clearly seen in Fig. 7(c) with a shell 3.7. Accelerated aging test
of palmitic acid having thickness of 10 nm. In order to examine
closely the morphology of the particles, only one of two particles Formation of surface oxide layer on nano-aluminum particles is
was focused under transmission electron microscope. Line profile greatly influenced in presence of humidity, and temperature, which
pictures of individual particles as shown in Fig. 7(d and e) con- increases aging. In the limited time period of the experiments,
firms the fact that aluminum particles are passivated by palmitic method of accelerated aging is adopted, which is mainly a function
acids which are seen to be agglomerated. Line profile analysis of of temperature at fixed higher humidity [15]. In accelerated aging
Fig. 7(d), reveals that the thickness of passivating layer is 10 nm on a test samples under investigation are exposed to humidity (90%) at
150 nm sized particle, evident from the sudden change in the inten- fixed temperatures and active aluminum content was determined
sity count near 10 nm, which is in agreement with the TEM image using cerimetric back titration method. Accelerated aging studies
shown in Fig. 7(c). Intensity count for aluminum are low than the were carried out for the samples subjected to 90% humidity at 60 ◦ C,
palmitic acid, which is due to the high mass absorption coefficient of for 10 h [5,10,35]. About 50 mg nano-aluminum sample was taken
aluminum. Periodic change in the intensity of Fig. 7(d) qualitatively out periodically after two hours up to 10 h and active aluminum
demonstrates presence of more than one particle. SAED analysis of content was determined using cerrimetric titration method. It is
the blue squared area of Fig. 7(e) for IP-n-Al:pa sample is presented seen from Fig. 9 that AAC content decreases significantly for IP-n-
in Fig. 7(f). Diffracting planes of aluminum and ␣-palmitic acid are Al:air and ALEX sample whereas it remains fairly constant for for
indicated by Al for aluminum peaks and pa for palmitic acid. Com- IP-n-Al:pa sample which indicates effectiveness of passivation.
parison of SAED data with JCPDS data of aluminum Al (JCPDS #
85-1327) and palmitic acid (␣-Palmitic Acid ref. JCPDS # 08-0787) 3.8. Aging study
confirms presence of both the phases. The observation of crystalline
palmitic acide using TEM is in agreement with the literature [32]. The passivated samples were stored at normal temperature and
Thickness of the passivating layer was found to be 10 nm, which is pressure conditions in open atmosphere for the period of four years
 V.L. Mathe et al. / Applied Surface Science 368 (2016) 16–26 23

Fig. 7. TEM analysis of IP-ANP:pa. (a and b) TEM images used for the determination of particle size distribution. (c) TEM image of IP-ANP:pa sample indicating core shell
structure with shell thickness of 10 nm. (d) Line profile analysis for the determination of thickness of the passivation. The sudden change in the intensity count near 10 nm,
implies 10 nm thickness of the passivation. (e) Line profile analysis of the agglomerated particles. (f) Selected Area Electron Diffraction (SAED) pattern of the blue square area
marked in Fig. 6(b), depicted planes are confirmed from XRD data of Al (Ref. JCPDS # 85-1327) and PA (␣-PalmiticAcid Ref. JCPDS # 08-0787). (g) particle size distribution
plotted, with the % mass contribution in the following table.
24 V.L. Mathe et al. / Applied Surface Science 368 (2016) 16–26

Fig. 8. Comparison of active aluminum content in both the samples, IP-n-Al:air and IP-n-Al:pa obtained by cerimetric and TGA analysis for four samples synthesized using
same operating parameters which reveals reproducibility of the results.

99 Alex ANP inferred from FTIR data and is an evidence of passivation of palmitic
IP-n-Al:pa acid onto nano-aluminum. The palmitic acid passivation protects
96 IP-n-Al:air aluminum nano particle from adsorption of OH group and hence
93
Active Aluminum Content (%)

slows down the oxidation process.

Enhanced AAC of IP-n-Al:pa demonstrated by its nearly com-
90
plete oxidation, attributed to the nearly constant diffusion of
87 oxygen through the oxide shell. Continuous weight gain indicates
conversion of aluminum (Al = 2.7 g/cm3 ), to alumina (Al2 O3 =
84
3.78 g/cm3 ). Trunov et al. [29] described the phase change and
81 growth in aluminum oxide from amorphous phase to crys-
78 talline ␥ phase, and crystalline ␥ phase to ␣ phase. TGA plot
of as-synthesized IP-n-Al:pa (Fig. 5) demonstrate the formation
75 of aluminum oxide (2Al + (3/2)O2 → Al2 O3 ), which starts around
72 500 ◦ C. The weight gain (m) confirms the % amount of metallic
aluminum converted to aluminum oxide, which is m × [atomic
0 2 4 6 8 10 weight of 2Al/atomic weight of (3/2)O2 ] = m × 1.124 [25]. For
Aging Time (hour)
IP-n-Al:pa samples, phase transition to ␥-alumina occurs around
615 ◦ C and phase transition to ␣-alumina occurs around 942 ◦ C.
Fig. 9. Accelerated aging test of ANP. Beyond 1000 ◦ C growth of hollow alumina [23] takes place, which
is evident from the reduction in % weight. ␥-alumina phase is com-
paratively more porous phase than ␣-alumina phase. This leads
in order to study the aging effect. An aging effect has been studied to the continous diffusion of ambient oxygen, contributing to the
using XRD, TEM and thermogravimetric analysis. Fig. 10(a–c) shows growth and conversion from ␥-phase to ␣-phase of alumina, which
XRD, TEM images and TGA plots of aged IP-n-Al:pa and IP-n-Al:air continous till 942 ◦ C, hence continous increase in the % weight is
sample. observed. During the in situ passivation process of IP-n-Al:pa sam-
ples, the formation ␣-phase alumina around ANP is prevented by
4. Discussions the passivation of palmitic acid, which is confirmed from the XRD
analysis of IP-n-Al:pa (Fig. 3). It results in the continous and com-
X-ray diffraction data of the freshly prepared samples show plete oxidation of IP-n-Al:pa samples. Consquently higher AAC of
presence of crystalline aluminium for both, air passivated as well IP-n-Al:pa is the combined result of availability of higher metallic
as palmitic acid passivated samples alongwith small content of aluminum and prevention of the formation of ␣-alumina shell on
alumina. An estimate of metallic aluminium content based on ANP cores, both are due to in situ passivation.
thermogravaometric analysis suggests that alongwith crystalline It is seen from Fig. 6(c) that core is lighter and the shell is
alumina amorphous alumina is also present. Eventhough TEM pic- darker whereas Fig. 7(c) indicates that core is darker and shell is
tures of frershly prepared aluminum nano particles does not show lighter. The contract observed is due to difference in mass absorp-
clear signature of amorphous alumina, SAED pattern (Fig. 6d) shows tion coefficient of core and shell. Particle size distribution obtained
the presence of diffused diffraction ring corresponding to amor- from TEM images shown in Fig. 7(a–c), is summarized in the his-
phous alumina phase. The amorphous oxide layer is not protective togram depicted in Fig. 7(g). Particle size distribution is found to
and leads to further diffusion of oxygen when exposed to air. be in the range of less than 10 nm to about 250 nm with maxi-
Passivation of nano-aluminum by palmitic acid is expected to mum number of particles having size of 60 nm. Even though the
protect oxidation of aluminum. Passivation of nano-aluminum by number of particles having size less than 100 nm are more in num-
palmitic acid results into the formation of aluminum palmitate as ber, mass of the particle having size more than 100 nm is more.
 V.L. Mathe et al. / Applied Surface Science 368 (2016) 16–26 25

Fig. 10. (a) X-ray diffraction patterns, (b) and (c) TEM images of IP-n-Al:air, (d) thermogravometric plots of passivated samples (e) and (f) TEM images of of IP-n-Al:pa, after
aging for the period of 4 years.

60.4% mass of the IP-n-Al:pa is in the range from 100 to 200 nm for IP-n-Al:air samples. In case of IP-n-Al:pa sample thickness of
and only 17% mass is possessed by the particles having size less passivating layer was found to almost same and no signature of
than 100 nm. For IP-n-Al:air sample the mass of the particles hav- crystalline alumina is observed. Thermogravametric analysis of the
ing size less than 100 nm is double than that of IP-n-Al:pa sample, aged samples indicates that the weight gain observed for IP-n-Al:pa
i.e. 34%. The particles having size less than 100 nm are expected to sample is almost double than that of IP-n-Al:air sample in the mea-
oxidize to greater extent due high reactivity. TGA analysis of the sured temperature range. However, this weight gain is less than
IP-n-Al:pa sample indicates maximum oxidation during TGA mea- that of freshly prepared IP-n-Al:pa particles which indicates incom-
surements [mIP-ANP:pa = 84% → AACIP-ANP:pa = 94%, ref table (4)] in pleat oxidation of aluminum nano particles or transformation of
the temperature range of 500–1050 ◦ C. The enhanced oxidation is part of metallic aluminium content into amorphous alumina due
attributed to protection provided by in situ passivation of palmitic to diffusion of oxygen through the layer of palmitic acid.
acid which leads to high metallic aluminum content as compared
to IP-n-Al:air sample.
It is noted from Fig. 5 that the weight gain in case of IP-n-Al:pa 5. Conclusions
is almost continuous with a small hump around 600 ◦ C unlike IP-
n-Al:air or ALEX sample. The continuous weight gain above 500 ◦ C, Aluminum nano particles have been synthesized using ther-
i.e. continuous oxidation is due to continuous diffusion of oxygen mal plasma route. It was possible to in situ passivate aluminum
towards core aluminum and diffusion of core aluminum towards nano particles in the thermal plasma reactor. Both air and palmitc
oxygen at the periphery. acid passivation was carried out during the synthesis in the ther-
Cerimetric titration also supports the inferences based on mal plasma reactor. Characterizations of the samples using X-ray
thermogravometric analysis for oxide passivated nano particles, diffraction (XRD), Fourier Transform Infrared Spectroscopy (FTIR),
whereas it shows disagreement for the IP-n-Al:pa which is due to transmission electron microscopy (TEM) and BET surface area
presence of palmitic acid as a passivating layer. Accelarated aging analyzer demonstrated highly crystalline spherical particles. X-
test carried out for all the sample under investigation indicates ray diffraction data analysis confirms the single phase formation
fairly constant value of active aluminum content as a function of whereas there is no significant change in the crystal structure after
aging time for IP-n-Al:pa unlike IP-n-Al:air and Alex which sup- passivation except small signature of aluminum oxide phase. FTIR
ports effective passivation of nano-aluminum particles by palmitic analysis of the synthesized product confirms the passivation of Al
acid. nano particles by palmitic acid. TEM analysis reveals that air pas-
It is interesting to report the aging behavior of IP-n-Al:pa and sivated nano-aluminum possesses 5 nm oxide layer on to 50 nm
IP-n-Al:air. It is seen from the XRD data of the passivated nano- aluminum core whereas palmitic acid passivated aluminum pos-
aluminum particles that air passivated nano particles allow the sesses 10 nm palmitic acid shell onto the 150 nm aluminum core.
growth of crystalline alumina whereas palmitic acid passivated It is observed that passivation of ANP by palmitic acid shift the
nano particles does not allow growth of crystalline alumina. The size distribution to higher side which supports enhancement in
growth of crystalline alumina was found to grow in preferred ori- active aluminum content. Thermo-gravimetric analysis demon-
entation. TEM images analysis of the aged samples indicates that strate higher AAC of 94% for palmitic acid passivated aluminum
thickness of oxide layer has slightly increased and signature of particles. In situ passivated nano-aluminum particles were found
crystalline alumina is seen from HRTEM images and SAED pattern to be fairly stable even in a 90% humidity environment at 60 ◦ C
26 V.L. Mathe et al. / Applied Surface Science 368 (2016) 16–26

for 10 h. Detailed analysis of aged passivated samples also depicts [16] A.A. Gromov, Stabilization of metal nanoparticles—a chemical approach,
palmitic acid passivation on to metallic aluminum is effective one. Chem. Eng. Technol. (2009) 1049–1060, http://dx.doi.org/10.1002/ceat.
200900022.
[17] K. Park, A. Rai, M.R. Zachariah, Characterizing the coating and size-resolved
Acknowledgements oxidative stability of carbon-coated aluminum nanoparticles by
single-particle mass-spectrometry, J. Nanoparticle Res. 8 (2006) 455–464,
http://dx.doi.org/10.1007/s11051-005-9012-2.
The author are thankful to HEMRL, Pune for funding the research [18] A.B. Nawale, N.S. Kanhe, K.R. Patil, S.V. Bhoraskar, V.L. Mathe, A.K. Das,
project. S.V. Bhoraskar is also thankful to UGC, New Delhi for the Magnetic properties of thermal plasma synthesized nanocrystalline nickel
award of emeritus professor fellowship. The author acknowledges ferrite (NiFe2 O4 ), J. Alloys Compd. 509 (2011) 4404–4413, http://dx.doi.org/
10.1016/j.jallcom.2011.01.057.
Savitribai Phule Pune University, Pune for providing finantial sup- [19] S.A. Raut, N.S. Kanhe, S.V. Bhoraskar, A.K. Das, V.L. Mathe, Thermal plasma
port. processed ferro-magnetically ordered face-centered cubic iron at room
temperature, J. Appl. Phys. 116 (2014) 163913, http://dx.doi.org/10.1063/1.
4899244.
References [20] S. Karmakar, A.B. Nawale, N.P. Lalla, V.G. Sathe, S.K. Kolekar, V.L. Mathe, et al.,
Gas phase condensation of few-layer graphene with rotational stacking faults
[1] M.M. Mench, K.K. Kuo, C.L. Yeh, Y.C. Lu, Comparison of thermal behavior of in an electric-arc, Carbon N.Y. 55 (2013) 209–220, http://dx.doi.org/10.1016/j.
regular and ultra-fine aluminum powders (Alex) made from plasma explosion carbon.2012.12.029.
process, Combust. Sci. Technol. 135 (1998) 269–292, http://dx.doi.org/10. [21] I. Banerjee, S. Karmakar, N.V. Kulkarni, A.B. Nawale, V.L. Mathe, A.K. Das, S.V.
1080/00102209808924161. Bhoraskar, Effect of ambient pressure on the crystalline phase of nano TiO2
[2] A. Gromov, A. Ilyin, U. Förter-Barth, U. Teipel, Characterization of aluminum particles synthesized by a dc thermal plasma reactor, J. Nanoparticle Res. 12
powders: II. Aluminum nanopowders passivated by non-inert coatings, (2010) 581–590, http://dx.doi.org/10.1007/s11051-009-9627-9.
Propellants Explos. Pyrotech. 31 (2006) 401–409, http://dx.doi.org/10.1002/ [22] N.S. Kanhe, A.B. Nawale, R.L. Gawade, V.G. Puranik, S.V. Bhoraskar, A.K. Das,
prep.200600055. et al., Understanding the growth of micro and nano-crystalline AlN by
[3] A.A. Gromov, U. Förter-Barth, U. Teipel, Aluminum nanopowders produced by thermal plasma process, J. Cryst. Growth 339 (2012) 36–45, http://dx.doi.org/
electrical explosion of wires and passivated by non-inert coatings: 10.1016/j.jcrysgro.2011.11.011.
characterisation and reactivity with air and water, Powder Technol. 164 [23] N.V. Kulkarni, S. Karmakar, S.N. Asthana, A.B. Nawale, A. Sheikh, S.P. Patole,
(2006) 111–115, http://dx.doi.org/10.1016/j.powtec.2006.03.003. et al., Study on growth of hollow nanoparticles of alumina, J. Mater. Sci. 46
[4] A.A. Gromov, A.P. Il’in, U. Foerter-Barth, U. Teipel, Effect of the passivating (2011) 2212–2220, http://dx.doi.org/10.1007/s10853-010-5059-2.
coating type, particle size, and storage time on oxidation and nitridation of [24] N. Wu, T. Wu, Enhanced phase stability for tetragonal zirconia in precipitation
aluminum powders, Combust. Explos. Shock Waves 42 (2006) 177–184, synthesis, J. Am. Ceram. Soc. 83 (2000) 3225–3227.
http://dx.doi.org/10.1007/s10573-006-0036-4. [25] L. Chen, W. Song, J. Lv, X. Chen, C. Xie, Research on the methods to determine
[5] Y. Li, W. Song, C. Xie, D. Zeng, A. Wang, M. Hu, Influence of humidity on the metallic aluminum content in aluminum nanoparticles, Mater. Chem. Phys.
thermal behavior of aluminum nanopowders, Mater. Chem. Phys. 97 (2006) 120 (2010) 670–675, http://dx.doi.org/10.1016/j.matchemphys.2009.12.030.
127–131, http://dx.doi.org/10.1016/j.matchemphys.2005.07.064. [26] S. Wang, Dynamical effects of the oxide layer in aluminum nanoenergetic
[6] T.D. Fedotova, O.G. Glotov, V.E. Zarko, Application of cerimetric methods for materials, Propellants Explos. Pyrotech. 30 (2005) 148–155, http://dx.doi.org/
determining the metallic aluminum content in ultrafine aluminum powders, 10.1002/prep.200400097.
Propellants Explos. Pyrotech. 32 (2007) 160–164, http://dx.doi.org/10.1002/ [27] M.A. Trunov, S.M. Umbrajkar, M. Schoenitz, J.T. Mang, E.L. Dreizin, Oxidation
prep.200700017. and melting of aluminum nanopowders, J. Phys. Chem. B 110 (2006)
[7] P. Puri, V. Yang, Effect of particle size on melting of aluminum at nano scales, 13094–13099, http://dx.doi.org/10.1021/jp0614188.
J. Phys. Chem. C 111 (2007) 11776–11783. [28] A. Rai, K. Park, L. Zhou, M.R. Zachariah, Understanding the mechanism of
[8] Y. Huang, G.A. Risha, V. Yang, R.A. Yetter, Effect of particle size on combustion aluminium nanoparticle oxidation, Combust. Theory Model. 10 (2006)
of aluminum particle dust in air, Combust. Flame 156 (2009) 5–13, http://dx. 843–859, http://dx.doi.org/10.1080/13647830600800686.
doi.org/10.1016/j.combustflame.2008.07.018. [29] M.A. Trunov, S.M. Umbrajkar, M. Schoenitz, J.T. Mang, E.L. Dreizin, Oxidation
[9] J. Gesner, M.L. Pantoya, V.I. Levitas, Effect of oxide shell growth on and melting of aluminum nanopowders, J. Phys. Chem. B 110 (2006)
nano-aluminum thermite propagation rates, Combust. Flame 159 (2012) 13094–13099.
3448–3453, http://dx.doi.org/10.1016/j.combustflame.2012.06.002. [30] Y. Li, R.K. Kalia, A. Nakano, P. Vashishta, Size effect on the oxidation of
[10] C. Dubois, P.G. La, C. Roy, É.P. De Montréal, Q. Hc, Polymer-grafted metal aluminum nanoparticle: Multimillion-atom reactive molecular dynamics
nanoparticles for fuel applications, J. Propuls. Power 23 (2007) 651–658, simulations, J. Appl. Phys. 114 (2013) 134312-1–134312-10.
http://dx.doi.org/10.2514/1.25384. [31] V.Y. Gertsman, Q.S.M. Kwok, TEM investigation of nanophase aluminum
[11] Y.S. Kwon, A.A. Gromov, J.I. Strokova, Passivation of the surface of aluminum powder, Microsc. Microanal. 11 (2005) 410–420.
nanopowders by protective coatings of the different chemical origin, Appl. [32] R. Benages, L. Bayés, R. Cordobilla, E. Moreno, T. Calvet, M.A. Cuevas-Diarte, A
Surf. Sci. 253 (2007) 5558–5564, http://dx.doi.org/10.1016/j.apsusc.2006.12. comparative study of several techniques to obtain fatty acid nanoparticles:
124. palmitic acid, Cryst. Growth Des. 9 (2009) 1762–1766, http://dx.doi.org/10.
[12] L. Guo, W. Song, M. Hu, C. Xie, X. Chen, Preparation and reactivity of 1021/cg8008128.
aluminum nanopowders coated by hydroxyl-terminated polybutadiene [33] S. Liu, M. Ye, A. Han, X. Chen, Preparation and characterization of energetic
(HTPB), Appl. Surf. Sci. 254 (2008) 2413–2417, http://dx.doi.org/10.1016/j. materials coated superfine aluminum particles, Appl. Surf. Sci. 288 (2014)
apsusc.2007.09.043. 349–355, http://dx.doi.org/10.1016/j.apsusc.2013.10.031.
[13] S.W. Chung, E.a. Guliants, C.E. Bunker, D.W. Hammerstroem, Y. Deng, M.a. [34] P.A. Jelliss, S.W. Buckner, S.W. Chung, A. Patel, E.A. Guliants, C.E. Bunker, The
Burgers, et al., Capping and passivation of aluminum nanoparticles using use of 1,2-epoxyhexane as a passivating agent for core e shell aluminum
alkyl-substituted epoxides, Langmuir 25 (2009) 8883–8887, http://dx.doi.org/ nanoparticles with very high active aluminum content, Solid State Sci. 23
10.1021/la901822h. (2013) 8–12, http://dx.doi.org/10.1016/j.solidstatesciences.2013.06.
[14] A.A. Gromov, Y.I. Strokova, a.a. Ditts, Passivation films on particles of 001.
electroexplosion aluminum nanopowders: a review, Russ. J. Phys. Chem. B 4 [35] A. Sossi, E. Duranti, C. Paravan, L.T. Deluca, A.B. Vorozhtsov, A.A. Gromov,
(2010) 156–169, http://dx.doi.org/10.1134/S1990793110010239. et al., Non-isothermal oxidation of aluminum nanopowder coated by
[15] B.J. Thomas, C.E. Bunker, E.A. Guliants, S.E. Hayes, A. Kheyfets, K.M. Wentz, hydrocarbons and fluorohydrocarbons, Appl. Surf. Sci. 271 (2013) 337–343,
et al., Synthesis of aluminum nanoparticles capped with copolymerizable http://dx.doi.org/10.1016/j.apsusc.2013.01.197.
epoxides, J. Nanopart. Res. 15 (2013) 1729, http://dx.doi.org/10.1007/s11051-
013-1729-8.