Quantitative analysis of surface sites on carbon dots and their

Jun 27, 2019 - ... analysis of surface sites on carbon dots and their interaction with ... CDs with and without Fe(III) were carried out and the stabi...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Cite This: Anal. Chem. XXXX, XXX, XXX−XXX

Quantitative Analysis of Surface Sites on Carbon Dots and Their Interaction with Metal Ions by a Potentiometric Titration Method Zhe Wang,†,‡ Yu Xie,§ Zhen Lei,† Yuexiang Lu,*,‡ Guoyu Wei,‡ Shuang Liu,‡ Chao Xu,‡ Zhicheng Zhang,∥ Xiangke Wang,†,⊥ Linfeng Rao,*,∥ and Jing Chen*,‡

Downloaded via UNIV AUTONOMA DE COAHUILA on July 20, 2019 at 11:48:05 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

The MOE Key Laboratory of Resource and Environmental System Optimization, School of Environment and Chemical Engineering, North China Electric Power University, Beijing 102206, People’s Republic of China ‡ Institute of Nuclear and New Energy Technology, Collaborative Innovation Center of Advanced Nuclear Energy Technology, Beijing Key Lab of Radioactive Waste Treatment, Tsinghua University, Beijing 100084, People’s Republic of China § Department of Environmental Engineering and Earth Sciences, Clemson University, 342 Computer Court, Anderson, South Carolina 29625, United States ∥ Chemical Sciences Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, California 94720, United States ⊥ Hebei Key Lab of Power Plant Flue Gas Multi-Pollutants Control, Department of Environmental Science and Engineering, North China Electric Power University, Baoding 071003, People’s Republic of China S Supporting Information *

ABSTRACT: Carbon dots (CDs) possess abundant functional groups on their surface which are related to their application in various fields such as sensing, imaging, and catalysis. Understanding the amount and properties of these functional groups and their interaction with metal ions is essential but has posed longstanding challenges because of the diverse and complex structures of CDs. In this work, potentiometric titration is demonstrated as an effective method to figure out the categories and amounts of functional groups. Surface complexation modeling with the FITEQL program was applied to the quantification of the surface sites on CDs with the titration data. Then with the obtained molar concentrations of the surface sites, the pKas of these surface sites were calculated with the Hyperquad program. Finally, titration experiments of CDs with and without Fe(III) were carried out and the stability constants of Fe(III) and ArgCDs were simulated on the Hyperquad program. By utilizing the stability constants, the distribution of Fe(III) species at different pHs and the concentrations of Fe(III) and CDs were also investigated. This potential method might be used for characterizing the surface sites on other CDs or even other soluble nanoparticles as well as for investigating the interactions of the surface sites with different metal ions.

1. INTRODUCTION Carbon dots (CDs) are a novel class of water-soluble fluorescent carbon nanoparticles1 with abundant functional groups on their surface. They have contributed to advances in bioimaging,2 photocatalysis,3 sensors,4,5 and adsorption areas6 due to their low toxicity, favorable energy and charge transfer, and unique optical characters.7,8 Especially, by taking advantage of their abundant functional groups and fluorescent property, researchers have widely applied CDs for the fluorescent detection of metal ions which possess a deleterious influence on the environment and human health.9−11 Recently, there have also been increasing reports on applying CDs for the adsorption of metal ions.12−14 However, because of the complex and diverse structures and the inhomogeneous composites of the CDs, the detection/adsorption properties of CDs are difficult to be well understood, predicted, or optimized.15,16 © XXXX American Chemical Society

With changes in raw materials, synthesis method/conditions, and postsynthesis functionalization, the type and amount of these functional groups will be ever-changing and result in diverse properties of the CDs, such as water solubility, fluorescence intensity, and interaction with target molecules.17−19 Commonly used analysis methods, such as Fourier transform infrared (FT-IR) and X-ray photoelectron (XPS), can only give qualitative characterization or semiquantitative comparison of the functional groups on different kinds of CDs but cannot provide the quantitative information on the interaction between CDs and metal ions.20 Fluorescent titration methods21 are useful for quantitative analysis of the fluorescent response of CDs to metal ions, but they cannot Received: March 8, 2019 Accepted: June 27, 2019 Published: June 27, 2019 A

DOI: 10.1021/acs.analchem.9b01225 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

technique could be broadened to other carbon dots prepared with various raw materials and other water-soluble nanoparticles, as well as the interactions between nanoparticles and metal ions.

reflect the real interaction between them. This is because the fluorescent quenching process is complicated, which may be caused by several different mechanisms. Also, the fluorescent titration methods are not suitable for investigating metal ions with strong intrinsic fluorescence or which cause weak fluorescent response to the CDs. As a result, developing a general method for the quantitative analysis of the surface sites on CDs and their interaction with metal ions is of great importance not only for understanding the relationship between the surface chemistry and diverse properties but also for optimizing the synthesis and application of the CDs. Potentiometric titration is a powerful and cost-effective method for quantifying the functional groups existing on different materials22 and modeling the stability constants between ligands and metal ions.23,24 In comparison with the other methods mentioned above, the titration method could provide more detailed information and in-depth properties, such as the content and pKa of the functional groups. It could also reveal the formation of stable and deprotonated complexes between ligands and metal ions, which could guide the applications in adsorption, migration, and other areas.25,26 It has been widely used for confirming the pKas and stability constants of small molecular ligands27 in which the structure and groups are quite clear. Recently, this method has been broadened to nanoparticle materials functionalized with certain small molecules for quantifying the redox properties and modeling the surface reaction kinetics.15,28−30 In the field of carbon nanomaterials, however, the titration method is still under development because of the more complex species and amounts of functional groups. The successful attempt with graphene oxide (GO) proved the feasibility of this method for carbon materials.31 Xie et al. employed potentiometric titration data with surface complexation modeling methods to simulate the pKas of carboxylate functional groups on GO. The stability constants of Eu(III) and U(VI) to GO were further calculated with the assistance of adsorption data. However, this method is not suitable for investigating the interaction between CDs and metal ions, because the CDs are much smaller with excellent solubility and cannot be separated as easily as GO. Mesquita’s group32 applied the titration data for charactering the acid functional groups on one kind of CDs, while the interaction of CDs and metal ions were not discussed in that paper. In this study, we propose a novel strategy by treating the CDs as a kind of special molecule with complex functional groups. The acid/base potentiometric titrations of CDs with and without metal ions were conducted for further simulation. Two kinds of representative CDs denoted as ArgCDs and GlyCDs were prepared with citric acid and different amino acids. Surface complexation modeling with the FITEQL program was applied for the quantification of the surface sites on CDs with the titration data. Then with the molar concentrations of the surface sites obtained above, the pKas of these surface sites were calculated with the Hyperquad program. Finally, the titration experiments of ArgCDs with and without Fe(III) were carried out and the stability constants of Fe(III) and ArgCDs were simulated on the Hyperquad program. By utilizing the stability constants, the distribution of Fe(III) species at different pHs and the concentrations of Fe(III) and ArgCDs were also investigated. To the best of our knowledge, this is the first quantitative analysis of functional groups on different CDs and calculation of the stability constants of metal ions with CDs systematically with potentiometric methods. We believe this potential

2. EXPERIMENTAL SECTION 2.1. Materials. Citric acid, L-arginine (Arg), and L-glycine (Gly) were purchased from Aladdin. Sodium perchlorate and Fe(NO3)3 metal salts were obtained from Alfa Aesar. Sodium hydroxide titrant solution was purchased from Metrohm. Perchloric acid was obtained from Sigma-Aldrich. Boiled/ cooled Milli-Q water was used for the preparation of all solutions. All reagents were reagent grade or higher and were used with no further purification. 2.2. Characterizations. Transmission electron microscope (TEM) images were obtained on a JEM-2010 (JEOL) instrument with an accelerating voltage of 120 kV. Fluorescence spectroscopy of CDs was carried out with a FluoroMax-4 spectrophotometer. The KBr pellet method was applied for FT-IR spectra on a Nicolet NEXUS 470 FT-IR spectrophotometer. A 250XI spectrometer was employed for X-ray photoelectron spectroscopy (XPS) analysis of CDs. 2.3. Preparation of Carbon Dots. The carbon dots (CDs) were synthesized through a simple hydrothermal method. In brief, 5 mM of citric acid and 10 mM of amino acid (Arg or Gly) were sufficiently dissolved in 10 mL of deionized water and then transferred to a 50 mL autoclave made of poly(tetrafluoroethylene) (Teflon). The autoclave was treated in an oven for 6 h at 180 °C and cooled down naturally. After that, carbon dots with a yellow color were placed in a membrane bag (Mw = 1000 Da) to remove the unreacted precursors. After 24 h, the solution was taken out carefully and dried in a vacuum oven at 80 °C for 2 days to obtain the solid products. To make identification easier, the carbon dots made from different amino acids were abbreviated as ArgCDs and GlyCDs. They were used for further characterization and application. 2.4. Potentiometric Titration. Potentiometric titrations of carbon dots as well as of carbon dot and metal ion solutions were investigated on an autotitration system at a fixed temperature of 25 ± 0.1 °C by circulating the water bath. The system was composed of a double-jacketed glass titration vessel, a computer, and a pH meter (Model 713, Metrohm) connected to a pH electrode (Model 8102, Orion) for the recording of the electromotive force (EMF, in millivolts). Because of the low solubility of KClO4, we replaced the original electrode inner solution (3 M KCl) with 1 M NaCl. To prevent CO2 interference, pure Ar gas was employed at the headspace of the titration vessel. Before each titration experiment, the electrode was calibrated with standard acid and base solutions to acquire parameters for calculating the concentration of H+ with the EMF values. The EMF values can be expressed by the equations E = E0 + RT /F ln [H+] + γH[H+]

(1)

E = E0 + RT /F ln (K w /[OH−]) + γOH[OH−]

(2)

where R, F, and T are the gas constant, Faraday constant, and the temperature in K, respectively. Kw = [H+][OH−] and the γH[H+] and γOH[OH−] terms represent the electrode junction potentials for the hydrogen or the hydroxide ions, assumed to be proportional to the concentration of the hydrogen or hydroxide ions. B

DOI: 10.1021/acs.analchem.9b01225 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry Multiple titrations of carbon dots with different concentrations (0.162, 0.324, and 0.648 g/L) in 0.1 M NaClO4 were conducted as the pH was changed from 2.5 to 11. The titration data were analyzed with FITEQL 4.0 to obtain the surface site density of the protonated groups and deprotonated groups on CDs. The net proton concentration (Net[H+]) in the suspension was calculated with eq 3 Net[H+] = (ca − cb)/v

(3)

where ca is the total [H+] added, cb is the total [OH−], and v is the total volume of the suspension in the titration. In addition, solutions containing different amounts of ferric ions were titrated with a solution of ArgCDs in 0.1 M NaClO4. The titration data were further calculated with the Hyperquad 2008 program to acquire the stability constants of the complex between ArgCDs and ferric ions.33,34

3. RESULTS AND DISCUSSION 3.1. Determination of the Surface Site on Carbon Dots. 3.1.1. Potentiometric Titration of Carbon Dots. With the hydrothermal method, two kinds of amino acid (L-arginine and L-glycine) and citric acid were selected as raw materials for the preparation of carbon dots. For convenience, the carbon dots prepared from different amino acids were named ArgCDs and GlyCDs. They were first characterized through various means. As shown in Figure S1, the TEM images showed that the two kinds of CDs have a uniform particle size and the fluorescence spectra displayed that both of them possessed fluorescent properties and excitation-dependent behavior. ArgCDs and GlyCDs also contained a great number of functional groups, as determined through the FT-IR and XPS analysis results (Figure S2). In this paper, the potentiometric titration method was applied for the quantitative analysis of the functional groups on the surface of CDs. The molar ratio between citric acid and amino acid was 1:2, and the carboxyl (−COOH) and amino (−NH2) groups dehydrated to form the complex CDs at a certain temperature and pressure. From the chemical formulas given in Figure 1a,b, the theoretical ratios of −COOH and −NH2 were about 5:4 on ArgCDs and 5:2 on GlyCDs. The pH values tested by a pH meter were about 6.5 for ArgCDs and 3.5 for GlyCDs, which agreed well with the raw ratio and also indicated the neutral and acidic properties, respectively, of the two kinds of CDs. Then the potentiometric titration method was used for further quantitative analysis. Figure 1c,d gives the titration results of ArgCDs and GlyCDs at different concentrations. For a concentration of 0.324 g/L in Figure 1c, the initial pH was adjusted to about 2.7 as needed in the titration experiment. A volume of NaOH was used up to 4.3 mL instead of 3.9 mL for the ideal system to reach pH 7, indicating the faint acidity of ArgCDs. The consumption of NaOH was about 4.3 mL to make pH 7 in Figure 1c, and the variation tendency of the three curves slowed down with the increasing concentration of ArgCDs. This should be caused by the existence of acidic (−COOH) and basic (−NH2 and −NH) functional groups on ArgCDs. However, the phenomenon of GlyCDs was not the same. To make the pH arrive at 7, the consumption of NaOH solution was increased from 5.9 to 10.1 mL with a rising concentration of GlyCDs, which showed that the GlyCDs was a kind of acidic carbon dot (Figure 1d). This can be easily understood from the fragment structure in Figure 1b, where the basic −NH2 groups on Gly were occupied to form amide

Figure 1. Possible formation mechanisms of (a) ArgCDs and (b) GlyCDs. Potentiometric titration results of the different concentrations of (c) ArgCD and (d) GlyCD solutions. INaClO4 = 0.1 mol/L, T = 25 ± 0.1 °C.

bonds with −COOH and large numbers of redundant −COOH groups were exposed on the GlyCDs. The acid and base information on the functional groups on CDs could be obtained from the titration results and showed no difference with the pH of the CD solutions. 3.1.2. Surface Site Modeling of Carbon Dots with FITEQL. The FITEQL program was selected for the description of the category and amount of the functional groups on CDs with the titration data. As required in the program, the net proton concentration (Net[H+]) was calculated for the y axis in Figure 2a,b. Figure 2a shows that, with an increasing concentration of ArgCDs, the pHPZC (point of zero charge) was intersected at one point at 6.28, and the ArgCDs had good buffer capacity in both the positive and negative Net[H+]. Furthermore, the pHPZC value of 6.28 again verified the neutral character of ArgCDs. As presented in Figure 2b, with an increase in the concentration of GlyCDs from 0.162 to 0.648 g/L, the titration curves moved to more negative Net[H+]. This meant that to reach the same pH value, for a higher concentration of GlyCDs, more NaOH was needed. In addition, the pHPZC value of GlyCDs decreased from 3.41 to 3.0 with an increase in concentration. This change in the pHPZC indicated the existence of acidic functional groups on GlyCDs and a similar acidic nature has already been observed for graphene oxide by previous studies.30,35,36 The surface complexation modeling (SCM) was applied to simulate the titration curves to quantize the concentrations (in molarity) of functional groups on CDs.37 Although multiple functional groups might exist on the CDs, to simplify the simulation process, the SCM applied two kinds of surface sites to simulate the titration data. For the ArgCDs, the two kinds of surface sites were assigned as acidic and basic sites, noted as site A and site B. For the GlyCDs, because of their acidic nature, the surface sites were appointed as more strongly acidic sites and more weakly acidic sites, named site S and site W, respectively. As shown in Figure 2c,d, the agreements between the model and data were quite good at various concentrations. C

DOI: 10.1021/acs.analchem.9b01225 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 2. Base titration (dots) and modeling (lines) fitted for different concentrations of (a) ArgCDs and (b) GlyCDs (0.162 to 0.648 g/L). Linear relationship of (c) ArgCD and (d) GlyCD mass concentration and modeled surface site density. Note that the data of the y axis in (c) and (d) were the output of the FITEQL program.

deprotonation, first from AH to A− and then from BH to B. This was quite in agreement with the acidic and basic properties of sites A and B. However, the deprotonation process for GlyCDs in Figure 3b was different, and before the terminated deprotonation of site S, site WH launched its deprotonation journey. This was understandable because of the acidic properties of both site S and site W. The protonation constants (pKas) of the surface sites were accordingly calculated, as displayed in Table 3. In comparison with the pKas of CDs and different functional groups, it was not hard to find that sites A and B should be the carboxylic and the amino groups existing as different forms on ArgCDs, respectively (Figure 3c). The pKas for these functional groups may exhibit significant differences depending on their position and neighboring groups anchored on the surface of ArgCDs. For GlyCDs, the pKas could be the two kinds of carboxylic groups under different circumstances (Figure 3d). As viewed against the structure of citric acid and glycine, the carboxylic groups on the GlyCDs should possess different pKas. In addition, due to the low content of amino groups on the GlyCDs as well as the high acidity of the GlyCDs, the GlyCDs expressed the property of only carboxyl groups. Thus, the pKas for the surface sites on ArgCDs and GlyCDs were modeled with the titration method and they agreed well with the properties and titration results of the CDs.38,39 3.2. Interactions between ArgCDs and Fe(III). Carbon dots have been widely used for the detection and adsorption of metal ions, especially ferric ions, as they can interact with carbon dots and quench the fluorescence of carbon dots. The unclear mechanism makes the detection/adsorption properties of CDs difficult to be well understood, predicted, or optimized. Here, the ArgCDs were selected to combine with Fe(III), and the stability constants and the species distribution between ArgCDs and Fe(III) were modeled with the Hyperquad program. To some extent, we hoped this insight into the

The molar concentrations of the surface sites for ArgCDs and GlyCDs with different concentrations are optimized and displayed in Figure 2c,d, respectively. It is clear that the amount of sites (mol/L) increased linearly with the increasing concentration of CDs (g/L) and the slope (mol/g) provided the exact amount of the surface sites on CDs (Tables 1 and 2). Thus, the surface sites on ArgCDs and GlyCDs were simulated and quantized with this FITEQL program. Table 1. Surface Site Modeling Describing ArgCDs with the FITEQL Program number ArgCDs (g/L) site A (mol/L) site B (mol/L)

1

2

3

slope (mol/g)

0.162 0.00293 0.0015

0.324 0.0136 0.00319

0.648 0.0297 0.00536

0.00536 0.00352

Table 2. Surface Site Modeling Describing GlyCDs with the FITEQL Program number GlyCDs (g/L) site S (mol/L) site W (mol/L)

1

2

3

slope (mol/g)

0.162 0.0011 0.000445

0.324 0.0022 0.00089

0.648 0.0043 0.00178

0.00657 0.00275

3.1.3. Modeling the pKas of the Carbon Dots. The molar concentration results fitted above were imported into the Hyperquad program to model the pKas of the surface sites and the protonation process. The titration data with different concentration of CDs were modeled together with the Hyperquad program (Figure 3 and Figure S3). They could be best fitted with the modeled curves in Figure 3. For ArgCDs in Figure 3a, the modeling results contained two steps of D

DOI: 10.1021/acs.analchem.9b01225 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 3. pKa modeling results of the surface sites on (a) ArgCDs and (b) GlyCDs with a concentration of 0.324 g/L by the Hyperquad program. Possible structure and classification of surface sites on (c) ArgCDs and (d) GlyCDs.

3.2.2. Stability Constant Modeling of ArgCDs and Fe(III). The representative titration experiments of the complexation of Fe(III) with ArgCDs are shown in Figure 5. To make the simulation results more accurate, the titration results with various concentrations of ArgCDs and Fe(III) were applied for modeling. Fitting with the titration data, the best model included the formation of two Fe(III) complexes, Fe(OH)A+ and Fe(OH)B2+. The stability constants of the complexes were calculated and are exhibited in Table 4.40,41 With an increasing concentration of ArgCDs (Figure 5a), more surface sites were offered for combination with Fe(III), and the percentage of Fe(OH)A+ and Fe(OH)B2+ increased along with a decrease in Fe(OH)2+ and Fe(OH)2+. As presented in Figure 5b, with increasing concentration of Fe(III), the Fe(III) became excess in comparison with the fixed amount of ArgCDs, causing a slowly increasing percentage of the hydrolysate of Fe(III) and a decrease in Fe(OH)A+ and Fe(OH)B2+ complexes in the meantime. 3.2.3. Species Distribution of ArgCDs and Fe(III). Generally, the distribution of the metal−ligand complexes in the suspension was controlled by the pH value of the aqueous systems. Herein, the stability constants calculated above were applied in the Hyss2009 program to forecast the species distribution in solutions of different pH. The curves of different complexes of Fe(III) in the range of pH 2−8 are given in Figure 6. At the initial pH 2, great numbers of H+ occupied the surface sites of ArgCDs and about 65% Fe(III) existed as ferric iron (Fe3+). The one-stage hydrolysis to Fe(OH)2+ also existed at this pH and half of them combined with site A on ArgCDs to form the complex Fe(OH)A+. With a slowly increasing pH of the system, the amount of Fe3+ began to decrease and transformed into more Fe(OH)2+ and Fe(OH)2+. The spare Fe(OH)2+ went to site B to generate the complex Fe(OH)B2+. As a sustainable growth of the pH, the peak of Fe(OH)A+ at pH 3.5, the disappearance of Fe3+, and the formation of Fe(OH)3 at pH 4.5 were discovered successively. It was at pH 5.5 that the maximum amount of Fe(OH)B2+ was observed and more than 60% of Fe(III)

Table 3. Protonation Constants (pKas) of Surface Sites on ArgCDs and GlyCDs ArgCDs pKa

GlyCDs

site A

site B

site S

site W

3.12

9.18

2.16

4.98

interaction could help to reveal the inner mechanism of CDs and metal ions. 3.2.1. Potentiotitration of Carbon Dots and Metal Ions. The titration process for ArgCDs with Fe(III) solutions was carried out first. During the titration, a visible color change of the suspension was observed at higher pH, which indicated the formation of the hydrolyzed metal ions. Thus, the titration data below pH 7 were selected for further analysis and are presented in Figure 4. The titration curve shifted to the right

Figure 4. Potentiometric titration curves of ArgCDs with and without Fe(III) metal ions. INaClO4 = 0.1 mol/L, T = 25 ± 0.1 °C.

after the addition of Fe(III), and this was because the hydrogen ions on the surface sites of ArgCDs were replaced by Fe(III) and they were released into the suspension, meaning that more NaOH solution was needed to reach the same pH in the titration process. This phenomenon implied that the interactions between the surface sites on ArgCDs and Fe(III) and the stability constants were modeled accordingly. E

DOI: 10.1021/acs.analchem.9b01225 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 5. Stability constants modeling results using the Hyperquad program of the ArgCDs−Fe(III) system with different concentrations of (a) ArgCDs (the concentration of Fe(III) was fixed at 0.32 mM) and (b) Fe(III) (the concentration of ArgCDs was fixed at 0.324 g/L).

7a, the concentration of ArgCDs was fixed at 1 mM and that of Fe(III) covered a wide range from 0.01 to 100 mM. From left to right, the proportion of the two complexes decreased slowly and disappeared finally with the increasing concentration of Fe(III). This was because most of the Fe(III) was hydrolyzed to Fe(OH)3 or other species at higher concentration, leading to the much lower ratio of the complexes. Figure 7b is the Fe(OH)A+ and Fe(OH)B2+ complex distribution with a fixed concentration (1 mM) of Fe(III) and an increasing amount of ArgCDs. From right to left, the ratio of the two complexes increased and became the main species of Fe(III), which could be attributed to the fast-growing surface sites on ArgCDs. Most of the Fe(III) could combine with the redundant site A and site B on ArgCDs. In addition, the proportion between ArgCDs and Fe(III) could also affect the species distribution of Fe(III). The portrait orientation of Figure 7 shares the same proportion and it shows the similar changing tendencies of complexes Fe(OH)A+ and Fe(OH)B2+. With a decreasing proportion of AgCDs, the ratio of the two complexes became lower and lower and disappeared at last. At a higher value of CDs:Fe(III), the ratio of the two complexes was not the same, which was because the low concentration of Fe(III) decreased the probability of molecular collision between Fe(OH)2+ and surface sites on ArgCDs. It was more difficult to form the complexes at a lower concentration of Fe(III).

Table 4. Stability Constants Calculated by the Hyperquad Program for the Surface Sites of ArgCDs with Fe(III) protonation equilibria −

H + A → AH H+ + B → BH− Fe3+ + OH− + A− → Fe(OH)A+ Fe3+ + OH− + B → Fe(OH)B2+ Fe3+ + OH− → Fe(OH)2+ Fe3+ + 2OH− → Fe(OH)2+ +

log K 3.12 ± 0.03 9.18 ± 0.03 15.68 ± 0.10 19.56 ± 0.07 11.14 21.2

Figure 6. Species distribution of Fe(III) in the ArgCDs−Fe(III) system with an increase in pH from 2 to 8.

existed as Fe(OH)2+. As the pH grew higher, Fe(OH)A+ and Fe(OH)B2+ vanished gradually and Fe(OH)3 changed to be the main species in the solutions at pH 8. Not only pH but also the concentration and proportion of Fe(III) and ArgCDs could influence the distributions of Fe(III) species. To make the graphs clearer and more concise, the Fe(OH)A+ and Fe(OH)B2+ complex curves were selected to demonstrate the variation tendency in Figure 7; the complete graphs are given in Figure S4. As shown in Figure

4. CONCLUSION In summary, with the potentiometric titration method, two representative carbon dots (CDs) were selected for quantifying the surface sites on CDs and calculating the pKas of these sites, as well as the stability constants of CDs with metal ions. The CDs prepared from different amino acids showed distinct functional groups and properties. For neutral ArgCDs, the pKas F

DOI: 10.1021/acs.analchem.9b01225 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 7. Species distribution of Fe(III) combined with the surface sites on ArgCDs at various concentrations of Fe(III) and ArgCDs: (a) different concentrations of Fe(III) with a fixed 1 mM of ArgCDs; (b) different concentrations of ArgCDs with a fixed 1 mM of Fe(III).

■

for sites A and B were 3.12 and 9.18, which could be attributed to the carboxylic and amino groups on CDs, respectively. For acidic GlyCDs, the pKas for sites S and W were 2.16 and 4.98, which could be ascribed to two kinds of carboxylic group under different circumstances. The modeled pKas agreed very well with the titration data and also matched with the properties of the raw materials of these two CDs. With the molar concentration and the pKas of the CDs, the stability constants of ArgCDs with Fe were modeled and the best-fitted model included two Fe(III) complexes, Fe(OH)A+ and Fe(OH)B2+. The log K values for the two complexes were 15.68 and 19.56, respectively. By utilizing the stability constant calculated above, the species distribution of the Fe(III) complexations was analyzed at different pHs of the suspension and concentrations of ArgCDs and Fe(III). The species distribution of Fe(III) changed with the concentration and relative proportion of ArgCDs and Fe(III). The method established above is a potential method and could provide a novel way to analyze the properties of the functional groups on carbon dots prepared with various materials or methods. It also could be used for the study of the interactions between these carbon dots and different kinds of metal ions, which could provide useful information in many application areas of the CDs, such as sensing, adsorption, and photocatalysis. Furthermore, the application of this method may also be broadened to investigate other water-soluble nanoparticles with complex functional groups and their interaction with metal ions, providing a powerful analysis tool for analytical and environmental areas.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail for Y.L.: [email protected]. *E-mail for L.R.: [email protected]. *E-mail for J.C.: [email protected]. ORCID

Yuexiang Lu: 0000-0003-2755-7733 Chao Xu: 0000-0001-5539-4754 Zhicheng Zhang: 0000-0002-2192-3846 Xiangke Wang: 0000-0002-3352-1617 Linfeng Rao: 0000-0002-1873-0066 Author Contributions

The manuscript was written through contributions of all authors. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the National Natural Science Foundation of China (Grant Nos. 51425403, 21775087) and the Fundamental Research Funds for the Central Universities (2019MS046). Z.W. also thanks the China Scholarship Council for financial support to study at Lawrence Berkeley National Laboratory in the US.

■

REFERENCES

(1) Xu, X.; Ray, R.; Gu, Y.; Ploehn, H. J.; Gearheart, L.; Raker, K.; Scrivens, W. A. J. Am. Chem. Soc. 2004, 126, 12736−12737. (2) Tan, X.; Li, Y.; Li, X.; Zhou, S.; Fan, L.; Yang, S. Chem. Commun. 2015, 51, 2544−2546. (3) Jiang, J.; Ye, G.; Wang, Z.; Lu, Y.; Chen, J.; Matyjaszewski, K. Angew. Chem., Int. Ed. 2018, 57, 12037−12042. (4) Zhu, S.; Meng, Q.; Wang, L.; Zhang, J.; Song, Y.; Jin, H.; Zhang, K.; Sun, H.; Wang, H.; Yang, B. Angew. Chem. 2013, 125, 4045−4049. (5) Wang, Z.; Xu, C.; Lu, Y.; Chen, X.; Yuan, H.; Wei, G.; Ye, G.; Chen, J. Sens. Actuators, B 2017, 241, 1324−1330. (6) Wang, Z.; Xu, C.; Lu, Y.; Wu, F.; Ye, G.; Wei, G.; Sun, T.; Chen, J. ACS Appl. Mater. Interfaces 2017, 9, 7392−7398.

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.9b01225. Experimental section, detailed characterization information on the two carbon dots, and supporting figures (PDF) G

DOI: 10.1021/acs.analchem.9b01225 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry (7) Lim, S. Y.; Shen, W.; Gao, Z. Chem. Soc. Rev. 2015, 44, 362−381. (8) Vallan, L.; Canton-Vitoria, R.; Gobeze, H. B.; Jang, Y.; Arenal, R.; Benito, A. M.; Maser, W. K.; D Souza, F.; Tagmatarchis, N. J. Am. Chem. Soc. 2018, 140, 13488−13496. (9) Ma, Y.; Xu, G.; Wei, F.; Cen, Y.; Ma, Y.; Song, Y.; Xu, X.; Shi, M.; Muhammad, S.; Hu, Q. J. Mater. Chem. C 2017, 5, 8566−8571. (10) Tan, C.; Su, X.; Zhou, C.; Wang, B.; Zhan, Q.; He, S. RSC Adv. 2017, 7, 40952−40956. (11) Wang, Z.; Lu, Y.; Yuan, H.; Ren, Z.; Xu, C.; Chen, J. Nanoscale 2015, 7, 20743−20748. (12) Gogoi, N.; Barooah, M.; Majumdar, G.; Chowdhury, D. ACS Appl. Mater. Interfaces 2015, 7, 3058−3067. (13) Wang, J.; Zhang, W.; Yue, X.; Yang, Q.; Liu, F.; Wang, Y.; Zhang, D.; Li, Z.; Wang, J. J. Mater. Chem. A 2016, 4, 3893−3900. (14) Wang, Z.; Xu, C.; Lu, Y.; Wei, G.; Ye, G.; Sun, T.; Chen, J. Chem. Eng. J. 2018, 344, 480−486. (15) Fedin, I.; Talapin, D. V. J. Am. Chem. Soc. 2014, 136, 11228− 11231. (16) Vallan, L.; Urriolabeitia, E. P.; Ruipérez, F.; Matxain, J. M.; Canton-Vitoria, R.; Tagmatarchis, N.; Benito, A. M.; Maser, W. K. J. Am. Chem. Soc. 2018, 140, 12862−12869. (17) Baker, S. N.; Baker, G. A. Angew. Chem., Int. Ed. 2010, 49, 6726−6744. (18) Li, H.; Sun, C.; Ali, M.; Zhou, F.; Zhang, X.; MacFarlane, D. R. Angew. Chem. 2015, 127, 8540−8544. (19) Boles, M. A.; Ling, D.; Hyeon, T.; Talapin, D. V. Nat. Mater. 2016, 15, 141. (20) Gorgulho, H. F.; Mesquita, J. P.; Gonçalves, F.; Pereira, M. F. R.; Figueiredo, J. L. Carbon 2008, 46, 1544−1555. (21) Zhu, X.; Zhang, Z.; Xue, Z.; Huang, C.; Shan, Y.; Liu, C.; Qin, X.; Yang, W.; Chen, X.; Wang, T. Anal. Chem. 2017, 89, 12054− 12058. (22) Black, S.; Ferrell III, J. R. Energy Fuels 2016, 30, 1071−1077. (23) Yang, C.; Powell, B. A.; Zhang, S.; Rao, L. Radiochim. Acta 2015, 103, 707−717. (24) Duster, T. A.; Szymanowski, J. E.; Na, C.; Showalter, A. R.; Bunker, B. A.; Fein, J. B. Colloids Surf., A 2015, 466, 28−39. (25) Strelko, V.; Malik, D. J.; Streat, M. Carbon 2002, 40, 95−104. (26) Gorgulho, H.; Mesquita, J.; Goncalves, F.; Manuel Pereira, F.; Figueiredo, J. Carbon 2008, 46, 1544−1555. (27) Sun, X.; Tian, G.; Xu, C.; Rao, L.; Vukovic, S.; Kang, S. O.; Hay, B. P. Dalton T. 2014, 43, 551−557. (28) Arrigo, R.; Hävecker, M.; Wrabetz, S.; Blume, R.; Lerch, M.; McGregor, J.; Parrott, E. P.; Zeitler, J. A.; Gladden, L. F.; KnopGericke, A. J. Am. Chem. Soc. 2010, 132, 9616−9630. (29) Brozek, C. K.; Hartstein, K. H.; Gamelin, D. R. J. Am. Chem. Soc. 2016, 138, 10605−10610. (30) Pan, Z.; Li, W.; Fortner, J. D.; Giammar, D. E. Environ. Sci. Technol. 2017, 51, 9219−9226. (31) Xie, Y.; Helvenston, E. M.; Shuller-Nickles, L. C.; Powell, B. A. Environ. Sci. Technol. 2016, 50, 1821−1827. (32) Alves, L. A.; de Castro, A. H.; de Mendonça, F. G.; de Mesquita, J. P. Appl. Surf. Sci. 2016, 370, 486−495. (33) Gans, P.; Sabatini, A.; Vacca, A. Talanta 1996, 43, 1739−1753. (34) Alderighi, L.; Gans, P.; Ienco, A.; Peters, D.; Sabatini, A.; Vacca, A. Coord. Chem. Rev. 1999, 184, 311−318. (35) Zhao, G.; Li, J.; Ren, X.; Chen, C.; Wang, X. Environ. Sci. Technol. 2011, 45, 10454−10462. (36) Sun, Y.; Wang, Q.; Chen, C.; Tan, X.; Wang, X. Environ. Sci. Technol. 2012, 46, 6020−6027. (37) Estes, S. L.; Arai, Y.; Becker, U.; Fernando, S.; Yuan, K.; Ewing, R. C.; Zhang, J.; Shibata, T.; Powell, B. A. Geochim. Cosmochim. Acta 2013, 122, 430−447. (38) Ferdousi, B. N.; Islam, M. M.; Okajima, T.; Ohsaka, T. Int. J. Electrochem. Sci. 2016, 11, 6215−6228. (39) Sastre, S.; Casasnovas, R.; Munoz, F.; Frau, J. Phys. Chem. Chem. Phys. 2016, 18, 11202−11212. (40) Dodge, C. J.; Francis, A. J. Environ. Sci. Technol. 1997, 31, 3062−3067.

(41) Dodge, C. J.; Francis, A. J. Environ. Sci. Technol. 2002, 36, 2094−2100.

H

DOI: 10.1021/acs.analchem.9b01225 Anal. Chem. XXXX, XXX, XXX−XXX