Colloids and Surfaces A 535 (2017) 96–104

Contents lists available at ScienceDirect

Colloids and Surfaces A

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

Application of Turbiscan in the homoaggregation and heteroaggregation of MARK

copper nanoparticles
⁎
Xuejiao Qi, Ya’nan Dong, Hongtao Wang , Chen Wang, Fengting Li
State Key Laboratory of Pollution Control and Resource Reuse, Key Laboratory of Yangtze River Water Environment, Ministry of Education, College of Environmental
Science and Engineering, Tongji University, Shanghai 200092, China

G RA P H I C A L AB S T R A C T

A R T I C L E I N F O A B S T R A C T

Keywords: With the development of industries, copper nanoparticles (Cu NPs) have been abundantly discharged into
Turbiscan natural water and may threaten the safety of aquatic environments. The stability (such as homoaggregation and
Copper nanoparticles heteroaggregation) of Cu NPs in aqueous phase may affect their toxicity. Turbiscan, including three kinds of data
Homoaggregation processing methods (transmitted intensity (T), variation of average transmitted intensity (ΔT) and Turbiscan
Heteroaggregation
stability index (TSI), were used to investigate the homoaggregation and heteroaggregation of Cu NPs with humic
Kaolin
acid (HA) and kaolin in aqueous phase. T and TSI were used to analyze Cu NPs-kaolin and Cu NPs-HA-kaolin
systems, respectively, whereas T and ΔT were used to analyze Cu NPs and Cu NPs-HA systems. Results showed
that the stability of the system is influenced by the dissolution and sedimentation of Cu NPs, and the aggregation
and sedimentation of Cu NPs, HA and kaolin. When pH is 4, the dissolution of Cu NPs is the main factor affecting
the system stability. Kaolin may reduce the stability of system by sedimentation or impeding the dissolution of
Cu NPs. When pH is 8, the aggregation and sedimentation of Cu NPs mainly affect the system stability. Kaolin
renders the system unstable by promoting the aggregation of Cu NPs. In addition, HA improves the stability of
the system by inhibiting the aggregation of Cu NPs and kaolin when pH = 4 and 8. Ionic strength reduces the

⁎
Corresponding author.
E-mail addresses: hongtao@tongji.edu.cn, wanght010@gmail.com (H. Wang).

http://dx.doi.org/10.1016/j.colsurfa.2017.09.015
Received 19 June 2017; Received in revised form 9 September 2017; Accepted 11 September 2017
Available online 12 September 2017
0927-7757/ © 2017 Elsevier B.V. All rights reserved.
X. Qi et al. Colloids and Surfaces A 535 (2017) 96–104

stability of system by condensing electric double layer. Therefore, Turbiscan can be used to study both homo-
aggregation and heteroaggregation in a relatively long period (12 h), and three kinds of data processing methods
can be applied based on the properties of the samples.

1. Introduction Turbiscan was used as the main method to analyze the homoaggrega-
tion and heteroaggregation of Cu NPs under different conditions. The
Nanoparticles (NPs), such as copper nanoparticles (Cu NPs), tita- effects of pH, IS, HA and kaolin were studied. We selected the most
nium oxide nanoparticles (TiO2 NPs), zinc oxide nanoparticles and applicable data processing method to study the aggregation behavior of
cerium oxide nanoparticles have been widely used in worldwide in- Cu NPs under different environment conditions.
dustries, such as in cosmetics, lubricating oils and pesticides [1,2].
Approximately 10%-30% NPs were discharged into natural water in 2. Materials and methods
Asia, 3%-17% in Europe and 4%-19% in North America in 2013[3]. The
amount of Cu NPs and CuO NPs used has recently reached 200 t/a [4]. 2.1. Chemicals
Cu NPs may threaten the security of environment and human health
[5], and can be toxic to fungus [6], aquatic and terrestrial plants [7,8], Cu NPs were purchased from Jingchun Biochemical Technology
invertebrates [9], marine worms [10], mussels [10] and clams[11]. The Company, Shanghai, China (Aladdin Reagent, C103843–50 G). The
toxicity and removal efficiency of Cu NPs are related to its fate and particle size of Cu NPs varies from 10 nm to 30 nm, and the purity is
transportation in water and are affected by the concentration, ionic over 99%. Cu NP stock solution (500 mg/L) was prepared by adding
strength (IS), pH and presence of organic matter [12,13]. Therefore, it 0.5 g Cu NPs into 1 L distilled water with ultrasonic treatment for
is necessary to investigate the migration and transformation of Cu NPs, 20 min. The Cu NP concentration used in this study was 10 mg/L,
including stability, aggregation, dissolution and precipitation, as well as which was prepared by diluting the stock solution using distilled water.
the effects of ionic strength, pH and organic matters. HA was purchased from Sigma Aldrich Company (AR, 53680-10G).
Environmental conditions, including pH, IS, natural organic matters The main components and characteristics of HA used in this study were
(NOM) and other natural particles (such as kaolin and hematite), affect previously reported [42–44]. The main elements of HA are S, C, Fe and
the stability and aggregation of NPs [14–19]. The pH affects the sta- N. The weight-average molecular weight and number-average mole-
bility of NPs by influencing their electrical property [16,20]. High IS cular weight of HA are 20,032 and 9787, respectively [42–44]. HA
may improve the aggregation of NPs by condensing the thickness of the stock solution was prepared as follows: HA powder was first added into
electric double layer [14,16]. NOM and clay particles can be adsorbed distilled water, and then the pH was adjusted to about 11.00 using
on the surface of NPs and affect the surface charge, thereby changing 0.1 mol/L NaOH solution. Magnetic stirring was performed for 24 h to
the stability [21–24]. Humic acid (HA) is a typical NOM, that affects the improve HA dissolution. Finally, the solution was filtered by 0.22 μm
aggregation behavior of several NPs [25]. HA can reduce the removal membrane and was stored at 4 °C. All the HA in the stock solution can
efficiency of TiO2 NPs and CuO NPs through electrostatic force of at- pass through the membrane. The total organic carbon (TOC) of the
traction, steric hindrances, and bridging effect [23,26]. HA possesses 10 mg/L HA solution is about 4.5 mg/L. The desired TOC in this study
various structures and may cause different influences on the aggrega- (0.1 mg/L, 1 mg/L and 10 mg/L) was achieved through dilution.
tion of NPs [27]. Kaolin, a clay material commonly found in natural Kaolin was purchased from Sinopharm, Shanghai, China (AR,
water, can affect the aggregation behavior of NPs because of its varying 20020528). Given that IS was controlled by adding NaCl in this study,
electrical properties under different conditions [28]. In a previous removing calcium was necessary to avoid interference. Sodium-mod-
study, kaolin was either sodium-modified to remove the calcium con- ified method [29] was used to treat kaolin, and the process is shown in
tent [29] or was full dispersed [30]. Fig. S1. The experimental concentration of kaolin is 50 mg/L.
To investigate the homoaggregation and heteroaggregation of NPs, NaCl was purchased from Sinopharm, Shanghai, China (AR). The
three methods, namely, UV −spectrophotometry [31–33], dynamic desired concentrations (0, 0.1 and 1 mol/L) of NaCl were prepared by
light scattering (DLS) [15,24,34] and Turbiscan Tower have been used adding NaCl into distilled water. Three kinds of buffer, including
[35,36]. UV-spectrophotometry is often used to analyze samples with NaCH3COO-HCH3COO (pH = 4, 5, 6), Na2HPO4-NaH2PO4 (pH = 7)
characteristic peaks and proper concentration range. It’s usually used to and H3BO3-Na2B4O7 (pH = 8, 9), were used to control the pH in this
study the homoaggregation in a single-phase system [14,16,18,37,38]. study. All reagents used to prepare the buffer were purchased from
For heteroaggregation study, DLS is usually used to investigate the Sinopharm, Shanghai, China (AR).
aggregation between NPs and kaolin by measuring the particle size.
This method is applicable to systems wherein the size of NPs increases
slowly and stabilizes within a short time, but is not appropriate for the 2.2. Experiment 1: measurement of dissolved and sediment quantity and
measurement of large particles [22]. However, the particle size in- hydrodynamic diameter
creases very quickly within a short time for most heterogeneous sys-
tems. Turbidimeter and laser diffraction (LD) can also be used to In the absence of HA, 10 mg/L Cu NP suspension was prepared, and
measure the stability of particles by measuring the turbidity and par- was filtered through 0.22 μm membrane to obtain the supernatant after
ticle size, respectively [39–41]. Turbidimeter can only measure tur- 2 h. The concentration of Cu NPs in the supernatant was measured to
bidity, and sometimes it’s difficult to analyze stability based on tur- find the concentration of dissolved copper. The dissolved quantity
bidity [41]. LD can measure the particle size, and the heteroaggregation (DQCu) of Cu NPs was calculated as Eq. (1). In the presence of HA, the
of nanoparticles can be determined according to the changes in their sample was digested before measurement. To measure the total con-
particle size. However, a continuous change in particle size cannot be centration of dissolved and suspending copper, the unfiltered suspen-
measured by LD [39,40]. Turbiscan Tower can analyze the samples by sion was digested after 2 h both in the absence and presence of HA. The
evaluating the transmission light or back-scattering light. This device sediment quantity (SQCu) of Cu NPs was calculated as Eq. (2).
scans the samples every 40 μm and can obtain the data at different
DissolvedquantityofCuNPs(%)
heights of the sample. Turbiscan can also be used to conduct long
concentration of dissolved Cu
(> 12 h) continuous monitoring. =
In this study, we tested Turbiscan on a typical ternary system. total concentration of Cu in initial suspension (1)

97
X. Qi et al. Colloids and Surfaces A 535 (2017) 96–104

SedimentquantityofCuNPs(%) appropriate to use when the transmitted intensity changes over height
concentraion of dissolved and suspending Cu is small.
=1− TSI can be calculated as formula (5):
total concentration of Cu in initial suspension (2)

For measuring hydrodynamic diameter, Cu NPs were dispersed into ∑h Scani (h) − Scani−1 (h)
water through ultrasonic dispersion, and the concentration is 10 mg/L.
TSI = ∑ H
i (5)
DLS was used to measure the hydrodynamic diameter.
where H: Sample height from bottom of the cell to the meniscus;
2.3. Experiment 2: preparation of samples for turbiscan measurement Scani-1(h): intensity of scanning light (time is i-1, height is h).
TSI is usually used for binary systems, especially the unstable
Buffer of different pH levels (pH = 4–9) and stock solutions of Cu system. The higher TSI value, the more unstable the system is.
NPs, NaCl, HA and kaolin were prepared. Before measurement, the In conclusion, the T value implies the changes in transmission in-
stock solution of Cu NPs was dispersed again by ultrasonic dispersion. tensity over time and height. When the change in transmitted intensity
When prepare the sample, diluted water, buffer, NaCl, HA or kaolin over height is small, ΔT value would be suitable. TSI is usually used for
were first added into the tube, then Cu NPs was added into the tube, binary systems, especially the unstable system.
and the concentration of Cu NPs should be 10 mg/L. And then put the
tube into the instrument to test immediately. Aggregation of Cu NPs 2.5. Characterization and software
began as soon as Cu NPs was added into the tube. It takes about 4 s from
the addition of Cu NPs to the beginning of the measurement. Malvern Zetasizer (Nano ZS90, UK) was used to measure the zeta
potential. The morphologies of Cu NPs and kaolin were characterized
2.4. Analyzing instrument- turbiscan by transmission electron microscopy (TEM, TEM-2011, Japan).
Inductively coupled plasma optical emission spectrometer (ICP-OES,
Turbiscan Tower (Formulaction, France) was used to investigate the Optima 2100 DV, USA) was used to measure the concentration of
mechanism and process of homoaggregation and heteroaggregation in copper after digestion. The 2100Q Protable Turbidimeter (HACH, USA)
this study. Multiple light scattering is the fundament of Turbiscan [45]. was used to measure the turbidity of kaolin suspension. The Shimadzu
Two kinds of signal, including transmission light and back-scattering TOC V-CPN (TOC-LCPH, Japan) was used to measure the TOC of water
light, could be achieved (Fig. 1) [45]. In general, signal value of containing HA. DLS (DLS, Malvern, UK) was used to measure the hy-
transmission light was used to analyze clear liquids, whereas that of drodynamic diameter.
back-scattering light was used to analyze non-clear liquids or samples ANOVA [46] was used to evaluate the statistical significance among
with high concentration. Transmission light was used in this study different results. These results were affected by different environmental
because the samples were clear liquid. These samples were prepared in conditions, such as IS, pH value and HA concentration.
a 20 mL glass vial with a height of 42 mm. The signal value was ob-
tained every 40 μm of the sample, and once takes about 20 s. 3. Results and discussion
There are three kinds of data processing methods for the signal
value of transmission light, include transmitted intensity (T), variation 3.1. Characterization of Cu NPs and kaolin
of average transmitted intensity (ΔT) and Turbiscan stability index
(TSI). T can be calculated as formulas (1)- (2): TEM results of Cu NPs and kaolin are shown in Fig. S2. The particle
2ri
size of Cu NPs and kaolin is about 50–100 nm and 200 nm, respectively.
T(λ, ri) = T0 e− λ (1) DLS results of Cu NPs and kaolin suspension are shown in Fig. S3. The
hydrodynamic diameter of Cu NPs and kaolin was measured to be
2d around 122–190 nm and 200 nm, respectively.
λ(d,ϕ) =
3ϕQs (2) The zeta potential of Cu NPs and kaolin are shown in Fig. 2. Fig. 2(a)
where T0: Transmittance of continuous phase; shows that when pH varied from 4 to 9, the Cu NPs were always ne-
ri: Vial inner radius; gatively charged when IS = 0. The absolute zeta potential decreased
λ: mean free-path of photon; with the increase in IS, and Cu NPs became highly unstable. This is due
d: Average particle size; to the condensation of electrical double layer at high IS [47]. Fig. 2 (b)
Ø: Volume concentration of dispersion phase; shows that the isoelectric point (pHIEP) of kaolin is about 4.5. It was
Qs: Optical Parameters according to Mie theory. reported that the Si-O surface of kaolin is negatively charged, whereas
T is affected by the average diameter of particles and the mean free- the Al-O surface of kaolin is positively charged when pH is less than
path of photon. According to T, we can obtain the changes in trans-
mission intensity over time and height.
ΔT can be calculated using the following formulas (3)-(4):
Hu
∑Hl Scani (h) − Scan 0 (h)
Ti =
Hu − Hl (3)

ΔT = Ti − T0 (4)

where Ti: average transmitted intensity (time is i);

Hu: upper limit;
Hl: lower limit;
Scani (h): intensity of scanning light (time is i, height is h);
Scan0 (h): intensity of scanning light (time is 0, height is h);
T0: average intensity of transmission. Fig. 1. Schematic diagram and parameters of Turbiscan (Noted: part from Turbisoft-LAB
user guide).
ΔT shows the changes in average transmitted intensity, and is

98
X. Qi et al. Colloids and Surfaces A 535 (2017) 96–104

Fig. 2. Zeta potential of (a) Cu NPs and (b) kaolin at different pH.

(pHIEP). However, both Si-O and Al-O surfaces carry negative charges the bottom. ΔT and T are combined to analyze the homoaggregation of
when pH > (pHIEP) [29]. Cu NPs when pH = 7, 8, 9.
The effects of pH and IS on Cu NPs stability are shown in Fig. 3
3.2. Homoaggregation of Cu NPs and the effects of HA (a)–(c). Fig. 3(a) shows that at IS = 0 and pH=4–6, ΔT increased
sharply within the first 2.5 h, indicating the instability of the system.
3.2.1. Homoaggregation of Cu NPs The increase became sharp when the pH is low. ΔT became constant
To investigate the homoaggregation of Cu NPs, the changes in after 2.5 h, revealing that the system stabilized. Fig. 3(d) shows that
transmitted intensity (T) of Cu NPs at different height of samples over more than 75% Cu NPs dissolved after 2 h when pH = 4, 5 or 6.
time were analyzed (Fig. S4). It was shown that when pH = 4–6, T Therefore, the destabilization of Cu NPs system within 2 h was caused
changed over time, but the differences between different heights of by the dissolution of Cu NPs. In addition, about 18% Cu NPs were se-
samples at the same time were small. Therefore, when pH = 4–6, the dimented when pH = 4, 5, 6. The homoaggregation of Cu NPs may
average transmitted intensity can represent the changes in the different occur during sedimentation. When the pH was 7, 8 or 9, ΔT increased
height of the sample, and ΔT was used to study the homoaggregation of gradually during the entire process (until 12 h). However, T increased
Cu NPs. When pH = 7–9, T at the top of tube was higher than that at with the increasing height of the tube at the same time when pH = 7, 8

Fig. 3. Effects of pH and IS on stability and dissolved and sediment quantity of Cu NPs. Changes of stability under different conditions: (a) [NaCl] = 0 M; (b) [NaCl] = 0.1 M; (c)
NaCl = 1.0 M; Dissolved and sediment quantity under different conditions: (d) [NaCl] = 0 M; (e) [NaCl] = 0.1 and 1.0 M (sedimentation time = 2 h).

99
X. Qi et al. Colloids and Surfaces A 535 (2017) 96–104

or 9 (Fig. S4). This is because of the sedimentation of Cu NPs. Fig. S4 increased gradually again (Fig. 3(b) and (c)). The level of increase or
(d), (e) and (f) show sedimentation occurred during the first 8 h. When decrease was higher when [NaCl] = 1.0 M. This is because the increase
pH was 7, 8 or 9, the Cu NP system was more chemically stable than of IS contributed to the condensing electric double layer and resulted in
when pH = 4, 5, 6 because of the low DQCu under this condition. Al- the decrease of the absolute value of zeta potential [16], which causes
though Fig. S4 and Fig. 3(d) show sedimentation occurred when the quick aggregation of Cu NPs and the formation of lots of large
pH = 7, 8, 9, the process was slow. Therefore, ΔT increased gradually particles. When [NaCl] = 0.1 and 1.0 M, aggregation is the dominant
with the homoaggregation and sedimentation of Cu NPs. In addition, factor that affects ΔT during the first 4 h. In the first 30 min, homo-
the zeta potential of Cu NPs is less than −32 mV when the pH was 7, 8 aggregation occured and some large particles sedimented, which
and 9, as shown in Fig. 2(a). Therefore, Cu NPs system was more stable caused the increase of ΔT. However, 30 min later, as more and more
when pH = 7, 8, 9 than at acidic condition. large particles formed, sedimentation became very slowly because the
Fig. 3(d) and (e) show that the dissolved quantity of Cu NPs is about tube is filled with many too much large particles. The slow sedi-
70%- 80% under acidic condition (pH = 4, 5 and 6), and that under the mentation reduced the transmission intensity and ΔT. After 3 or 4 h,
condition of pH = 7, 8 and 9 is about 3%. This finding indicates that Cu sedimentation became the main factor affecting transmission intensity.
NPs tend to dissolve much easier and faster under acidic condition than The transmission intensity and ΔT increased with the sedimentation of
under pH ≥ 7. According to Pourbaix diagram for Cu, the Cu2+ form large particles. Finally, the aggregation and sediment occurred gradu-
prevails when pH < 7, whereas Cu(OH)2 (S), Cu (S) and Cu2O (S) ally, and ΔT increased slowly.
prevail when pH ≥7. With the decrease in pH, more H+ will react with
Cu and produce Cu2+. Therefore, Cu NPs tend to dissolve at an acidic 3.2.2. Effects of HA on homoaggregation of Cu NPs
pH and not when pH ≥ 7, and the dissolved quantity of Cu NPs in- Section 3.3.1 shows that the rules of dissolution, sedimentation and
creased with the decreasing pH. aggregation of Cu NPs were similar when pH = 4, 5, 6 or pH = 7, 8, 9.
When the concentration of NaCl is 0.1 M or 1.0 M, ΔT increased Therefore, the following experiments were performed at the condition
faster than IS = 0 (Fig. 3 (b) and (c)).When pH = 4, 5, 6, ΔT became of pH = 4 and 8.
constant within 2 h ([NaCl] = 0.1 M) or 1.5 h ([NaCl] = 1.0 M). Fig. 4 shows the changes of ΔT of Cu NPs in the presence of HA at
Fig. 3(d) indicates that DQCu decreased and SQCu increased with the the condition of pH = 4 and 8 and IS = 0 M, 0.1 M, 1.0 M NaCl. When
increase in IS. Hence, ΔT was affected by dissolution, aggregation and pH = 4, ΔT was lower in the presence of HA than HA = 0 M. As shown
sedimentation of Cu NPs [48]. The dissolution of Cu NPs played a in Fig. S5 (a), when the concentration of HA = 0 M, 0.1 M, 1 M and
dominant role in the changes in ΔT when pH was 4, 5 and 6. When 10 M, DQCu was 79.2%, 76.6%, 73.3% and 72.9%, and SQCu was
pH = 7, 8 and 9, ΔT increased sharply within 30 min, decreased, and 16.8%, 16.4%, 15.7% and 14.4%, respectively. Therefore, the addition

Fig. 4. Changes of ΔT and zeta potential at the presence of HA at different pH and IS (a)-(f) changes of ΔT over time: (a) pH = 4, IS = 0 M; (b) pH = 4, IS = 0.1 M; (c) pH = 4,
IS = 1.0 M; (d) pH = 8, IS = 0 M; (e) pH = 8, IS = 0.1 M; (f) pH = 8, IS = 1.0 M. (g)-(h) changes of zeta potential (g) pH = 4; (h) pH = 8.

100
X. Qi et al. Colloids and Surfaces A 535 (2017) 96–104

of HA may restrict the dissolution, and also restrict aggregation or se- changes of ΔT.
diment of Cu NPs [48–51]. Although both Cu NPs and HA are nega- IS also affects the stability of HA-Cu NPs system when pH = 4 and
tively charged (Fig. 2(a), Table S1), Cu NPs interact with HA (Fig. S6). 8. Fig. S5 shows that addition of NaCl restricted the dissolution and
The zeta potential of Cu NPs- HA system is shown in Fig. 4 (g). The improved the sediments by condensing electric double layer [14,16]. As
absolute value of zeta potential of the system increased with increasing shown in Fig. 4(g) and (h), the absolute value of zeta potential de-
concentration of HA, which means stability of the system was improved creases with the increasing of IS. Therefore, the addition of IS may
resulting in the decrease of ΔT. When pH is 8, ΔT is much lower in the reduce the stability of the whole system.
presence of HA than HA = 0 M. It is also because the addition of HA The ANOVA analysis results show that the effects of HA is not sig-
improved the stability of Cu NPs system by restricting the dissolution nificant (p > 0.05), but the influences of IS is significant (p < 0.05)
and aggregation (Fig. S5 (b)) and changing electric properties both for pH = 4 and 8. Therefore, the effect of IS on aggregation of
(Fig. 4(h)). nanoparticles is more significant than that of HA.
Fig. 4(f) indicates that ΔT increased sharply during the first 30 min,
decreased from 30 min to 3 h, and increased again after 3 h. This is also
3.3. Heteroaggregation of Cu NPs and kaolin and the effects of HA
caused by the addition of NaCl, which contributes to the condensed
electric double layer. The reason is the same with Cu NPs without HA
3.3.1. Heteroaggregation of Cu NPs and kaolin
(Fig. 3(b) and (c)). During the first 30 min, homoaggregation occurred
Fig. 5 shows the changes of transmitted intensity of kaolin-Cu NPs
and some large particles sedimented causing the increase of ΔT. From
system at different height over time. As shown in Fig. 5(b), transmitted
30 min to 3 h, the sediment process was inhibited because too many
intensity of kaolin-Cu NPs system varied a lot at different heights.
large particles have filled the tube. Three hours later, the sediment of
Therefore, T was chosen to study the stability of kaolin-Cu NPs system.
particles caused the gradual increase of ΔT. Therefore, the addition of
Fig. 5(a) indicates that when pH = 4, the color of the curves changes
NaCl led to the quick aggregation and slow sediment, and caused the
uniformly so that the variation rate at different heights was almost the

Fig. 5. Changes of T of Cu NPs at different height over time (a) pH = 4; (b) pH = 8.

101
X. Qi et al. Colloids and Surfaces A 535 (2017) 96–104

same. When pH = 4, most of Cu NPs were dissolved (DQCu = 75.48%), −37.00 ± 0.02, −45.67 ± 0.85 and −43.27 ± 0.55 mV when
while only 18.0% Cu NPs and 32.0% kaolin sedimented. Therefore, it pH = 8, respectively (Table S2). Electrostatic attractive force does not
was a homogeneous system. However, when pH = 8, the colors turned contribute to the heteroaggregation because both Cu NPs and kaolin are
non-uniformly (Fig. 5(b)). When reaction time was the same, T was negatively charged. With the increase of particle concentration, parti-
higher at higher sample height, which indicated that the aggregation cles are closer to each other and Brownian motion of particles becomes
and sediment occurred. When pH = 8, DQCu is only 4.4%, and 45.3% weaker [52,53]. Cu NPs may be adsorbed on kaolin and aggregation
Cu NPs and 35.2% kaolin sedimented. Therefore, Cu NPs and kaolin may happen, which leads to the decrease of stability of the system and
aggregated and the particle size increased and then sedimented to cause TSI increased after addition of kaolin.
the differences of T. IS also affected the stability of Cu NPs-kaolin system. As shown in
Because Cu NPs-kaolin system is not a homogeneous system, ΔT Fig. 6(c) and (d), TSI increased at higher IS. IS can affect the ag-
can’t represent it well. TSI was chosen to analyze the process. Fig. 6(a) gregation of Cu NPs and kaolin by influencing the dissolution and se-
shows that when pH = 4, TSI increased fast because of the dissolution dimentation of Cu NPs. As shown in Table S3, DQCu decreased and SQCu
of Cu NPs (DQCu = 79.3%). After the addition of kaolin, TSI increased increased with the increasing of NaCl concentration when pH]4 or 8.
continually, but the growing rate was slower than Cu NPs during the Van der Waals forces between kaolin and Cu NPs due to compressing
first 2 h. One reason may be that kaolin inhibited the dissolution of Cu electric double layer and charge screening may be the main mechanism
NPs (DQCu = 75.5%) in the first 2 h causing the decrease of growing for the influence of IS [54].
rate of TSI, but it contributed to the heteroaggregation of this system.
Another possible reason may be that some Cu2+ is absorbed on kaolin. 3.3.2. Effects of HA on heteroaggregation of Cu NPs and kaolin
However, the changes of particle size of kaolin can be ignored because As shown in Fig. 7, TSI of the ternary system increased continually
of the small particle size of Cu2+. Furthermore, the change of TSI of when pH = 4 and 8. Compared with Fig. 6, the addition of HA de-
kaolin was slow. Therefore, change of transmission intensity, as well as creased TSI value. When pH = 4, SQCu and SQKaolin decreased to
TSI of Cu NPs-kaolin system grows slowly than Cu NPs during the first 14.28% and 26.4%, respectively. HA may adhere on the surface of Cu
2 h. NPs and kaolin, inhibiting the sedimentation contributing to the ag-
As shown in Fig. 6(b), the addition of kaolin resulted in the rapid gregation of Cu NPs and kaolin (Fig. S7). Therefore, the variation of TSI
increase of TSI. DQCu was measured to be only 4.4%, while SQCu and in ternary system was caused by dissolution of Cu NPs and aggregation
SQkaolin was 45.3% and 35.2%, respectively. Therefore, when pH = 8, between Cu NPs and kaolin particles at pH = 4. When pH = 8, how-
the addition of kaolin will promote the aggregation and sediment ever, the dissolution quantity of Cu NPs is lower. SQCu and SQKaolin
greatly. Zeta potential of Cu NPs, kaolin and Cu NPs- kaolin system are decreased from 45.3% and 35.2% to 39.68% and SQKaolin 30.13%,

Fig. 6. Changes of TSI of different system at different pH and IS over time (a) pH = 4; (b) pH = 8; (c) pH = 4, Cu NPs-kaolin system; (d) pH = 8, Cu NPs-kaolin system.

102
X. Qi et al. Colloids and Surfaces A 535 (2017) 96–104

Fig. 7. Changes of TSI of Cu NPs-HA-kaolin ternary system with presence of IS at different pH over time (a) pH = 4; (b) pH = 8.

respectively. HA impeded the sedimentation of Cu NPs, but kaolin composites with antifungal and bacteriostatic properties, Chem. Mater. 17 (2005)
5255–5262.
promoted the aggregation of Cu NPs and kaolin. Therefore, TSI varia- [2] P. Yi, J.J. Pignatello, M. Uchimiya, J.C. White, Heteroaggregation of cerium oxide
tion of ternary system was caused by aggregation between Cu NPs and nanoparticles and nanoparticles of pyrolyzed biomass, Environ. Sci. Technol. 49
kaolin particles when pH = 8. (2015) 13294–13303.
[3] A.A. Keller, A. Lazareva, Predicted releases of engineered nanomaterials: from
In addition, IS also affected the stability of the ternary system by global to regional to local, Environ. Sci. Technol. Lett. 1 (2014) 65–70.
influencing the dissolution and sedimentation of Cu NPs and con- [4] A.A. Keller, S. McFerran, A. Lazareva, S. Suh, Global life cycle releases of en-
tributing to the aggregation of Cu NPs, kaolin and HA by condensing gineered nanomaterials, J. Nanopart. Res. 15 (2013) 1–17.
[5] M.R. Wiesner, G.V. Lowry, P. Alvarez, D. Dionysiou, P. Biswas, Assessing the risks of
electric double layer [14,16], as can be seen in Fig. 7. manufactured nanomaterials, Environ. Sci. Technol. 40 (2006) 4336–4345.
[6] V. Shah, P. Dobiášová, P. Baldrian, F. Nerud, A. Kumar, S. Seal, Influence of iron
4. Conclusions and copper nanoparticle powder on the production of lignocellulose degrading
enzymes in the fungus Trametes versicolor, J. Hazard. Mater. 178 (2010)
1141–1145.
The homoaggregation and heteroaggregation affect the stability as [7] J. Shi, A.D. Abid, I.M. Kennedy, K.R. Hristova, W.K. Silk, To duckweeds (Landoltia
well as fate and transport of Cu NPs in aqueous phase. In this study, punctata), nanoparticulate copper oxide is more inhibitory than the soluble copper
Turbiscan Tower was used to investigate the stability of Cu NPs at in the bulk solution, Environ. Pollut. 159 (2011) 1277–1282.
[8] Z. Wang, X. Xie, J. Zhao, X. Liu, W. Feng, J.C. White, B. Xing, Xylem- and phloem-
different conditions. Three different data processing methods, including Based transport of CuO nanoparticles in maize (Zea mays L.), Environ. Sci. Technol.
T, ΔT and TSI were used to analyze different system based on the 46 (2012) 4434–4441.
properties of the system. The results showed that the stability of the [9] S.K. Hanna, R.J. Miller, D. Zhou, A.A. Keller, H.S. Lenihan, Accumulation and
toxicity of metal oxide nanoparticles in a soft-sediment estuarine amphipod, Aquat.
whole system is influenced by the dissolution and sediments of Cu NPs, Toxicol. 142-143 (2013) 441–446.
and the aggregation of Cu NPs, HA and kaolin. For Cu NPs system, the [10] P.E. Buffet, M. Richard, F. Caupos, A. Vergnoux, H. Perrein-Ettajani, A. Luna-
dissolution of Cu NPs was the main factor of the system stability when Acosta, F. Akcha, J.C. Amiard, C. Amiard-Triquet, M. Guibbolini, A mesocosm study
of fate and effects of CuO nanoparticles on endobenthic species (Scrobicularia
pH = 4. The sediments and aggregation of Cu NPs played the most plana, Hediste diversicolor), Environ. Sci. Technol. 47 (2013) 1620–1628.
important role in the stability when pH = 8. The addition of HA im- [11] T. Gomes, J.P. Pinheiro, I. Cancio, C.G. Pereira, C. Cardoso, M.J. Bebianno, Effects
proved the stability of the system by inhibiting the dissolution and of copper nanoparticles exposure in the mussel Mytilus galloprovincialis, Environ.
Sci. Technol. 45 (2011) 9356–9362.
aggregation of Cu NPs and kaolin. However, kaolin may reduce the [12] J.R. Conway, A.S. Adeleye, J. Gardeatorresdey, A.A. Keller, Aggregation, dissolu-
stability of the system by inhibiting the dissolution of Cu NPs when tion, and transformation of copper nanoparticles in natural waters, Environ. Sci.
pH = 4 and promoting the aggregation of Cu NPs when pH = 8. Technol. 49 (2015) 2749–2756.
[13] L.F. Wang, N. Habibul, D.Q. He, W.W. Li, X. Zhang, H. Jiang, H.Q. Yu, Copper
release from copper nanoparticles in the presence of natural organic matter, Water
Acknowledgements Res. 68 (2015) 12–23.
[14] Y. Han, D. Kim, G. Hwang, B. Lee, I. Eom, P.J. Kim, M. Tong, H. Kim, Aggregation
The research was partially supported by special fund of State Key and dissolution of ZnO nanoparticles synthesized by different methods: influence of
ionic strength and humic acid, Colloids Surfaces A Physicochem. Eng. Asp. 451
Joint Laboratory of Environment Simulation and Pollution Control (No. (2014) 7–15.
16K07ESPCT) and Science and Technology Commission of Shanghai [15] D. Zhou, A.I. Abdel-Fattah, A.A. Keller, Clay particles destabilize engineered na-
Municipality (No.15230724300). The work was also partially funded noparticles in aqueous environments, Environ. Sci. Technol. 46 (2012) 7520–7526.
[16] F. Piccapietra, L. Sigg, R. Behra, Colloidal stability of carbonate-coated silver na-
by the Foundation of State Key Laboratory of Pollution Control and noparticles in synthetic and natural freshwater, Environ. Sci. Technol. 46 (2012)
Resource Reuse, China (No. PCRRE16007). 818–825.
[17] D. Lin, S. Ma, K. Zhou, F. Wu, K. Yang, The effect of water chemistry on homo-
aggregations of various nanoparticles: specific role of Cl− ions, J. Colloid Interface
Appendix A. Supplementary data Sci. 450 (2015) 272–278.
[18] I.L. Gunsolus, M.P. Mousavi, K. Hussein, P. Bühlmann, C.L. Haynes, Effects of humic
Supplementary data associated with this article can be found, in the and fulvic acids on silver nanoparticle stability, dissolution, and toxicity, Environ.
Sci. Technol. 49 (2015) 8078–8086.
online version, at http://dx.doi.org/10.1016/j.colsurfa.2017.09.015. [19] Y. Feng, X. Liu, K.A. Huynh, J.M. McCaffery, L. Mao, S. Gao, K.L. Chen,
Heteroaggregation of graphene oxide with nanometer- and micrometer-sized he-
References matite colloids: influence on nanohybrid aggregation and microparticle sedi-
mentation, Environ. Sci. Technol. (2017) 6821–6828.
[20] A. Praetorius, J. Labille, M. Scheringer, A. Thill, K. Hungerbuehler, J.Y. Bottero,
[1] N. Cioffi, L. Torsi, N. Ditaranto, G. Tantillo, L. Ghibelli, L. Sabbatini, T. Bleve- Heteroaggregation of titanium dioxide nanoparticles with model natural colloids
Zacheo, M. D'Alessio, P.G. Zambonin, E. Traversa, Copper nanoparticle/polymer under environmentally relevant conditions, Environ. Sci. Technol. 48 (2014)

103
X. Qi et al. Colloids and Surfaces A 535 (2017) 96–104

10690–10698. kinetics using UV-Visible spectrophotometry upon discharge in complex environ-

[21] J.D. Hu, Y. Zevi, X.M. Kou, J. Xiao, X.J. Wang, Y. Jin, Effect of dissolved organic ments, Sci. Total Environ. 539 (2016) 7–16.
matter on the stability of magnetite nanoparticles under different pH and ionic [38] M. Tejamaya, I. Römer, R.C. Merrifield, J.R. Lead, Stability of citrate, PVP, and PEG
strength conditions, Sci. Total Environ. 408 (2010) 3477–3489. coated silver nanoparticles in ecotoxicology media, Environ. Sci. Technol. 46
[22] J. Labille, C. Harns, J.Y. Bottero, J. Brant, Heteroaggregation of titanium dioxide (2012) 7011–7017.
nanoparticles with natural clay colloids, Environ. Sci. Technol. 49 (2015) [39] M. Kowalska, Żbikowska, A. application of a laser diffraction method for determi-
6608–6616. nation of stability of dispersion systems in food and chemical industry, J. Disper.
[23] L. Miao, C. Wang, J. Hou, P. Wang, Y. Ao, S. Dai, B. Lv, Effects of pH and natural Sci. Technol. 34 (2013) 1447–1453.
organic matter (NOM) on the adsorptive removal of CuO nanoparticles by periph- [40] F. Wang, Y. Liu, Y. Zhang, S. Hu, Experimental study on the stability of asphalt
yton, Environ. Sci. Pollut. Res. 22 (2015) 1–9. emulsion for CA mortar by laser diffraction technique, Construct. Buil. Mater. 28
[24] H. Wang, Y.N. Dong, M. Zhu, X. Li, A.A. Keller, T. Wang, F. Li, Heteroaggregation of (2012) 117–121.
engineered nanoparticles and kaolin clays in aqueous environments, Water Res. 80 [41] Y. Zhu, A. Marchuk, M.L. Bennett, Rapid method for assessment of soil structural
(2015) 130–138. stability by turbidimeter, Soil Sci. Soc. Am. J. 80 (2016) 1629–1637.
[25] A. Philippe, G.E. Schaumann, Interactions of dissolved organic matter with natural [42] D. Fetsch, J. Havel, Capillary zone electrophoresis for the separation and char-
and engineered inorganic colloids: a review, Environ. Sci. Technol. 48 (2014) acterization of humic acids, J. Chromatogr. A 802 (1998) 189–202.
8946–8962. [43] Z.G. Yu, S. Orsetti, S.B. Haderlein, K.H. Knorr, Electron transfer between sulfide and
[26] H. Wang, J. Qi, A.A. Keller, M. Zhu, F. Li, Effects of pH, ionic strength and humic humic acid: electrochemical evaluation of the reactivity of sigma-Aldrich humic
acid on the removal of TiO2 nanoparticles from aqueous phase by coagulation, acid toward sulfide, Aquat. Geochem. 22 (2016) 117–130.
Colloids Surfaces A Physicochem. Eng. Asp. 450 (2014) 161–165. [44] M.A.G.T. Hoop, V. Den, H.P. Leeuwen, Influence of molar mass distribution on the
[27] S. Ghosh, H. Mashayekhi, P. Bhowmik, B. Xing, Colloidal stability of Al2O3 nano- complexation of heavy metals by humic material, Colloids Surf. A Physicochem.
particles as affected by coating of structurally different humic acids, Langmuir Acs Eng. Asp. 120 (1997) 235–242.
J. Surf. Colloids 26 (2010) 873–879. [45] A. Carrentero, R. Villepin, L. Brunel, J. Carries, Turbiscan Lab User Guide,
[28] J.C. Miranda-Trevino, C.A. Coles, Kaolinite properties, structure and influence of Formulaction, France, 2005.
metal retention on pH, Appl. Clay Sci. 23 (2003) 133–139. [46] B.G. Tabachnick, L.S. Fidell, Experimental Designs Using ANOVA, (2007).
[29] E. Tombácz, M. Szekeres, Surface charge heterogeneity of kaolinite in aqueous [47] A.M.E. Badawy, T.P. Luxton, R.G. Silva, K.G. Scheckel, M.T. Suidan, T.M. Tolaymat,
suspension in comparison with montmorillonite, Appl. Clay Sci. 34 (2006) Impact of environmental conditions (pH, ionic strength, and electrolyte type) on the
105–124. surface charge and aggregation of silver nanoparticles suspensions, Environ. Sci.
[30] W. Yu, J. Gregory, L.C. Campos, N. Graham, Dependence of floc properties on Technol. 44 (2010) 1260–1266.
coagulant type, dosing mode and nature of particles, Water Res. 68 (2015) [48] N. Odzak, D. Kistler, R. Behra, L. Sigg, Dissolution of metal and metal oxide na-
119–126. noparticles under natural freshwater conditions, Environ. Chem. 12 (2014)
[31] M.S. Usman, N.A. Ibrahim, K. Shameli, N. Zainuddin, W.M. Yunus, Copper nano- 138–148.
particles mediated by chitosan: synthesis and characterization via chemical [49] M. Luo, X. Qi, T. Ren, Y. Huang, A.A. Keller, H. Wang, B. Wu, H. Jin, F. Li,
methods, Molecules 17 (2012) 14928–14936. Heteroaggregation of CeO2 and TiO2 engineered nanoparticles in the aqueous
[32] M. Nasrollahzadeh, S.M. Sajadi, Y. Mirzaei, An efficient one-pot synthesis of 1,4- phase: application of turbiscan stability index and fluorescence excitation-emission
disubstituted 1,2,3-triazoles at room temperature by green synthesized Cu NPs matrix (EEM) spectra, Colloids Surf. A Physicochem. Eng. Asp. (2017).
using Otostegia persica leaf extract, J. Colloid Interface Sci. 468 (2016) 156–162. [50] A. Praetorius, J. Labille, M. Scheringer, A. Thill, K. Hungerbühler, J.Y. Bottero,
[33] J. Xiong, Y. Wang, Q. Xue, X. Wu, Synthesis of highly stable dispersions of nano- Heteroaggregation of titanium dioxide nanoparticles with model natural colloids
sized copper particles using L-ascorbic acid. Green Chem, Green Chem. 13 (2011) under environmentally relevant conditions, Environ. Sci. Technol. 48 (2014)
900–904. 10690–10698.
[34] K.A. Huynh, J.M. Mccaffery, K.L. Chen, Heteroaggregation of multiwalled carbon [51] M. Zhu, H. Wang, A.A. Keller, T. Wang, F. Li, The effect of humic acid on the
nanotubes and hematite nanoparticles: rates and mechanisms, Environ. Sci. aggregation of titanium dioxide nanoparticles under different pH and ionic
Technol. 46 (2012) 5912–5920. strengths, Sci. Total Environ. 487 (2014) 375–380.
[35] T. Yadav, A.A. Mungray, A.K. Mungray, A comparative analysis of TiO2 nano- [52] R.J. Hunter, Foundations of Colloid Science, second ed., OXFORD, New York, 2001.
particle dispersion in various biological extracts, RSC Adv. 5 (2015) 64421–64432. [53] Terence Cosgrove, Colloid Science: Principles, Methods and Applications, second
[36] M. Wiśniewska, K. Terpiłowski, S. Chibowski, T. Urban, V.I. Zarko, V.M. Gun’ko, ed., Wiley, UK, 2010.
Stability of colloidal silica modified by macromolecular polyacrylic acid (PAA) [54] M. Luo, Y. Huang, M. Zhu, Y.N. Tang, T. Ren, J. Ren, H. Wang, F. Li, Properties of
−application of turbidymetry method, J. Macromol. Sci. Part A 50 (2013) 639–643. different natural organic matter influence the adsorption and aggregation behavior
[37] P. Lodeiro, E.P. Achterberg, J. Pampín, A. Affatati, M.S. El-Shahawi, Silver nano- of TiO2 nanoparticles, J. Saudi Chem. Soc. (2016) 1–8.
particles coated with natural polysaccharides as models to study AgNP aggregation

104