Highly Selective Fluorescence Determination of ... - ACS Publications

Subscriber access provided by GAZI UNIV

Article

Highly Selective Fluorescence Determination of the Hematin Level in Human Erythrocytes with No Need for Separation from Bulk Hemoglobin Lijuan Ji, Li Chen, Ping Wu, Dominic Francis Gervasio, and Chenxin Cai Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b00131 • Publication Date (Web): 04 Mar 2016 Downloaded from http://pubs.acs.org on March 5, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Highly Selective Fluorescence Determination of the Hematin Level in Human Erythrocytes with No Need for Separation from Bulk Hemoglobin

Lijuan Ji,a Li Chen,a Ping Wu,*a Dominic F. Gervasiob and Chenxin Cai*a a

Jiangsu Key Laboratory of New Power Batteries, Jiangsu Collaborative Innovation Center of

Biomedical Functional Materials, National and Local Joint Engineering Research Center of Biomedical Functional Materials, College of Chemistry and Materials Science, Nanjing Normal University, Nanjing 210097, P.R. China. b

Department of Chemical & Environmental Engineering, University of Arizona, 1133 East James E.

Rogers Way, Tucson, Arizona 85721, United States.

* Corresponding author, E-mail: [email protected] (P. Wu); [email protected] (C. Cai).

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 28

ABSTRACT: Hematin-induced fluorescence quenching of the boron-doped graphene quantum dots (BGQDs) allows determination of hematin concentration in human erythrocytes with no need for separating hematin from hemoglobin before doing the assay. The BGQDs are made by oxidizing a graphite anode by holding the voltage between a graphite rod and a Pt cathode at 3 V for 2 hours in an aqueous borax solution at pH=7; then filtering the borate solution with BGQDs and dialyzing the borate from the filtrate leaving a solution of BGQDs in water. The fluorescence intensity of BGQDs is measurable in real time, and its quenching is very sensitive to the concentration of hematin in the system, but not to other coexisting biological substances. The analytical signal is defined as ∆F = 1–F/F0, where F0 and F are the fluorescence intensities of the BGQDs before and after interaction with hematin, respectively. There is a good linear relationship between ∆F and hematin concentration, ranging from 0.01 to 0.92 µM with the limit of detection (LOD) being ~0.005 ± 0.001 µM at an S/N of 3. This new method is sensitive, label free, simple, and inexpensive and many tedious procedures related to sample separation and preparation can be omitted, implying this method has potential for applications in clinical examinations and disease diagnoses. For example, the determination of the hematin level in two kind of red blood cell samples, healthy human erythrocytes and sickle cell erythrocytes, gives the average concentrations of hematin to be ~(23.1 ± 4.9) µM (average of five samples) for healthy red cell cytosols and ~(52.5 ± 9.5) µM (average of two samples) for sickle red cell cytosols.

Keywords: Boron-doped graphene quantum dots; Hematin; Erythrocyte; Hemoglobin; Fluorescence quenching.


2

Page 3 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Hematin has a hydroxyl group (–OH) axially bound to an iron (III) ion, which is also bound to 4 nitrogen donors in the plane of the protoporphyrin IX macrocycle (the structure is depicted in Figure S1). Hematin is an oxidized form of heme, Fe (II)-protoporphyrin IX, which is the prosthetic group of proteins such as hemoglobin and myoglobin.1,2 When hemoglobin is oxidized to met-hemoglobin, the iron is oxidized from its Fe(II) to Fe(III) form, which binds a hydroxyl group, and the heme group is released from hemoglobin to the solution (in red cell cytosol) as hematin.3–4 Hematin is a potent hemolytic agent in vitro,5 and a perturbant of red cell membrane integrity in vivo,6 causing the membrane damage7 because it catalyzes the generation of reactive oxygen species (ROS) such as hydrogen peroxide, superoxide radicals, and the hydroxyl radicals etc., which mediate lipid peroxidation.8–10 In addition to its oxidative effects on membrane lipids, hematin, even at low concentrations of micromolar level, can also exert profound detergent-like effects on the red cell membrane, leading to the conformational changes in skeletal proteins such as spectrin and protein 4.1R.11 This kind of injury will result a series of functional consequences, such as disruption of proteinprotein linkages and disintegration of membrane skeletal structure.11 Furthermore, in aqueous solutions at a physiological pH, hematin has a tendency to aggregate itself and adhere strongly to different kinds of surfaces, such as red cell surface, leading to the higher adhesion of sickle red cells to leucocytes and the endothelium.12–15 The deleterious influence which hematin exerts on the red cells calls for a facile method for the rapid and precise determination of hematin levels (concentrations) in human erythrocytes. However, we are aware of few reports of ways to assay the concentration of hematin in erythrocytes. Liu et al.16 reported a method for determination of hematin in sickle and healthy human erythrocytes based on the adsorption characteristics of hematin. Prior to this assay, it was necessary to separate hematin from the bulk hemoglobin and its derivatives by ion-exchange liquid chromatography under high-ionic strength conditions (2 M NaCl) exploiting the different charges carried by hematin and hemoglobin. Aich et al.17,18 reported an approach based on enzymatic catalysis and chemiluminescence for assaying hematin in healthy human erythrocytes. They first lysed erythrocytes and isolated hematin ACS Paragon Plus Environment

3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 28

from hemoglobin by dialysis based on the different sizes of the hematin and hemoglobin, then used the hematin to reconstitute horseradish peroxidase (HRP) from an excess of the apoenzyme (apo-HRP) and finally determined the HRP reaction rate from the evolution of the emitted luminescence in the presence of H2O2 and luminol. The luminescence intensity is proportional to the rate of catalytic decomposition of H2O2. This rate, in turn, is proportional to the concentration of reconstituted HRP, whose concentration equals that of the initial hematin under the conditions of excess of apo-HRP. Hence, the luminescence intensity is measured to proportional to the concentration of the hematin. Although the above two methods for assaying hematin are well-established, they are time-consuming and require tedious procedures to separate the hematin from hemoglobin before performing the assay. Other attempts to directly measure the hematin level in red cell cytosol have been hampered by the lack of a reliable method to separate hematin from the bulk hemoglobin and other denatured hemoglobin species. Most of works on heme quantification has focused on the determination of the total heme content, including the free heme and the heme quantities bound to globins.1,19–24 Note that the free heme is found in the form of hematin, because the dissociation of heme from ferrous hemoglobin is extremely slow, and heme mostly dissociates from methemoglobin in the form of hematin. The methods employed in the studies of total heme content mostly rely on spectroscopic responses of heme and heme-containing proteins such as hemoglobin, myoglobin, and cytochrome c etc.;20,21,25 HPLC,26 electrochemical methods,22–24 capillary electrophoresis,27 and mass spectrometry19,28 etc. have also been used for in recent reports. However, these methods do not offer an approach for directly assay of hematin (free heme) in erythrocytes. Thus, there is still a great need and desire to find a facile and highly sensitive method for the direct assay of hematin in human erythrocytes with no needing for separating hematin from hemoglobin prior to measuring the hematin levels. Up to now, assaying hematin in erythrocytes with no need for separating hematin from hemoglobin has remained a great challenge. The incentive for this work was the deleterious effect of hematin and the lack of easy ways to measure hematin concentration in erythrocytes. Here, we report a new method to meet this challenge, to ACS Paragon Plus Environment

4

Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


determine the hematin level in human erythrocytes by a facile, sensitive measurement, and with no need for separating hematin from hemoglobin in the erythrocyte lysates prior to the measurement. Our approach is based on the fluorescence quenching of boron-doped graphene quantum dots (BGQDs) by hematin. BGQDs were synthesized by oxidizing a graphite rod in aqueous borax solution with a Pt cathode at a constant voltage of 3 V. The BGQDs possess both sp2-bonded carbon atoms and abundant surface oxygen-containing functional groups (hydroxyl, carbonyl, and carboxylic acid groups etc.).29 These groups facilitate binding of moieties (such as hematin) to the surface of BGQDs through π–π stacking, electrostatic interaction, and chemical reactions, etc. Moreover, BGQDs have unique properties, such as, chemical inertness, low cytotoxicity, biocompatibility, and rich fluorescence, and so can be expected to be compatible with biological measurements.29 BGQDs emit a strong fluorescence under excitation of 300–440 nm in an excitation wavelength-dependent manner. We found that this fluorescence can be efficiently quenched by the presence of hematin and that hemoglobin and myoglobin do not quench the fluorescence of BGQDs. The changes in the fluorescence intensities of the BGQDs could be monitored in real time and directly related to the amount of hematin, which establishes the basis for measuring hematin levels in human erythrocytes. That hemoglobin does not quench the fluorescence of BGQDs shows this assay is selective to hematin and there is no need to separate hematin from hemoglobin prior to assay. In addition, this assay is label free, fairly simple to do, and inexpensive. All of these characteristics of this method for assaying hematin suggest that this new method offers a new practical tool for clinical examinations and diagnoses of diseases. EXPERIMENTAL SECTION Synthesis of Boron-Doped Graphene Quantum Dots (BGQDs). BGQDs were synthesized by a constant potential electrolysis method, which was performed on a CHI 660B electrochemical workstation (CH Instruments) in a two-compartment two-electrode cell with the sample volume of ∼20 mL. The high purity graphite rod (99.9%, 3 mm in diameter, Shanghai Carbon Co. Ltd.) was used as an anode and was put into aqueous 0.1 M borax (99.5%, Sigma-Aldrich) solution (the solution pH was ACS Paragon Plus Environment

5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 28

adjusted to ~7), the length of the graphite rod below the solution level was about 10 cm. A Pt sheet (1.5 cm × 1.5 cm) was used as a cathode. The voltage between the anode and cathode was controlled at 3 V. Assuming the Pt cathode was pinned by the proton reduction reaction at pH 7, then its potential was around –0.4 V versus the normal hydrogen electrode (NHE) and the graphite anode would have to be at around +2.6 V versus NHE to maintain the cell voltage at 3 V. This voltage (+2.6 V) is high enough to drive the electrolyte ions into the graphene layers and to oxide the C–C bonds to form the B–C bond and BGQDs.29,30 When the voltage was applied, a current of ~0.13 mA flowed between the anode and cathode. As the electrolysis time increased, the color of the solution in the anode area changed from colorless to pale yellow, and finally to brown. At the same time a lot of gas was seen to evolve from the cathode, implying the reduction of protons to hydrogen and the oxidation and dissolution of the graphite anode leading to the formation of the BGQDs. After 2-h electrolysis, BGQDs were collected by filtering the resulting solution using 0.22-µm microporous nylon membrane to remove the precipitated graphite oxide and graphite particles. Then the obtained pale-yellow solution was dialyzed over deionized water in a dialysis bag (retained molecular weight 3500 Da) for 48 h to remove the electrolyte of borax, the deionized water was changed every 12 h. The concentration of the aqueous BGQD solution was found by taking a known volume of the aqueous BGQD solution and evaporating it to dryness and weighing the dry mass of the BGQDs, which gives a concentration of 15 µg per milliliter. Thus, about 320 µg BQGDs can be obtained by passing one coulomb charge. The quantum yield (QY) of the synthesized aqueous BGQD solution was estimated to be about 5% by comparing the integrated photoluminescence intensities (excited at 380 nm) and the absorbance values (at 380 nm) using rhodamine 6G in ethanol solution (QY = 95%) as a reference. Characterizations. Transmission electron microscopic (TEM) images of the synthesized BGQDs were recorded on a JEOL JEM-2100F transmission electron microscope operating at an accelerating voltage of 200 kV. The TEM samples were prepared by drying a droplet of the BGQDs solution on a Cu grid. The surface characteristics of the synthesized BGQDs were examined by X-ray photoelectron spectroscopy (XPS), which was measured with ESCALAB 250 XPS spectrometer (VG Scientifics) ACS Paragon Plus Environment

6

Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


using a monochromatic Al Kα line at 1486.6 eV. Binding energies were calibrated with respect to the C1s peak at 284.6 eV. Peak fit analysis was performed using the XPS PEAK program (version 4.0). Xray diffraction (XRD) patterns of the BGQDs were recorded on a Rigaku/Max-3A X-ray diffractometer with Cu Kα radiation (λ = 0.15418 nm). UV-vis spectra were measured on a Cary 5000 UV-vis-NIR spectrometer (Varian). The FTIR transmission spectra were recorded at a resolution of 4 cm−1 using a Nexus 670 FT-IR spectrophotometer (Nicolet Instruments) for sample on a KBr disk. Raman spectra were recorded on a Labram HR 800 microspectrometer (Jobin Yvon, France) with an excitation source of 514 nm. Typically, Raman spectra were acquired from an area of 1 mm × 1 mm on the substrate. The collection time of each Raman spectrum was 10 s over a spectral range from 500 to 2000 cm–1. Fluorescence and photoluminescence (PL, under an excitation wavelength of 360 nm) spectra were collected using a FluoroSENS fluorescence spectrophotometer (Gilden Photonics) equipped with a Xenon lamp excitation source. The distribution and the mean sizes of the BGQDs were analyzed by a BI-200SM dynamic light scattering instrument (DLS; Brookhaven Instruments). Electrochemical Measurements. The redox characteristics of the synthesized BGQDs were studied by cyclic voltammetry, which were performed on an Autolab PGSTAT302N electrochemical station (Metrohm) using a glassy carbon (GC, 3 mm in diameter, CH Instruments) electrode. A Pt sheet (1.5 cm × 1.5 cm) and a saturated calomel electrode (SCE) were used as counter electrode and reference electrode, respectively. All electrochemical measurements were performed at room temperature (22 ± 1 °C). Prior to use, GC electrode was sequentially polished with metallographic abrasive paper (no. 6) and slurries of 0.3 and 0.05 µm alumina to create a mirror finish. It was then sonicated for about 1 min in absolute ethanol and then double-distilled water to remove traces of alumina grit from its surface. After the electrode was rinsed thoroughly with double distilled water and dried at ambient temperature, 15 µL of BGQDs aqueous suspension (200 µg mL–1) was cast onto the surface of the electrode with a microsyringe (the mass of BGQDS loaded on electrode surface is about 3 µg), and solvent (water) was allowed to be evaporated in air at ambient temperature. The redox features of the BGQDs were studied in freshly dried dimethylformamide (DMF) solution containing 0.1 M Bu4NBF4 (tetrabutylammonium ACS Paragon Plus Environment

7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 28

tetrafluoroborate, from Sahn Chemical Technology) as electrolyte. The BGQDs on the surface of GC electrode in DMF solution is fairly stable, since excepting a slight decrease in current over the initial five cycles (decrease by ~3%), the electrochemical response was virtually unchanged even 200 cycles of continuous scanning. Hematin-Induced Fluorescence Quenching of BGQDs. To evaluate the fluorescence quenching ability of hematin to BGQDs, the fluorescence of the BGQDs in the presence of various concentrations of hematin (Sigma-Aldrich) was recorded under excitation of 380 nm. In a typical experiment, 15 µg of BGQDs were dispersed in 1 mL phosphate buffer (PBS, 0.1 M, pH 7.4), and the fluorescence spectrum was recorded. Then, various concentrations of hematin (0.01–2.0 µM) were added into the system and allowed to interact with BGQDs. After a stable fluorescence of the solution was attained (usually within 5 min), the fluorescence signal for BGQDs with a particular concentration of hematin was recorded. Hematin solution was prepared by dissolving hematin powder in dimethyl sulfoxide (DMSO) at a concentration of ~1 mg mL–1, then diluted it by a mixture of 66.5% ethanol, 17% acetic acid, and 16.5% water (v/v). The final concentration of hematin was spectrophotometrically determined using the solvent-specific extinction coefficient 144 mM–1cm–1 at 398 nm.31 For evaluating the selectivity of the fluorescence quenching of the hematin to BGQDs, several biological substances such as hemoglobin (Hb), myoglobin (Mb), ascorbic acid (AA), L-cysteine (LCys), L-glycine (L-Gly), glutathione (GSH), nicotinamide adenine dinucleotide (reduced form, NADH), nicotinamide adenine dinucleotide phosphate (NADPH), serum (provided by Xianlin Hospital, Nanjing), K+, and Na+ ions were randomly selected as potential species that could quench the fluorescence of BGQDs and interfere with the quantitative observation of fluorescence quenching of BGQDs by hematin. Preparation of Red Cell Hemolysate. To determine the hematin level in red cell cytosol, we prepared the red cell hemolysate using the procedures reported by Aich et al.17 Blood samples from healthy adults and patients with sickle cell disease were provided by the Xianlin Hospital (Nanjing, China) following a protocol approved by the Xianlin Hospital Committee for Protection of Human Subjects. The time ACS Paragon Plus Environment

8

Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


between blood collection and the start of its analysis was approximately 1 h. To prepare the hemolysate, we added 45 mL of isotonic 0.9% NaCl solution into 4 mL of blood sample, then gently agitated the tube to homogenize the suspension and centrifuged the diluted blood for 20 minutes. The supernatant was decanted. After washed three times by the above procedures, the red blood cells were suspended into 50 mL deionized water. The red cells ruptured under the influence of the osmotic pressure difference between the cells and the water. We then centrifuged this suspension to compress the cell membranes at the tube bottom, collected the hemolysate, and stored it at 5 oC in a sterile tube for further analysis. To evaluate the dilution of the hemolysate compared to the red cell cytosol, we converted hemoglobin to its cyan-met form using Drabkin’s reagent (Sigma, Product code: D5941). The concentration of hemoglobin in the hemolysate was spectrophotometrically determined using an extinction coefficient 1.512 mL mg–1cm–1 at 540 nm for cyanomethemoglobin.32 We assumed that the hemoglobin concentration in the red cell cytosol of healthy red cells is 330 mg mL–1 (corresponding to mean corpuscular hemoglobin concentration (MCHC) of 33 g dL–1) and calculated the dilution ratio as the ratio of two concentrations. Typical hemoglobin concentrations in hemolysate were ~1.8 mg mL–1 (~28 µM), corresponding to dilution factor of 183. We then used this dilution factor to covert the measured concentration of hematin in hemolysate into that in red cell cytosol. Calculation Details of the Interaction of Hematin with BGQDs. The interaction of hematin with BGQDs was theoretically calculated based on DFT using Gaussian 03 (revision B.03; Gaussian Inc., Pittsburgh, PA, 2003).33 The geometric optimizations were performed using a DFT method with Becke’s hybrid three-parameter nonlocal exchange functional combined with the Lee-Yang-Parr gradient-corrected correlation functional (B3LYP). The 6-31G (d, p) basis set was used for all elements. For these calculations, we constructed a model containing 28 hexagonal rings with delocalized π electrons (C77BH24) for BGQDs. The carbon atoms at the edge of the graphene were terminated with hydrogen atoms. For simulating hematin adsorption, we placed hematin near the BGQDs plane at a distance of 3 Å and the optimized the geometric structures of the adsorbed hematin at the surface of ACS Paragon Plus Environment

9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 28

BGQDs. The adsorption energy (Ead), defined as the zero-point energy (ZPE) difference between the total system including the adsorbed molecules and the isolated system, was derived from frequency analysis after the initial structure optimization.34 The energy of the isolated system is the sum of the energies of BGQDs and the isolated adsorbed molecule. Thus, a negative value of Ead indicates that the adsorption process is energetically favorable. The effects of the interaction of hematin with BGQDs on the energy gap between LUMO (the lowest unoccupied molecular orbital) and HOMO (the highest occupied molecular orbital) energy levels of BGQDs was also evaluated based on the calculation. RESULTS AND DISCUSSION We synthesized the BGQDs to assay of the hematin level in red cells based on the fluorescence quenching of BGQDs by hematin. The synthesized BGQDs solution remains homogeneous with a clear pale yellow color for several months (at least for four months) at ambient temperature without any perceptible aggregation and color change (Figure S2), which could be further verified by the almost unchanged PL spectra after four months (Figure S3). This feature is crucial for practical sensing applications, because the synthesized BGQDs must be water-soluble and stable toward ambient environment when they are used for sensing applications. The high dispersion and physical stability of the solution of BGQDs is probably due to the oxygen-containing groups present at their surface and edges of the BGQDs, as evidenced from FTIR spectrum (Figure S4A), which shows the stretching mode of O−H (∼3442 cm−1), stretching vibration of C=O (∼1578 cm−1), sp2 hybridized C=C (∼1431 cm−1), C−OH groups (∼1338 cm−1), and the deformation mode of C−O−C in epoxide moieties (∼1129 cm−1).35–38 Note that the peak at ~1405 cm–1 is the asymmetric stretching vibration mode of B–O39 (Figure S4B), confirming the incorporation of boron during the synthesis of the BGQDs. TEM image shows that the synthesized BGQDs are virtually monodispersed with a narrow distribution of diameters between 3 to 5 nm in (Figure 1A), which is in good agreement with sizing by dynamic light scattering (DLS analysis, Figure S5). HRTEM image (the inset in Figure 1A) indicates the high crystallinity of the synthesized BGQDs with a lattice spacing of 0.241 nm, which corresponds ACS Paragon Plus Environment

10

Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


to the lattice constant in (1220) plane of graphitic (sp2) carbon. However, this spacing is slightly smaller than that of pristine graphite (0.242 nm),40,41 implying the boron atoms have been doped into the conjugated carbon backbone leading to somewhat disordered structures. The further evidence of the Bdoping in the graphene lattice is provided by a Raman spectrum, which exhibits two peaks at 1347 and 1601 cm–1 corresponding to the well-defined D and G bands (Figure S6), respectively. The ratio of ID/IG in the Raman spectra of carbon materials is usually used to evaluate the disorder in the graphene structures. The G band is related to the E2g vibration mode of sp2 carbon domains and is associated with the degree of graphitization, while the D band is associated with structural defects and partially disordered structures of the sp2 domains. The value of ID/IG ratio is estimated to be ~0.94, which is higher than that reported for pristine graphene (0.79), suggesting a breakage of the hexagonal symmetry of the graphene,42 most likely due to the B-doping. These results clearly indicated that BGQDs have been synthesized by the electrolysis of graphite method.

Figure 1. (A) Typical TEM and HRTEM images of the synthesized BGQDs. (B) XPS spectrum of BGQDs. The inset shows the amplified XPS of B1s. (C and D) High-resolution XPS spectra of the C1s and B1s in BGQDs, respectively, and their related curve-fitted components.


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

XRD pattern of BGQDs shows a diffraction peak at ~24.7 ° (Figure S7), corresponding to the (002) plane of graphite. In contrast to the sharp peak of the pristine graphite,29 this diffraction peak is broader, indicating that the oxidation process during the BGQDs synthesis has introduced more active sites on the surface of BGQDs, which can be confirmed by XPS (see discussion presented below) and FTIR spectra (Figure S4A). Moreover, XRD peak of BGQDs shifts to a lower angle compared with that of pristine graphite (~26 °), implying that BGQDs have a larger interlayer spacing (calculated to be ~0.360 nm) than pristine graphene does (0.342 nm42). The increased interlayer spacing could be attributed to the weak π–π stacking of the graphene with more oxygen-containing functional groups at the edges of the layer of BGQDs. It is also noticed that the synthesized BGQDs do not show any diffractions at ~10 ° (2θ), a characteristic feature of graphene oxide (GO), suggesting that XRD characteristics of BGQDs are different from GO, though both of them have rich oxygen-containing groups. The detailed surface composition and bonding information of the doped B atom in the BGQDs can be studied by XPS spectra, which show the presence of B, C, and O elements with atomic percentages of ~3.2, 60.1, and 36.7%, and corresponding B1s, C1s, and O1s peaks at ~191, 284, and 531 eV (Figure 1B), respectively, further conforming the incorporation of the B atoms into the graphene by electrolysis of the graphite in borax solution. The B1s peak appears at a higher binding energy (191 eV) compared with that reported for pure boron (187 eV42), suggesting the B atoms coming from the borax have bonded to C atoms in the sp2 carbon network. The high resolution XPS peak of C1s indicates the presence of the C–B (283.9 eV43), C=C (284.6 eV), C–O (285.3 eV), C=O (286.5 eV), O–C=O (288.5 eV) bonds44–49 (Figure 1C). These results suggest that there are large amount of oxygen-containing groups presented at the surface of the synthesized BGQDs, agreeing with that obtained from FTIR spectrum. In the high resolution XPS spectrum, the B1s peak is deconvoluted into three peaks at 189.8, 190.6, and 191.3 eV, corresponding the B bonding form in C–B (189.8 eV), BC2O (190.6 eV), and BCO2 (191.3 eV)50,51 (Figure 1D), respectively. To estimate the relative amount of the C–B, BC2O, and BCO2 components in the BGQDs, we performed XPS measurements for five different BGQDs samples, which were synthesized using the same procedures. The average amount of the C–B, BC2O, and BCO2 ACS Paragon Plus Environment

12

Page 13 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


components is ~(24 ± 3.6), (55 ± 6.1), and (21 ± 4.7)%, respectively. These results further indicate the doping of the B atom into graphene and formation of the BGQDs.

Figure 2. (A) UV-vis spectrum of the BGQDs. The inset shows the photograph of the BGQDs under daylight and excitation of 365 nm (16 W). (B) Dependence of the photoluminescence (PL) spectra of BGQDs on excitation wavelength. (C) Photoluminescence excitation (PLE) spectrum of BGQDs excited by 530 nm. The concentration of BGQDs is ~15 µg mL–1.

After the synthesis of BGQDs was confirmed, their adsorption and excitation-dependent photoluminescence (PL) characteristics were studied. The UV-vis spectrum and optical images of the BGQDs are shown in Figure 2A. A typical adsorption peak at ∼230 nm was observed and is assigned to π−π* transition of aromatic sp2 domains.52,53 Two weak adsorption peaks at ∼302 and 370 nm was also observed and are assigned as an n−π* transition of B–C, which are indicative of B doping in graphene. It has been reported that isolated sp2-hybridized clusters with a size of ~3 nm within the carbon-oxygen matrix could generate band gaps consistent with blue emission due to the localization of the electronhole pairs.45,52,53 To check if our synthesized BGQDs emit a blue PL emission, the BGQDs solution was irradiated under 365 nm (16 W). It is interesting to observe that BGQDs emits a blue PL (Figure 2A, inset), while its solution shows no noticeable PL without UV irradiation, and is pale-yellow, transparent, and clear under daylight (Figure 2A, inset), suggesting that both the size and surface features make an important contribution to the observed blue PL emission from BGQDs. An excitation wavelengthdependent PL emission was observed with BGQDs as depicted in Figure 2B, which shows two prominent PL emission regions at ~465 and 530 nm, respectively. When excited with 300–340 nm ACS Paragon Plus Environment

13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 28

irradiation, the first PL peak (at ∼465 nm) dominates to the second peak (at ∼530 nm), and its intensity (the peak at 465 nm) increases with the wavelength of excitation. When the excitation wavelength increases to 360 nm, the second PL peak (at 530 nm) dominates to the first peak (at 465 nm), and its intensity decreases rapidly with the increase of the excitation wavelength to 440 nm, which is characteristic of PL for most luminescent carbon nanoparticles. As in the case of most of the luminescent carbon dots and graphene quantum dots, this excitation-dependent ﬂuorescence behaviour of the BGQDs reflects the effects of particles of different sizes and a distribution of different surface states.41,45,54,55 The PL intensities of the BGQDs as a function of extreme pH or high ionic strength in solution were investigated to evaluate stability of BGQDs under extreme conditions. As presented in Figure S8A, the BGQDs displayed stable PL even at extremes of pH conditions; they exhibited strong PL activities in the range of pH 3 to 11. The effects of the ionic strength on the PL activities were evaluated by recording the PL signal in solution containing different concentrations of NaCl (0.01 to 1 M). The PL intensities remained constant with the increase of ionic strength (Figure S8B), probably because there is almost no change in the ionization of the functional groups on the surface of BGQDs. These results suggest that synthesized BGQDs are highly stable even under extreme conditions. The photoluminescence excitation (PLE) spectrum of the BGQDs was further studied. The PLE spectrum shows two peaks at ∼360 and 270 nm under excitation of 530 nm, which is the strongest luminescence of the BGQDs (Figure 2C). The PLE peak at ∼270 nm (4.60 eV) is caused by electron transition from a σ orbital to LUMO, while the PLE peak located at ∼360 nm (3.45 eV) is attributed to an electron transition from a π orbital (HOMO) to LUMO. The energy difference between the σ and π orbitals is 1.15 eV, which is within the value expected for triplet carbenes (

Highly Selective Fluorescence Determination of ... - ACS Publications

Recommend Documents