Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Physico-chemical changes of ZnO nanoparticles with different size and surface chemistry under physiological pH conditions Gyeong-Hyeon Gwak a , Won-Jae Lee b , Seung-Min Paek b,∗∗ , Jae-Min Oh a,∗ a b
Department of Chemistry and Medical Chemistry, College of Science and Technology, Yonsei University, 220-710 Wonju, Gangwon, South Korea Department of Chemistry, Kyungpook National University, Taegu 702-701, South Korea
a r t i c l e
i n f o
Article history: Received 7 August 2014 Received in revised form 8 January 2015 Accepted 13 January 2015 Available online 25 January 2015 Keywords: ZnO Nanoparticles Size Surface chemistry Physiological pH condition Physico-chemical properties
a b s t r a c t We studied the physico-chemical properties of ZnO nanoparticles under physiological pH conditions (gastric, intestinal and plasma) as functions of their size (20 and 70 nm) and surface chemistry (pristine, l-serine, or citrate coating). ZnO nanoparticles were dispersed in phosphate buffered saline under physiological pH conditions and aliquots were collected at specific time points (0.5, 1, 4, 10 and 24 h) for further characterization. The pH values of the aqueous ZnO colloids at each condition were in the neutral to slightly basic range and showed different patterns depending on the original size and surface chemistry of the ZnO nanoparticles. The gastric pH condition was found to significantly dissolve ZnO nanoparticles up to 18–30 wt%, while the intestinal or plasma pH conditions resulted in much lower dissolution amounts than expected. Based on the X-ray diffraction patterns and X-ray absorption spectra, we identified partial phase transition of the ZnO nanoparticles from wurtzite to Zn(OH)2 under the intestinal and plasma pH conditions. Using scanning electron microscopy, we verified that the overall particle size and morphology of all ZnO nanoparticles were maintained regardless of the pH. © 2015 Published by Elsevier B.V.
1. Introduction Zinc oxide (ZnO) is one of the most well-known inorganic materials and is used in a broad array of applications such as drug delivery systems [1], food additives [2], cosmetic ingredients [3], food packaging additives [4], pigments [5], dyes [6], catalysis of methanol synthesis [7], optical device materials [8], semiconductors [9], and sensors [10]. Combined with recent progress in nanotechnology, nanosized ZnOs with dimensions on the scale of tens of nanometers are under development for high performance applications, with several now commercially available. Among the various applications of ZnO nanoparticles, many researchers have become especially interested in their potential for biomedical applications such as drug delivery and Zn supplementation [11]. ZnOs exhibit excellent biocompatibility as they can be dissolved in aqueous conditions to release Zn2+ , which in turn, is easily absorbed and secreted in biological systems [12]. Taking
∗ Corresponding author at: Yonsei University, Yonseidaegil-1, Heungeop-myeon, Wonju, Gangwon-do 220-710, South Korea. Tel.: +82 33 760 2368. ∗∗ Corresponding author at: Kyungpook National University, Sangyeok 3-dong, Buk-gu, Taegu 702-701, South Korea. Tel.: +82 53 950 5335. E-mail addresses: [email protected] (S.-M. Paek), [email protected] (J.-M. Oh). http://dx.doi.org/10.1016/j.colsurfb.2015.01.021 0927-7765/© 2015 Published by Elsevier B.V.
advantage of this property, ZnO is often added to foods such as breakfast cereals as a source of Zn. However, in spite of the high number of applications of ZnOs, only a few studies have investigated the physico-chemical properties and potential toxicity of nanoparticulates. Compared to their bulk counterparts, nanoparticles have a large surface area and high reactivity [12,13]. Therefore, nanoparticles can exhibit unexpected colloidal behaviors under physiological conditions, such as accelerated dissolution, phase transformation, and the formation of agglomerates/aggregates. As these colloidal behaviors are strongly related to the interface interactions between nano- and bio-systems which affect their potential toxicity, it is important to investigate the behaviors of ZnO nanoparticles under physiological conditions. It is generally accepted that particle size and surface chemistry are the major physico-chemical parameters of nanoparticles affecting their colloidal behaviors and biological reactivity [14]. Specifically, as the particle size decreases, the surface area and number of reactive surface sites increase, which often results in unexpected biochemical interactions not observed in the bulk state [11,15]. Surface chemistry is another important factor as it controls nano–bio interactions such as cellular adsorption and uptake [16]. In order to fully understand the nano–bio interface interactions and potential toxicities, it is important to first investigate the physico-chemical properties of ZnO nanoparticles under physiological conditions considering the particle size and surface
138
G.-H. Gwak et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
chemistry. In the present study, we prepared six different types of ZnO nanoparticles with various particle sizes and surface chemistries. ZnO nanoparticles with two different sizes (20 nm, ZnO20, and 70 nm, ZnO70) were purchased from commercial manufacturers. Using these particles, three different surface chemistries were prepared, namely, S-ZnO (l-serine), C-ZnO (citrate) and untreated. Considering the oral administration of ZnO nanoparticles, we studied three in vitro physiological conditions simulating the pH of the stomach, small intestine and plasma with pH values of 1.2, 6.8 and 7.4, respectively. Finally, we evaluated the solubility and changes of the crystal structure, size and local chemical environment of ZnO nanoparticles at each of the simulated pH conditions as functions of the particle size and surface chemistry. 2. Materials and methods 2.1. Materials Two kinds of ZnO nanoparticles with sizes of 20 nm (ZnO20) and 70 nm (ZnO70) were purchased from Sumitomo Osaka Cement Co. Ltd. (Chiba, Japan) and American Elements (Los Angeles, USA), respectively. HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid), sodium carbonate (Na2 CO3 ), l-serine and sodium citrate tribasic dihydrate (HOC(COONa)(CH2 COONa)2 ·2H2 O), which were utilized to modify the surface chemistry of the ZnO nanoparticles, were obtained from Sigma–Aldrich Co. LLC. (St. Louis, USA). Phosphate buffered saline (PBS) buffer solution was purchased from the Lonza Group Ltd. (Basel, Switzerland). 2.2. Surface chemistry modification of the ZnO nanoparticles Both ZnO20 and ZnO70 were modified with organic moieties to alter the surface chemistry and charge of the ZnO nanoparticles. In order to obtain ZnO nanoparticles with a positively charged organic surface, ZnO powder was suspended in a 1% l-serine solution prepared in 20 mM HEPES buffer (pH ∼ 6). The suspension was then stirred vigorously for 2 days. To prepare ZnO nanoparticles with a negatively charged organic surface, the same procedure was carried out using sodium citrate tribasic dihydrate instead of l-serine. The final concentration of ZnO in the suspension was adjusted to 1 wt/v%. The prepared l-serine and citrate-treated samples are abbreviated as S-ZnO and C-ZnO, respectively. The crystal structure and crystallinity of all ZnOs (ZnO20, S-ZnO20, C-ZnO20, ZnO70, S-ZnO70, and C-ZnO70) were characterized using an X-ray diffractometer (XRD) employing a Bruker ˚ In order D2PHASER with Ni-filtered CuK␣ radiation ( = 1.5406 A). to analyze the surfaces of the ZnO nanoparticles before and after surface modification, Fourier transform infrared spectra (FTIR) were obtained using PerkinElmer spectrum one FTIR spectrometer using KBr pellet technique. Scanning electron microscopy (SEM) images were obtained by a FEI Quanta 250 FEG at an acceleration potential of 30 kV in order to observe the size and morphology of the ZnO nanoparticles. 2.3. Evaluation of the physico-chemical properties of the ZnO nanoparticles under simulated physiological pH conditions In order to observe the physico-chemical property changes of the ZnO nanoparticles in physiological pH conditions as functions of their size and surface chemistry, we placed the ZnO samples in three different PBS buffer solutions at pH values of 1.2, 6.8 and 7.4 to simulate gastric, intestinal, and plasma conditions, respectively. We measured the time-dependent changes of ZnO considering the crystal structure, particle size/morphology, and local chemical environment around the Zn2+ ions. Each ZnO sample was dispersed in a pH adjusted PBS solution at a concentration of
Fig. 1. X-ray diffraction patterns of the ZnO nanoparticles: (a) ZnO20, (b) S-ZnO20, (c) C-ZnO20, (d) ZnO70, (e) S-ZnO70, and (f) C-ZnO70.
10 mg/mL. Aliquots were collected from the suspension at 0.5, 1, 4, 10 and 24 h, and centrifuged to separate the supernatant and precipitate. Supernatants were used to evaluate the time-dependent dissolution of ZnO using an atomic absorption spectrometer (AAS, PerkinElmer AAnalyst 400 with acetylene gas) or inductively coupled plasma-mass spectrometer (ICP-MS, PerkinElmer SCIEX with argon plasma). The precipitates were dried in a vacuum oven and further characterized as follows. The crystal structure and particle size/morphology of the ZnOs were evaluated by XRD and SEM. The crystallite sizes of the ZnOs were calculated according to the (1 0 1) peak of their XRD patterns utilizing Scherrer’s equation (Eq. (1)) [17]. =
0.9 ˇ cos
(1)
where , grain size; , X-ray wavelength (1.5406 A˚ for CuK␣); ˇ, full-width at half-maximum (FWHM) of diffraction peak in radian; , Bragg angle in radian. The local chemical environment around the Zn2+ ions in the various samples was investigated by Zn K-edge X-ray absorption near edge structure (XANES) analysis performed at the 8C beam line at the Pohang Accelerator Laboratory (PAL) in Korea. XANES spectra of samples were obtained at room temperature in the transmission mode, in which standard Zn metal foil was simultaneously measured to obtain an exact calibration of the edge position. Data acquisition and analysis were carried out using standard procedures reported previously [18,19]. 3. Results and discussion We first characterized all of the ZnO nanoparticles used in this study. In order to identify the crystal phase and crystallinity, X-ray diffraction patterns were obtained (Fig. 1). In the 2 range from 20◦ to 80◦ , all of the ZnOs showed typical diffraction peaks of the wurtzite phase (JCPDS No. 36-1451): (1 0 0), (0 0 2), (1 0 1), (1 0 2), (1 1 0), (1 0 3), (2 0 0), (1 1 2), (2 0 1), (0 0 4) and (2 0 2) at 31.7◦ , 34.4◦ , 36.2◦ , 47.5◦ , 56.6◦ , 62.8◦ , 66.3◦ , 67.9◦ , 69.0◦ , 72.5◦ and 76.9◦ , respectively. Although a slight shift of the peak positions of (1 0 1) and (1 0 3) between ZnO20 and ZnO70 was observed, as can be seen in a comparison of Fig. 1(a)–(c) with (d)–(f), the cell parameters of all the samples calculated from the XRD patterns showed the same ˚ values within three significant figures (a = 3.25 A˚ and c = 5.21 A), regardless of their size or surface chemistry. Due to the larger primary particle size of the ZnO70s, the ZnO70 samples (ZnO70,
G.-H. Gwak et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
139
S-ZnO70 and C-ZnO70) clearly exhibited higher intensities with respect to the diffraction peaks compared with the ZnO20 series (ZnO20, S-ZnO20 and C-ZnO20). The crystallite sizes calculated using Scherrer’s equation (Eq. (1)) were approximately 20 nm and 50 nm, respectively, for ZnO20s and ZnO70s. It is notable that the XRD patterns for the ZnO series with the same starting particle size were similar in terms of intensity and peak width, which indicates that no significant dissolution or phase transformation occurred during the surface modification with citrate or l-serine. In the FTIR spectra of the untreated ZnOs (Fig. S1), we observed characteristic peaks at 3450 cm−1 and 485 cm−1 , which can be attributed to OH stretching and Zn O stretching vibrations, respectively. Thus, the surfaces of as-purchased ZnO were thought to possess Zn OH moieties, regardless of the manufacturer. The infrared analysis of the surface-modified ZnO nanoparticles showed typical peaks corresponding to carboxylate and amines, suggesting that the surface of ZnO contained an intact structure of citrate or l-serine. The COO− terminal of l-serine molecules interact with Zn2+ ions at the surface of ZnO nanoparticles in a unidentate manner, which was confirmed by analyzing the wavenumber difference between asymmetric and symmetric stretching mode of carboxylate in Fig. S1. The zeta potential values of ZnO20, S-ZnO20, C-ZnO20, ZnO70, S-ZnO70 and C-ZnO70 at a pH of 7 were determined to be +28.8 ± 0.9, +26.3 ± 0.5, −34.4 ± 1.0, +26.3 ± 0.9, +26.1 ± 0.4 and −41.4 ± 0.6 mV, respectively, as previously reported [20]. This result indicates that the citrate treatment effectively modified the surface of ZnO nanoparticles to have a negative charge, while l-serine treatment maintained the original positive charge of the ZnOs. As we aimed to monitor the particle size and surface chemistry dependent behavior of ZnO nanoparticles when administered through an oral route, we applied a simple model of the physiological conditions of the digestive route. Specifically, the physico-chemical properties of metal oxide or hydroxide nanoparticles are usually more sensitive to pH than enzymatic reactions [21]. Thus, we utilized PBS buffer as a physiological condition without enzymes and adjusted the pH to 1.2, 6.8 and 7.4 to simulate gastric, intestinal, and plasma environments, respectively. For the first step, we monitored the time-dependent pH changes of each PBS buffer upon treatment with ZnOs. All of the different pH solutions exhibited an immediate pH increase (Fig. 2) upon the addition of the ZnOs, which can be attributed to the dissolution of Zn2+ (Eq. (2)) at the surface, resulting in the absorption of H+ and production of OH− . ZnO(s) + H+ (aq) Zn2+ (aq) + OH− (aq)
(2)
In the case of the simulated gastric solution, ZnO increased the pH from 1.2 to 6.5, reaching a plateau within 0.1 h (Fig. 2(A)). The pH values of the simulated intestinal (pH ∼ 6.8) and plasma (pH ∼ 7.4) solutions also exhibited a pH increase, resulting in a final pH value of ∼9.5 at 10 h. The difference of the pH (pH) before and after ZnO treatment was much higher for the gastric condition (pH = 5.6) compared with the intestinal and plasma conditions (pH = 2.6 and 2.1, respectively). As the pH increase depends mainly on the reaction between ZnO and H+ , the pH increase was more remarkable under the gastric condition, which had the largest abundance of H+ ions. While the gastric solution readily obtained pH equilibrium regardless of the size and surface chemistry of ZnO, the pH trends of the intestinal and plasma solutions were highly dependent on the particle size. The pH with the smaller ZnOs (ZnO20s) tended to increase faster than with larger particles (ZnO70s) (Figs. 2(B) and (C)). As smaller particles have larger specific surface areas (SBET = 27.7 m2 /g and 16.6 m2 /g for ZnO20 and ZnO70, respectively), the surface interactions between ZnO nanoparticles and H+ were more favorable with the ZnO20s. In addition, the increase of the pH was also slightly affected by the surface chemistry in which
Fig. 2. Time dependent pH changes of the physiological conditions simulating the pH of (A) the stomach (pH ∼1.2), (B) intestines (pH ∼6.8), and (C) plasma (pH ∼7.4) after incubation with ZnO nanoparticles. Each data point and error bar represents the average and standard deviation, respectively, obtained from triplicate experiments.
the pH with the uncoated ZnO nanoparticles increased faster than with coated nanoparticles. This effect was apparent for the 20 nm sized ZnO nanoparticles under the intestinal and plasma conditions (Figs. 2(B) and (C)). Specifically, ZnO20 (squares) resulted in a larger increase of the pH compared to the coated nanoparticles (circles: S-ZnO20, triangles: C-ZnO20). Thus, the surface coating agents may
140
G.-H. Gwak et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
Fig. 3. Time dependent cumulative dissolution of ZnO nanoparticles under gastric (pH ∼ 1.2) conditions. Each data point and error bar represents the average and standard deviation, respectively, obtained from triplicate experiments.
have blocked efficient interactions between the ZnO surface and H+ , decelerating the dissolution rate. However, the final pH was similar regardless of the surface chemistry, suggesting that the overall reaction equilibrium was not governed by the surface properties. As the particle size and surface chemistry of ZnO nanoparticles were determined to affect the pH increase, we evaluated the dissolution of ZnOs under each condition utilizing AAS and ICPMS. Monitoring the dissolution rates is essential to evaluate the physico-chemical properties of ZnOs and it was previously reported that dissolution strongly affects bioavailability [12]. Fig. 3 shows the time-dependent dissolution of the ZnO nanoparticles under gastric pH condition obtained by AAS. All of the ZnO nanoparticles showed prompt dissolution in the range of 18–30 wt%. Further, the amount of dissolved Zn2+ did not change significantly over a 24 h period, suggesting that the dissolution reached equilibrium at an early stage. This result is highly consistent with our observation of a fast increase of the pH with ZnO nanoparticles in the gastric pH buffer, as shown in Fig. 2(A). Although the cumulative dissolution appeared to be affected by the particle size and surface chemistry, we were unable to identify a statistically significant difference between the samples with a 99% reliability using the Student’s t-test. ZnO dissolution under intestinal (pH ∼ 6.8) and plasma (pH ∼ 7.4) conditions was very small and thus, it was further monitored by ICP-MS (Table 1). The cumulative dissolution measured by ICP-MS was less than 0.01 wt% under both conditions. Upon calculating the theoretical amount of dissolved ZnO based on the change of the pH for each solution, we identified a significant discrepancy between the calculated values and our measurements. Specifically, the gastric, intestinal, and plasma solutions showed pH values of 5.6, 2.6 and 2.1, respectively. Thus, we calculated that 6.310 × 10−3 , 1.581 × 10−8 and 3.950 × 10−9 mol of H+ and 6.310 × 10−9 , 2.506 × 10−6 and 3.137 × 10−6 mol of OH− were produced during the reaction under each condition, respectively. These values correspond to cumulative ZnO dissolutions of 25.3, 0.082 and 0.013 wt% for the gastric, intestinal and plasma conditions, respectively. The calculated value for the gastric condition was fairly comparable with the measured value. However, the calculated values for the intestinal and plasma conditions were much larger than the measured ones. Thus, we hypothesized that the behavior of the ZnO nanoparticles at relatively higher pH values (initial pH values of 6.8 and 7.4) was not only governed by dissolution (Eq. (2)), but also by Zn2+ hydrolysis in an aqueous system. It is well known that Zn2+ forms various hydrolysis products such as Zn2+ , Zn(OH)+ , Zn(OH)2 ,
Fig. 4. X-ray diffraction patterns of the ZnO nanoparticles after treatment for 24 h under physiological conditions simulating (A) the intestines (pH ∼ 6.8) and (B) plasma (pH ∼ 7.4): (a) ZnO20, (b) S-ZnO20, (c) C-ZnO20, (d) ZnO70, (e) S-ZnO70, and (f) C-ZnO70.
Zn(OH)3 − , Zn(OH)4 2− , Zn2 OH3+ , and Zn2 (OH)6 2− , depending on the pH [22]. Indeed, dissolved Zn2+ may have readily formed hydrolysis products which adsorb on the solid ZnO surface, thereby hindering dissolution. According to the Zn2+ hydrolysis curves (Fig. S2), the major hydrolysis products at pH ∼ 6.5 (final pH of the gastric condition) and pH ∼ 9 (final pH of the intestinal and plasma conditions) were Zn(OH)x 2−x and Zn(OH)2 , respectively. In particular, the pH range of 9–11 is optimum to precipitate Zn(OH)2 , which could cover ZnO nanoparticles in the intestinal and plasma conditions. In order to verify the presence of evolved Zn2+ hydrolysis products generated from ZnO nanoparticles under the intestinal and plasma conditions, we collected nanoparticles after reacting for 24 h and measured the resulting XRD patterns. As shown in Fig. 4(A), we could clearly observe a peak at 2 = 11.3◦ in all of the ZnOs reacted under the intestinal condition (pH ∼ 6.8). This peak was assigned as the (2 0 0) peak of the zinc hydroxide phase (Zn5 (OH)6 (CO3 )2 ; JCPDS No. 19-1458, hydrozincite), which is one of the possible hydrolysis products of Zn2+ . As the Zn2+ dissolved from ZnO could readily react with water and atmospheric CO2 , hydrozincite was thought to precipitate on the surface of the ZnO nanoparticles during the reaction. Furthermore, the pH in the intestinal and plasma conditions after ZnO treatment reached 9–10, which is a favorable condition for the precipitation of hydrozincite. ZnO nanoparticles treated in the plasma condition (pH ∼ 7.4) also showed evolution of a hydrozincite phase for the 70 nm samples (Fig. 4(B)). In order to confirm the formation of hydrozincite under simulated physiological conditions, we investigated the X-ray absorption spectra (XAS) of samples at the Zn K-edge. ZnOs with a wurtzite phase consist of ZnO4 tetrahedrons with shared corners, while hydrozincite is composed of both Zn(OH)4 tetrahedrons and Zn(OH)6 octahedrons. The XANES is sufficiently sensitive to detect subtle structural changes around specific elements. Thus, changes of the coordination information around Zn2+ could be obtained from the XANES analysis. We obtained the Zn K-edge XANES spectra of ZnO20 before and after treatment under the plasma simulating pH condition where small and bare ZnO nanoparticles with many reactive sites on the particle surface showed relatively remarkable phase transformation on the surface (Fig. 5). The Zn K-edge XANES spectrum of ZnO20 showed a subpeak at 9661 eV and an intense white line at 9667 eV (solid line in Fig. 5), which were attributed to the electron transition from the 1s to 4p orbital, as reported
G.-H. Gwak et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
141
Table 1 The cumulative dissolution of Zn2+ ions from ZnO nanoparticles (wt%) after treatment for 24 h under physiological conditions simulating the intestines (pH ∼ 6.8) and plasma (pH ∼ 7.4). All of the values were obtained from triplicate experiments. The data stand for the average values and corresponding standard deviations. ZnO20
C-ZnO20
ZnO70
S-ZnO70
C-ZnO70
Intestine (pH 6.8) (wt%) 0.00023 ± 0.00004 0.00514 ± 0.00028
S-ZnO20
0.00193 ± 0.00007
0.00018 ± 0.00001
0.00122 ± 0.00004
0.00291 ± 0.00010
Plasma (pH 7.4) (wt%) 0.00015 ± 0.00003 0.00879 ± 0.00034
0.00087 ± 0.00006
0.00011 ± 0.00004
0.00111 ± 0.00016
0.00108 ± 0.00014
Fig. 5. Zn K-edge X-ray absorption near edge structure (XANES) spectra of ZnO20 before (solid line) and after (open circles) treatment for 24 h under the plasma pH condition (pH ∼ 7.4).
previously [23,24]. Although the overall XANES pattern of ZnO20 in the plasma condition (open circle in Fig. 5) was similar to that of bare ZnO20, closer inspection revealed a slight difference. Specifically, the subpeak and white line of ZnO in the plasma condition increased from 9660.8 to 9661.3 eV and from 9666.6 to 9666.9 eV, respectively, compared to untreated ZnO. Furthermore, the white line intensity increased by ∼10% after plasma treatment. Importantly, the position and intensity of the white line are closely related to the coordination number. It has been reported that the peaks around the main edge shift to a higher energy and the white line intensities increase as the coordination number of zinc compounds increase [25]. Thus, the enhanced white line intensity observed in this study reflects expansion of the average coordination number around Zn2+ , which may be due to the evolution of hydrozincite with a Zn(OH)6 octahedra structure. These results are in good agreement with the XRD analysis in which a new XRD peak assigned to hydrozincite was detected. Therefore, our results suggest that a hydrozincite phase evolved from ZnO nanoparticles in the simulated plasma condition. As we observed that the ZnO nanoparticles partially dissolved and produced a certain amount of hydrozincite, we next examined the overall crystalline phase and crystallite size of the ZnO nanoparticles under each simulated condition. Fig. S3 shows the XRD patterns of the ZnOs after treatment for 24 h. All of the patterns showed a clear wurtzite phase, which was indifferent from the untreated ZnOs shown in Fig. 1, indicating that there was overall preservation of the phase during treatment. Although it is known that Zn2+ ions readily precipitate with phosphate ions, we could not find any evidence for the evolution of zinc phosphate in the structural analyses including XRD. It is thought that a small quantity of zinc phosphate could be formed, but the evolution of hydrozincite was prevalent. To analyze the time-dependent changes of the crystallinity, the crystallite sizes of all of the ZnOs at each time point were calculated using Scherrer’s equation with the (1 0 1) peaks. Each ZnO peak largely corresponded to initial crystallite sizes of
20 and 50 nm for the ZnO20s and ZnO70s, respectively, regardless of the time or pH (Fig. S4). As only a slight dissolution of Zn2+ was detected (Table 1), the ZnO nanoparticles were considered to maintain their crystallinity and size under both intestinal (pH ∼ 6.8) and plasma (pH ∼ 7.4) conditions. Interestingly, the ZnO nanoparticles in the gastric (pH ∼ 1.2) condition did not exhibit a significant decrease of the crystallite size despite the obvious dissolution (18–30 wt%) of Zn2+ . We could explain this phenomenon by simply calculating the particle size of both ZnO20 and ZnO70 after 18–30 wt% dissolution. Specifically, as the dissolution would naturally initiate at the surface of the nanoparticles, we hypothesized that the outermost sphere of the particles dissolved during the reaction. The dissolution of 18–30 wt% for the 20 nm and 70 nm particles resulted in a reduction of the diameter to 18 nm and 65 nm, respectively, meaning that there was a >90% maintenance of the original crystallite size. Thus, even dissolution of ZnO nanoparticles up to 30 wt% did not significantly affect the crystallite size. Finally, in order to confirm whether the ZnO particles maintained their original morphology, we obtained SEM images of all of the ZnOs before and after treatment under gastric conditions for 24 h. The original ZnO20s (Fig. S5(a)–(c)) exhibited a smooth and round-shaped primary particles with size of 28.03 ± 5.62 nm, while the original ZnO70s (Fig. S5(d)–(f)) showed a similar spherical morphology with a size of 77.45 ± 21.05 nm. After treatment for 24 h, we did not observe any significant changes of the morphology or surface roughness for all of the ZnO samples evaluated. The particle sizes of the ZnO20s and ZnO70s after treatment at a pH of 1.2 were determined to be 24.36 ± 7.19 nm (Fig. S5(a )–(c )) and 79.49 ± 21.68 nm (Fig. S5(d )–(f )), respectively. There was no significant difference between the 95% confidence intervals for the two conditions according to the Student’s t-test. The preservation of the particle size and morphology was also similar for the pH 6.8 and pH 7.4 conditions (Figs. S6 and S7). 4. Conclusions In order to investigate the physico-chemical properties of ZnO nanoparticles as a function of the particle size and surface chemistry, we observed the time-dependent physico-chemical changes in PBS buffer solutions simulating the pH of the stomach, intestine, and plasma. The ZnO nanoparticles were determined to enhance the pH in the physiological pH conditions to a neutral (in the case of the gastric condition) or basic range (in the case of the intestinal and plasma conditions), revealing a dependency on the size and surface chemistry. The dissolution of ZnO was significant under the gastric pH condition, whereas under the intestinal and plasma conditions, it was trivial. As the dissolution of ZnO under intestinal and plasma conditions was much less than expected based on the observed pH increase, we hypothesized that a phase transformation from ZnO to Zn(OH)2 occurred. Consistently, X-ray diffraction and X-ray absorption spectroscopy revealed the possible formation of hydrozincite on the surface of ZnO during the reaction. Based on Scherrer’s equation and the SEM analysis, the size and morphology of the ZnO nanoparticles were not seriously altered by the physiological pH conditions. Thus, we concluded that the ZnO nanoparticles mainly dissolved in the gastric condition and that the other physiological
142
G.-H. Gwak et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
conditions caused a phase transformation of the ZnO nanoparticles. While the particle size and surface chemistry are thought to affect interface interactions between nano- and bio-systems, these factors do not appear to govern the overall physico-chemical properties of the ZnO nanoparticles. Acknowledgements This work was financially supported by the Ministry of Science, ICT and Future Planning (MSIP) Korea Government and the Korea Industrial Technology Association (KOITA) as “A study on the programs to support a collaborative research among industry, academia and research institutes” (KOITA-2013-2-5), by Rural Development Administration (RDA, Korea) as “Cooperative Research Program for Agriculture Science & Technology Development” (PJ0105022015), and by a National Research Foundation of Korea (NRF) grant funded by the MSIP, Korea Government (20100024370). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.colsurfb. 2015.01.021. References [1] J.H. Yuan, Y. Chen, H.X. Zha, L.J. Song, C.Y. Li, J.Q. Li, X.H. Xia, Colloids Surf. B: Biointerfaces 76 (2010) 145.
[2] P. Whittaker, Am. J. Clin. Nutr. 68 (1998) 442S. [3] S.R. Pinnell, D. Fairhurst, R. Gillies, M.A. Mitchnick, N. Kollias, Dermatol. Surg. 26 (2000) 309. [4] R. Tankhiwale, S. Bajpai, Colloids Surf. B: Biointerfaces 90 (2012) 16. [5] S.K. Dhoke, A.S. Khanna, T.J.M. Sinha, Prog. Org. Coat. 64 (2009) 371. [6] M. Law, L.E. Greene, J.C. Johnson, R. Saykally, P. Yang, Nat. Mater. 4 (2005) 455. [7] M. Behrens, F. Studt, I. Kasatkin, S. Kühl, M. Hävecker, F. Abild-Pedersen, S. Zander, F. Girgsdies, P. Kurr, B.-L. Kniep, Science 336 (2012) 893. [8] L. Kumari, W.Z. Li, Cryst. Res. Technol. 45 (2010) 311. [9] A. Janotti, C.G. Van de Walle, Rep. Prog. Phys. 72 (2009) 126501. [10] T. Yumak, F. Kuralay, M. Muti, A. Sinag, A. Erdem, S. Abaci, Colloids Surf. B: Biointerfaces 86 (2011) 397. [11] M. Heinlaan, A. Ivask, I. Blinova, H.-C. Dubourguier, A. Kahru, Chemosphere 71 (2008) 1308. [12] H.-J. Paek, Y.-J. Lee, H.-E. Chung, N.-H. Yoo, J.-A. Lee, M.-K. Kim, J.K. Lee, J. Jeong, S.-J. Choi, Nanoscale 5 (2013) 11416. [13] N. Padmavathy, R. Vijayaraghavan, Sci. Technol. Adv. Mater. 9 (2008) 035004. [14] A. Albanese, P.S. Tang, W.C. Chan, Annu. Rev. Biomed. Eng. 14 (2012) 1. [15] M.D. Newman, M. Stotland, J.I. Ellis, J. Am. Acad. Dermatol. 61 (2009) 685. [16] J.Y. Kwon, S.Y. Lee, P. Koedrith, J.Y. Lee, K.-M. Kim, J.-M. Oh, S.I. Yang, M.-K. Kim, J.K. Lee, J. Jeong, Mutat. Res. Genet. Toxicol. Environ. Mutagen. 761 (2014) 1. [17] B.D. Cullity, Am. J. Phys. 25 (1957) 394. [18] S.M. Paek, H. Jung, Y.J. Lee, M. Park, S.J. Hwang, J.H. Choy, Chem. Mater. 18 (2006) 1134. [19] S.M. Paek, H. Jung, M. Park, J.K. Lee, J.H. Choy, Chem. Mater. 17 (2005) 3492. [20] K.-M. Kim, T.-H. Kim, H.-M. Kim, H.-J. Kim, G.-H. Gwak, S.-M. Paek, J.-M. Oh, Toxicol. Environ. Health Sci. 4 (2012) 121. [21] J.Y. Kim, S.-J. Choi, J.-M. Oh, T. Park, J.-H. Choy, J. Nanosci. Nanotechnol. 7 (2007) 3700. [22] C.F. Baes, R.E. Mesmer, Hydrolysis of Cations, Wiley, New York, 1976. [23] S. Larcheri, C. Armellini, F. Rocca, A. Kuzmin, R. Kalendarev, G. Dalba, R. Graziola, J. Purans, D. Pailharey, F. Jandard, Superlattices Microstruct. 39 (2006) 267. [24] J. Rose, I. Moulin, A. Masion, P.M. Bertsch, M.R. Wiesner, J.-Y. Bottero, F. Mosnier, C. Haehnel, Langmuir 17 (2001) 3658. [25] F. Sowrey, L. Skipper, D. Pickup, K. Drake, Z. Lin, M. Smith, R. Newport, Phys. Chem. Chem. Phys. 6 (2004) 188.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Physico-chemical changes of ZnO nanoparticles with different size and surface chemistry under physiological pH conditions Gyeong-Hyeon Gwak a , Won-Jae Lee b , Seung-Min Paek b,∗∗ , Jae-Min Oh a,∗ a b
Department of Chemistry and Medical Chemistry, College of Science and Technology, Yonsei University, 220-710 Wonju, Gangwon, South Korea Department of Chemistry, Kyungpook National University, Taegu 702-701, South Korea
a r t i c l e
i n f o
Article history: Received 7 August 2014 Received in revised form 8 January 2015 Accepted 13 January 2015 Available online 25 January 2015 Keywords: ZnO Nanoparticles Size Surface chemistry Physiological pH condition Physico-chemical properties
a b s t r a c t We studied the physico-chemical properties of ZnO nanoparticles under physiological pH conditions (gastric, intestinal and plasma) as functions of their size (20 and 70 nm) and surface chemistry (pristine, l-serine, or citrate coating). ZnO nanoparticles were dispersed in phosphate buffered saline under physiological pH conditions and aliquots were collected at specific time points (0.5, 1, 4, 10 and 24 h) for further characterization. The pH values of the aqueous ZnO colloids at each condition were in the neutral to slightly basic range and showed different patterns depending on the original size and surface chemistry of the ZnO nanoparticles. The gastric pH condition was found to significantly dissolve ZnO nanoparticles up to 18–30 wt%, while the intestinal or plasma pH conditions resulted in much lower dissolution amounts than expected. Based on the X-ray diffraction patterns and X-ray absorption spectra, we identified partial phase transition of the ZnO nanoparticles from wurtzite to Zn(OH)2 under the intestinal and plasma pH conditions. Using scanning electron microscopy, we verified that the overall particle size and morphology of all ZnO nanoparticles were maintained regardless of the pH. © 2015 Published by Elsevier B.V.
1. Introduction Zinc oxide (ZnO) is one of the most well-known inorganic materials and is used in a broad array of applications such as drug delivery systems [1], food additives [2], cosmetic ingredients [3], food packaging additives [4], pigments [5], dyes [6], catalysis of methanol synthesis [7], optical device materials [8], semiconductors [9], and sensors [10]. Combined with recent progress in nanotechnology, nanosized ZnOs with dimensions on the scale of tens of nanometers are under development for high performance applications, with several now commercially available. Among the various applications of ZnO nanoparticles, many researchers have become especially interested in their potential for biomedical applications such as drug delivery and Zn supplementation [11]. ZnOs exhibit excellent biocompatibility as they can be dissolved in aqueous conditions to release Zn2+ , which in turn, is easily absorbed and secreted in biological systems [12]. Taking
∗ Corresponding author at: Yonsei University, Yonseidaegil-1, Heungeop-myeon, Wonju, Gangwon-do 220-710, South Korea. Tel.: +82 33 760 2368. ∗∗ Corresponding author at: Kyungpook National University, Sangyeok 3-dong, Buk-gu, Taegu 702-701, South Korea. Tel.: +82 53 950 5335. E-mail addresses: [email protected] (S.-M. Paek), [email protected] (J.-M. Oh). http://dx.doi.org/10.1016/j.colsurfb.2015.01.021 0927-7765/© 2015 Published by Elsevier B.V.
advantage of this property, ZnO is often added to foods such as breakfast cereals as a source of Zn. However, in spite of the high number of applications of ZnOs, only a few studies have investigated the physico-chemical properties and potential toxicity of nanoparticulates. Compared to their bulk counterparts, nanoparticles have a large surface area and high reactivity [12,13]. Therefore, nanoparticles can exhibit unexpected colloidal behaviors under physiological conditions, such as accelerated dissolution, phase transformation, and the formation of agglomerates/aggregates. As these colloidal behaviors are strongly related to the interface interactions between nano- and bio-systems which affect their potential toxicity, it is important to investigate the behaviors of ZnO nanoparticles under physiological conditions. It is generally accepted that particle size and surface chemistry are the major physico-chemical parameters of nanoparticles affecting their colloidal behaviors and biological reactivity [14]. Specifically, as the particle size decreases, the surface area and number of reactive surface sites increase, which often results in unexpected biochemical interactions not observed in the bulk state [11,15]. Surface chemistry is another important factor as it controls nano–bio interactions such as cellular adsorption and uptake [16]. In order to fully understand the nano–bio interface interactions and potential toxicities, it is important to first investigate the physico-chemical properties of ZnO nanoparticles under physiological conditions considering the particle size and surface
138
G.-H. Gwak et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
chemistry. In the present study, we prepared six different types of ZnO nanoparticles with various particle sizes and surface chemistries. ZnO nanoparticles with two different sizes (20 nm, ZnO20, and 70 nm, ZnO70) were purchased from commercial manufacturers. Using these particles, three different surface chemistries were prepared, namely, S-ZnO (l-serine), C-ZnO (citrate) and untreated. Considering the oral administration of ZnO nanoparticles, we studied three in vitro physiological conditions simulating the pH of the stomach, small intestine and plasma with pH values of 1.2, 6.8 and 7.4, respectively. Finally, we evaluated the solubility and changes of the crystal structure, size and local chemical environment of ZnO nanoparticles at each of the simulated pH conditions as functions of the particle size and surface chemistry. 2. Materials and methods 2.1. Materials Two kinds of ZnO nanoparticles with sizes of 20 nm (ZnO20) and 70 nm (ZnO70) were purchased from Sumitomo Osaka Cement Co. Ltd. (Chiba, Japan) and American Elements (Los Angeles, USA), respectively. HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid), sodium carbonate (Na2 CO3 ), l-serine and sodium citrate tribasic dihydrate (HOC(COONa)(CH2 COONa)2 ·2H2 O), which were utilized to modify the surface chemistry of the ZnO nanoparticles, were obtained from Sigma–Aldrich Co. LLC. (St. Louis, USA). Phosphate buffered saline (PBS) buffer solution was purchased from the Lonza Group Ltd. (Basel, Switzerland). 2.2. Surface chemistry modification of the ZnO nanoparticles Both ZnO20 and ZnO70 were modified with organic moieties to alter the surface chemistry and charge of the ZnO nanoparticles. In order to obtain ZnO nanoparticles with a positively charged organic surface, ZnO powder was suspended in a 1% l-serine solution prepared in 20 mM HEPES buffer (pH ∼ 6). The suspension was then stirred vigorously for 2 days. To prepare ZnO nanoparticles with a negatively charged organic surface, the same procedure was carried out using sodium citrate tribasic dihydrate instead of l-serine. The final concentration of ZnO in the suspension was adjusted to 1 wt/v%. The prepared l-serine and citrate-treated samples are abbreviated as S-ZnO and C-ZnO, respectively. The crystal structure and crystallinity of all ZnOs (ZnO20, S-ZnO20, C-ZnO20, ZnO70, S-ZnO70, and C-ZnO70) were characterized using an X-ray diffractometer (XRD) employing a Bruker ˚ In order D2PHASER with Ni-filtered CuK␣ radiation ( = 1.5406 A). to analyze the surfaces of the ZnO nanoparticles before and after surface modification, Fourier transform infrared spectra (FTIR) were obtained using PerkinElmer spectrum one FTIR spectrometer using KBr pellet technique. Scanning electron microscopy (SEM) images were obtained by a FEI Quanta 250 FEG at an acceleration potential of 30 kV in order to observe the size and morphology of the ZnO nanoparticles. 2.3. Evaluation of the physico-chemical properties of the ZnO nanoparticles under simulated physiological pH conditions In order to observe the physico-chemical property changes of the ZnO nanoparticles in physiological pH conditions as functions of their size and surface chemistry, we placed the ZnO samples in three different PBS buffer solutions at pH values of 1.2, 6.8 and 7.4 to simulate gastric, intestinal, and plasma conditions, respectively. We measured the time-dependent changes of ZnO considering the crystal structure, particle size/morphology, and local chemical environment around the Zn2+ ions. Each ZnO sample was dispersed in a pH adjusted PBS solution at a concentration of
Fig. 1. X-ray diffraction patterns of the ZnO nanoparticles: (a) ZnO20, (b) S-ZnO20, (c) C-ZnO20, (d) ZnO70, (e) S-ZnO70, and (f) C-ZnO70.
10 mg/mL. Aliquots were collected from the suspension at 0.5, 1, 4, 10 and 24 h, and centrifuged to separate the supernatant and precipitate. Supernatants were used to evaluate the time-dependent dissolution of ZnO using an atomic absorption spectrometer (AAS, PerkinElmer AAnalyst 400 with acetylene gas) or inductively coupled plasma-mass spectrometer (ICP-MS, PerkinElmer SCIEX with argon plasma). The precipitates were dried in a vacuum oven and further characterized as follows. The crystal structure and particle size/morphology of the ZnOs were evaluated by XRD and SEM. The crystallite sizes of the ZnOs were calculated according to the (1 0 1) peak of their XRD patterns utilizing Scherrer’s equation (Eq. (1)) [17]. =
0.9 ˇ cos
(1)
where , grain size; , X-ray wavelength (1.5406 A˚ for CuK␣); ˇ, full-width at half-maximum (FWHM) of diffraction peak in radian; , Bragg angle in radian. The local chemical environment around the Zn2+ ions in the various samples was investigated by Zn K-edge X-ray absorption near edge structure (XANES) analysis performed at the 8C beam line at the Pohang Accelerator Laboratory (PAL) in Korea. XANES spectra of samples were obtained at room temperature in the transmission mode, in which standard Zn metal foil was simultaneously measured to obtain an exact calibration of the edge position. Data acquisition and analysis were carried out using standard procedures reported previously [18,19]. 3. Results and discussion We first characterized all of the ZnO nanoparticles used in this study. In order to identify the crystal phase and crystallinity, X-ray diffraction patterns were obtained (Fig. 1). In the 2 range from 20◦ to 80◦ , all of the ZnOs showed typical diffraction peaks of the wurtzite phase (JCPDS No. 36-1451): (1 0 0), (0 0 2), (1 0 1), (1 0 2), (1 1 0), (1 0 3), (2 0 0), (1 1 2), (2 0 1), (0 0 4) and (2 0 2) at 31.7◦ , 34.4◦ , 36.2◦ , 47.5◦ , 56.6◦ , 62.8◦ , 66.3◦ , 67.9◦ , 69.0◦ , 72.5◦ and 76.9◦ , respectively. Although a slight shift of the peak positions of (1 0 1) and (1 0 3) between ZnO20 and ZnO70 was observed, as can be seen in a comparison of Fig. 1(a)–(c) with (d)–(f), the cell parameters of all the samples calculated from the XRD patterns showed the same ˚ values within three significant figures (a = 3.25 A˚ and c = 5.21 A), regardless of their size or surface chemistry. Due to the larger primary particle size of the ZnO70s, the ZnO70 samples (ZnO70,
G.-H. Gwak et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
139
S-ZnO70 and C-ZnO70) clearly exhibited higher intensities with respect to the diffraction peaks compared with the ZnO20 series (ZnO20, S-ZnO20 and C-ZnO20). The crystallite sizes calculated using Scherrer’s equation (Eq. (1)) were approximately 20 nm and 50 nm, respectively, for ZnO20s and ZnO70s. It is notable that the XRD patterns for the ZnO series with the same starting particle size were similar in terms of intensity and peak width, which indicates that no significant dissolution or phase transformation occurred during the surface modification with citrate or l-serine. In the FTIR spectra of the untreated ZnOs (Fig. S1), we observed characteristic peaks at 3450 cm−1 and 485 cm−1 , which can be attributed to OH stretching and Zn O stretching vibrations, respectively. Thus, the surfaces of as-purchased ZnO were thought to possess Zn OH moieties, regardless of the manufacturer. The infrared analysis of the surface-modified ZnO nanoparticles showed typical peaks corresponding to carboxylate and amines, suggesting that the surface of ZnO contained an intact structure of citrate or l-serine. The COO− terminal of l-serine molecules interact with Zn2+ ions at the surface of ZnO nanoparticles in a unidentate manner, which was confirmed by analyzing the wavenumber difference between asymmetric and symmetric stretching mode of carboxylate in Fig. S1. The zeta potential values of ZnO20, S-ZnO20, C-ZnO20, ZnO70, S-ZnO70 and C-ZnO70 at a pH of 7 were determined to be +28.8 ± 0.9, +26.3 ± 0.5, −34.4 ± 1.0, +26.3 ± 0.9, +26.1 ± 0.4 and −41.4 ± 0.6 mV, respectively, as previously reported [20]. This result indicates that the citrate treatment effectively modified the surface of ZnO nanoparticles to have a negative charge, while l-serine treatment maintained the original positive charge of the ZnOs. As we aimed to monitor the particle size and surface chemistry dependent behavior of ZnO nanoparticles when administered through an oral route, we applied a simple model of the physiological conditions of the digestive route. Specifically, the physico-chemical properties of metal oxide or hydroxide nanoparticles are usually more sensitive to pH than enzymatic reactions [21]. Thus, we utilized PBS buffer as a physiological condition without enzymes and adjusted the pH to 1.2, 6.8 and 7.4 to simulate gastric, intestinal, and plasma environments, respectively. For the first step, we monitored the time-dependent pH changes of each PBS buffer upon treatment with ZnOs. All of the different pH solutions exhibited an immediate pH increase (Fig. 2) upon the addition of the ZnOs, which can be attributed to the dissolution of Zn2+ (Eq. (2)) at the surface, resulting in the absorption of H+ and production of OH− . ZnO(s) + H+ (aq) Zn2+ (aq) + OH− (aq)
(2)
In the case of the simulated gastric solution, ZnO increased the pH from 1.2 to 6.5, reaching a plateau within 0.1 h (Fig. 2(A)). The pH values of the simulated intestinal (pH ∼ 6.8) and plasma (pH ∼ 7.4) solutions also exhibited a pH increase, resulting in a final pH value of ∼9.5 at 10 h. The difference of the pH (pH) before and after ZnO treatment was much higher for the gastric condition (pH = 5.6) compared with the intestinal and plasma conditions (pH = 2.6 and 2.1, respectively). As the pH increase depends mainly on the reaction between ZnO and H+ , the pH increase was more remarkable under the gastric condition, which had the largest abundance of H+ ions. While the gastric solution readily obtained pH equilibrium regardless of the size and surface chemistry of ZnO, the pH trends of the intestinal and plasma solutions were highly dependent on the particle size. The pH with the smaller ZnOs (ZnO20s) tended to increase faster than with larger particles (ZnO70s) (Figs. 2(B) and (C)). As smaller particles have larger specific surface areas (SBET = 27.7 m2 /g and 16.6 m2 /g for ZnO20 and ZnO70, respectively), the surface interactions between ZnO nanoparticles and H+ were more favorable with the ZnO20s. In addition, the increase of the pH was also slightly affected by the surface chemistry in which
Fig. 2. Time dependent pH changes of the physiological conditions simulating the pH of (A) the stomach (pH ∼1.2), (B) intestines (pH ∼6.8), and (C) plasma (pH ∼7.4) after incubation with ZnO nanoparticles. Each data point and error bar represents the average and standard deviation, respectively, obtained from triplicate experiments.
the pH with the uncoated ZnO nanoparticles increased faster than with coated nanoparticles. This effect was apparent for the 20 nm sized ZnO nanoparticles under the intestinal and plasma conditions (Figs. 2(B) and (C)). Specifically, ZnO20 (squares) resulted in a larger increase of the pH compared to the coated nanoparticles (circles: S-ZnO20, triangles: C-ZnO20). Thus, the surface coating agents may
140
G.-H. Gwak et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
Fig. 3. Time dependent cumulative dissolution of ZnO nanoparticles under gastric (pH ∼ 1.2) conditions. Each data point and error bar represents the average and standard deviation, respectively, obtained from triplicate experiments.
have blocked efficient interactions between the ZnO surface and H+ , decelerating the dissolution rate. However, the final pH was similar regardless of the surface chemistry, suggesting that the overall reaction equilibrium was not governed by the surface properties. As the particle size and surface chemistry of ZnO nanoparticles were determined to affect the pH increase, we evaluated the dissolution of ZnOs under each condition utilizing AAS and ICPMS. Monitoring the dissolution rates is essential to evaluate the physico-chemical properties of ZnOs and it was previously reported that dissolution strongly affects bioavailability [12]. Fig. 3 shows the time-dependent dissolution of the ZnO nanoparticles under gastric pH condition obtained by AAS. All of the ZnO nanoparticles showed prompt dissolution in the range of 18–30 wt%. Further, the amount of dissolved Zn2+ did not change significantly over a 24 h period, suggesting that the dissolution reached equilibrium at an early stage. This result is highly consistent with our observation of a fast increase of the pH with ZnO nanoparticles in the gastric pH buffer, as shown in Fig. 2(A). Although the cumulative dissolution appeared to be affected by the particle size and surface chemistry, we were unable to identify a statistically significant difference between the samples with a 99% reliability using the Student’s t-test. ZnO dissolution under intestinal (pH ∼ 6.8) and plasma (pH ∼ 7.4) conditions was very small and thus, it was further monitored by ICP-MS (Table 1). The cumulative dissolution measured by ICP-MS was less than 0.01 wt% under both conditions. Upon calculating the theoretical amount of dissolved ZnO based on the change of the pH for each solution, we identified a significant discrepancy between the calculated values and our measurements. Specifically, the gastric, intestinal, and plasma solutions showed pH values of 5.6, 2.6 and 2.1, respectively. Thus, we calculated that 6.310 × 10−3 , 1.581 × 10−8 and 3.950 × 10−9 mol of H+ and 6.310 × 10−9 , 2.506 × 10−6 and 3.137 × 10−6 mol of OH− were produced during the reaction under each condition, respectively. These values correspond to cumulative ZnO dissolutions of 25.3, 0.082 and 0.013 wt% for the gastric, intestinal and plasma conditions, respectively. The calculated value for the gastric condition was fairly comparable with the measured value. However, the calculated values for the intestinal and plasma conditions were much larger than the measured ones. Thus, we hypothesized that the behavior of the ZnO nanoparticles at relatively higher pH values (initial pH values of 6.8 and 7.4) was not only governed by dissolution (Eq. (2)), but also by Zn2+ hydrolysis in an aqueous system. It is well known that Zn2+ forms various hydrolysis products such as Zn2+ , Zn(OH)+ , Zn(OH)2 ,
Fig. 4. X-ray diffraction patterns of the ZnO nanoparticles after treatment for 24 h under physiological conditions simulating (A) the intestines (pH ∼ 6.8) and (B) plasma (pH ∼ 7.4): (a) ZnO20, (b) S-ZnO20, (c) C-ZnO20, (d) ZnO70, (e) S-ZnO70, and (f) C-ZnO70.
Zn(OH)3 − , Zn(OH)4 2− , Zn2 OH3+ , and Zn2 (OH)6 2− , depending on the pH [22]. Indeed, dissolved Zn2+ may have readily formed hydrolysis products which adsorb on the solid ZnO surface, thereby hindering dissolution. According to the Zn2+ hydrolysis curves (Fig. S2), the major hydrolysis products at pH ∼ 6.5 (final pH of the gastric condition) and pH ∼ 9 (final pH of the intestinal and plasma conditions) were Zn(OH)x 2−x and Zn(OH)2 , respectively. In particular, the pH range of 9–11 is optimum to precipitate Zn(OH)2 , which could cover ZnO nanoparticles in the intestinal and plasma conditions. In order to verify the presence of evolved Zn2+ hydrolysis products generated from ZnO nanoparticles under the intestinal and plasma conditions, we collected nanoparticles after reacting for 24 h and measured the resulting XRD patterns. As shown in Fig. 4(A), we could clearly observe a peak at 2 = 11.3◦ in all of the ZnOs reacted under the intestinal condition (pH ∼ 6.8). This peak was assigned as the (2 0 0) peak of the zinc hydroxide phase (Zn5 (OH)6 (CO3 )2 ; JCPDS No. 19-1458, hydrozincite), which is one of the possible hydrolysis products of Zn2+ . As the Zn2+ dissolved from ZnO could readily react with water and atmospheric CO2 , hydrozincite was thought to precipitate on the surface of the ZnO nanoparticles during the reaction. Furthermore, the pH in the intestinal and plasma conditions after ZnO treatment reached 9–10, which is a favorable condition for the precipitation of hydrozincite. ZnO nanoparticles treated in the plasma condition (pH ∼ 7.4) also showed evolution of a hydrozincite phase for the 70 nm samples (Fig. 4(B)). In order to confirm the formation of hydrozincite under simulated physiological conditions, we investigated the X-ray absorption spectra (XAS) of samples at the Zn K-edge. ZnOs with a wurtzite phase consist of ZnO4 tetrahedrons with shared corners, while hydrozincite is composed of both Zn(OH)4 tetrahedrons and Zn(OH)6 octahedrons. The XANES is sufficiently sensitive to detect subtle structural changes around specific elements. Thus, changes of the coordination information around Zn2+ could be obtained from the XANES analysis. We obtained the Zn K-edge XANES spectra of ZnO20 before and after treatment under the plasma simulating pH condition where small and bare ZnO nanoparticles with many reactive sites on the particle surface showed relatively remarkable phase transformation on the surface (Fig. 5). The Zn K-edge XANES spectrum of ZnO20 showed a subpeak at 9661 eV and an intense white line at 9667 eV (solid line in Fig. 5), which were attributed to the electron transition from the 1s to 4p orbital, as reported
G.-H. Gwak et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
141
Table 1 The cumulative dissolution of Zn2+ ions from ZnO nanoparticles (wt%) after treatment for 24 h under physiological conditions simulating the intestines (pH ∼ 6.8) and plasma (pH ∼ 7.4). All of the values were obtained from triplicate experiments. The data stand for the average values and corresponding standard deviations. ZnO20
C-ZnO20
ZnO70
S-ZnO70
C-ZnO70
Intestine (pH 6.8) (wt%) 0.00023 ± 0.00004 0.00514 ± 0.00028
S-ZnO20
0.00193 ± 0.00007
0.00018 ± 0.00001
0.00122 ± 0.00004
0.00291 ± 0.00010
Plasma (pH 7.4) (wt%) 0.00015 ± 0.00003 0.00879 ± 0.00034
0.00087 ± 0.00006
0.00011 ± 0.00004
0.00111 ± 0.00016
0.00108 ± 0.00014
Fig. 5. Zn K-edge X-ray absorption near edge structure (XANES) spectra of ZnO20 before (solid line) and after (open circles) treatment for 24 h under the plasma pH condition (pH ∼ 7.4).
previously [23,24]. Although the overall XANES pattern of ZnO20 in the plasma condition (open circle in Fig. 5) was similar to that of bare ZnO20, closer inspection revealed a slight difference. Specifically, the subpeak and white line of ZnO in the plasma condition increased from 9660.8 to 9661.3 eV and from 9666.6 to 9666.9 eV, respectively, compared to untreated ZnO. Furthermore, the white line intensity increased by ∼10% after plasma treatment. Importantly, the position and intensity of the white line are closely related to the coordination number. It has been reported that the peaks around the main edge shift to a higher energy and the white line intensities increase as the coordination number of zinc compounds increase [25]. Thus, the enhanced white line intensity observed in this study reflects expansion of the average coordination number around Zn2+ , which may be due to the evolution of hydrozincite with a Zn(OH)6 octahedra structure. These results are in good agreement with the XRD analysis in which a new XRD peak assigned to hydrozincite was detected. Therefore, our results suggest that a hydrozincite phase evolved from ZnO nanoparticles in the simulated plasma condition. As we observed that the ZnO nanoparticles partially dissolved and produced a certain amount of hydrozincite, we next examined the overall crystalline phase and crystallite size of the ZnO nanoparticles under each simulated condition. Fig. S3 shows the XRD patterns of the ZnOs after treatment for 24 h. All of the patterns showed a clear wurtzite phase, which was indifferent from the untreated ZnOs shown in Fig. 1, indicating that there was overall preservation of the phase during treatment. Although it is known that Zn2+ ions readily precipitate with phosphate ions, we could not find any evidence for the evolution of zinc phosphate in the structural analyses including XRD. It is thought that a small quantity of zinc phosphate could be formed, but the evolution of hydrozincite was prevalent. To analyze the time-dependent changes of the crystallinity, the crystallite sizes of all of the ZnOs at each time point were calculated using Scherrer’s equation with the (1 0 1) peaks. Each ZnO peak largely corresponded to initial crystallite sizes of
20 and 50 nm for the ZnO20s and ZnO70s, respectively, regardless of the time or pH (Fig. S4). As only a slight dissolution of Zn2+ was detected (Table 1), the ZnO nanoparticles were considered to maintain their crystallinity and size under both intestinal (pH ∼ 6.8) and plasma (pH ∼ 7.4) conditions. Interestingly, the ZnO nanoparticles in the gastric (pH ∼ 1.2) condition did not exhibit a significant decrease of the crystallite size despite the obvious dissolution (18–30 wt%) of Zn2+ . We could explain this phenomenon by simply calculating the particle size of both ZnO20 and ZnO70 after 18–30 wt% dissolution. Specifically, as the dissolution would naturally initiate at the surface of the nanoparticles, we hypothesized that the outermost sphere of the particles dissolved during the reaction. The dissolution of 18–30 wt% for the 20 nm and 70 nm particles resulted in a reduction of the diameter to 18 nm and 65 nm, respectively, meaning that there was a >90% maintenance of the original crystallite size. Thus, even dissolution of ZnO nanoparticles up to 30 wt% did not significantly affect the crystallite size. Finally, in order to confirm whether the ZnO particles maintained their original morphology, we obtained SEM images of all of the ZnOs before and after treatment under gastric conditions for 24 h. The original ZnO20s (Fig. S5(a)–(c)) exhibited a smooth and round-shaped primary particles with size of 28.03 ± 5.62 nm, while the original ZnO70s (Fig. S5(d)–(f)) showed a similar spherical morphology with a size of 77.45 ± 21.05 nm. After treatment for 24 h, we did not observe any significant changes of the morphology or surface roughness for all of the ZnO samples evaluated. The particle sizes of the ZnO20s and ZnO70s after treatment at a pH of 1.2 were determined to be 24.36 ± 7.19 nm (Fig. S5(a )–(c )) and 79.49 ± 21.68 nm (Fig. S5(d )–(f )), respectively. There was no significant difference between the 95% confidence intervals for the two conditions according to the Student’s t-test. The preservation of the particle size and morphology was also similar for the pH 6.8 and pH 7.4 conditions (Figs. S6 and S7). 4. Conclusions In order to investigate the physico-chemical properties of ZnO nanoparticles as a function of the particle size and surface chemistry, we observed the time-dependent physico-chemical changes in PBS buffer solutions simulating the pH of the stomach, intestine, and plasma. The ZnO nanoparticles were determined to enhance the pH in the physiological pH conditions to a neutral (in the case of the gastric condition) or basic range (in the case of the intestinal and plasma conditions), revealing a dependency on the size and surface chemistry. The dissolution of ZnO was significant under the gastric pH condition, whereas under the intestinal and plasma conditions, it was trivial. As the dissolution of ZnO under intestinal and plasma conditions was much less than expected based on the observed pH increase, we hypothesized that a phase transformation from ZnO to Zn(OH)2 occurred. Consistently, X-ray diffraction and X-ray absorption spectroscopy revealed the possible formation of hydrozincite on the surface of ZnO during the reaction. Based on Scherrer’s equation and the SEM analysis, the size and morphology of the ZnO nanoparticles were not seriously altered by the physiological pH conditions. Thus, we concluded that the ZnO nanoparticles mainly dissolved in the gastric condition and that the other physiological
142
G.-H. Gwak et al. / Colloids and Surfaces B: Biointerfaces 127 (2015) 137–142
conditions caused a phase transformation of the ZnO nanoparticles. While the particle size and surface chemistry are thought to affect interface interactions between nano- and bio-systems, these factors do not appear to govern the overall physico-chemical properties of the ZnO nanoparticles. Acknowledgements This work was financially supported by the Ministry of Science, ICT and Future Planning (MSIP) Korea Government and the Korea Industrial Technology Association (KOITA) as “A study on the programs to support a collaborative research among industry, academia and research institutes” (KOITA-2013-2-5), by Rural Development Administration (RDA, Korea) as “Cooperative Research Program for Agriculture Science & Technology Development” (PJ0105022015), and by a National Research Foundation of Korea (NRF) grant funded by the MSIP, Korea Government (20100024370). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.colsurfb. 2015.01.021. References [1] J.H. Yuan, Y. Chen, H.X. Zha, L.J. Song, C.Y. Li, J.Q. Li, X.H. Xia, Colloids Surf. B: Biointerfaces 76 (2010) 145.
[2] P. Whittaker, Am. J. Clin. Nutr. 68 (1998) 442S. [3] S.R. Pinnell, D. Fairhurst, R. Gillies, M.A. Mitchnick, N. Kollias, Dermatol. Surg. 26 (2000) 309. [4] R. Tankhiwale, S. Bajpai, Colloids Surf. B: Biointerfaces 90 (2012) 16. [5] S.K. Dhoke, A.S. Khanna, T.J.M. Sinha, Prog. Org. Coat. 64 (2009) 371. [6] M. Law, L.E. Greene, J.C. Johnson, R. Saykally, P. Yang, Nat. Mater. 4 (2005) 455. [7] M. Behrens, F. Studt, I. Kasatkin, S. Kühl, M. Hävecker, F. Abild-Pedersen, S. Zander, F. Girgsdies, P. Kurr, B.-L. Kniep, Science 336 (2012) 893. [8] L. Kumari, W.Z. Li, Cryst. Res. Technol. 45 (2010) 311. [9] A. Janotti, C.G. Van de Walle, Rep. Prog. Phys. 72 (2009) 126501. [10] T. Yumak, F. Kuralay, M. Muti, A. Sinag, A. Erdem, S. Abaci, Colloids Surf. B: Biointerfaces 86 (2011) 397. [11] M. Heinlaan, A. Ivask, I. Blinova, H.-C. Dubourguier, A. Kahru, Chemosphere 71 (2008) 1308. [12] H.-J. Paek, Y.-J. Lee, H.-E. Chung, N.-H. Yoo, J.-A. Lee, M.-K. Kim, J.K. Lee, J. Jeong, S.-J. Choi, Nanoscale 5 (2013) 11416. [13] N. Padmavathy, R. Vijayaraghavan, Sci. Technol. Adv. Mater. 9 (2008) 035004. [14] A. Albanese, P.S. Tang, W.C. Chan, Annu. Rev. Biomed. Eng. 14 (2012) 1. [15] M.D. Newman, M. Stotland, J.I. Ellis, J. Am. Acad. Dermatol. 61 (2009) 685. [16] J.Y. Kwon, S.Y. Lee, P. Koedrith, J.Y. Lee, K.-M. Kim, J.-M. Oh, S.I. Yang, M.-K. Kim, J.K. Lee, J. Jeong, Mutat. Res. Genet. Toxicol. Environ. Mutagen. 761 (2014) 1. [17] B.D. Cullity, Am. J. Phys. 25 (1957) 394. [18] S.M. Paek, H. Jung, Y.J. Lee, M. Park, S.J. Hwang, J.H. Choy, Chem. Mater. 18 (2006) 1134. [19] S.M. Paek, H. Jung, M. Park, J.K. Lee, J.H. Choy, Chem. Mater. 17 (2005) 3492. [20] K.-M. Kim, T.-H. Kim, H.-M. Kim, H.-J. Kim, G.-H. Gwak, S.-M. Paek, J.-M. Oh, Toxicol. Environ. Health Sci. 4 (2012) 121. [21] J.Y. Kim, S.-J. Choi, J.-M. Oh, T. Park, J.-H. Choy, J. Nanosci. Nanotechnol. 7 (2007) 3700. [22] C.F. Baes, R.E. Mesmer, Hydrolysis of Cations, Wiley, New York, 1976. [23] S. Larcheri, C. Armellini, F. Rocca, A. Kuzmin, R. Kalendarev, G. Dalba, R. Graziola, J. Purans, D. Pailharey, F. Jandard, Superlattices Microstruct. 39 (2006) 267. [24] J. Rose, I. Moulin, A. Masion, P.M. Bertsch, M.R. Wiesner, J.-Y. Bottero, F. Mosnier, C. Haehnel, Langmuir 17 (2001) 3658. [25] F. Sowrey, L. Skipper, D. Pickup, K. Drake, Z. Lin, M. Smith, R. Newport, Phys. Chem. Chem. Phys. 6 (2004) 188.
