Tailoring chemically converted graphenes using a water-soluble pyrene derivative with a zwitterionic arm for sensitive electrochemiluminescence-based analyses
Tailoring chemically converted graphenes using a water-soluble pyrene derivative with a zwitterionic arm for sensitive electrochemiluminescence-based analyses
Author’s Accepted Manuscript Tailoring Chemically Converted Graphenes Using a Water-soluble Pyrene Derivative with a Zwitterionic Arm for Sensitive El...
Author’s Accepted Manuscript Tailoring Chemically Converted Graphenes Using a Water-soluble Pyrene Derivative with a Zwitterionic Arm for Sensitive Electrochemiluminescence-based Analyses Jihye Kwon, Seo Kyoung Park, Yongwoon Lee, Je Seung Lee, Joohoon Kim www.elsevier.com/locate/bios
To appear in: Biosensors and Bioelectronic Received date: 28 May 2016 Revised date: 3 August 2016 Accepted date: 5 August 2016 Cite this article as: Jihye Kwon, Seo Kyoung Park, Yongwoon Lee, Je Seung Lee and Joohoon Kim, Tailoring Chemically Converted Graphenes Using a Water-soluble Pyrene Derivative with a Zwitterionic Arm for Sensitive Electrochemiluminescence-based Analyses, Biosensors and Bioelectronic, http://dx.doi.org/10.1016/j.bios.2016.08.013 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Tailoring Chemically Converted Graphenes Using a Water-soluble Pyrene Derivative with a Zwitterionic Arm for Sensitive Electrochemiluminescence-based Analyses
Jihye Kwona1, Seo Kyoung Parka1, Yongwoon Leea, Je Seung Leea*, Joohoon Kima,b*
a
Department of Chemistry
b
KHU-KIST Department of Converging Science and Technology, Kyung Hee
Abstract We report a method to tailor chemically converted graphenes (CCGs) using a watersoluble pyrene derivative (1) with a zwitterionic arm, and the feasibility of the tailored CCGs to sensitive electrochemiluminescence (ECL)-based analyses. The compound 1 serves the dual purpose of improving the dispersion of the CCGs in aqueous solutions and further tailoring the catalytic activity of the CCGs with dendrimer-encapsulated catalytic nanoparticles. As a model system, we conjugated dendrimer-encapsulated Pt nanoparticles to the 1-functionalized CCGs on indium tin oxide (ITO) electrodes. The resulting ITOs exhibited significantly increased ECL emission of the luminol/H2O2 ECL system; i.e. two orders-of-magnitude enhancement 1
These authors contributed equally to this work.
in the ECL compared to that obtained from bare ITOs, which allowed a ca. 154 times more sensitive ECL-based analysis of cholesterol using the modified ITOs compared with the use of bare ITOs.
Keywords: Chemically converted graphene, Pyrene, Electrochemiluminescence, Indium tin oxide
1. Introduction Among several effective techniques developed for preparing graphene sheets, the chemical reduction of exfoliated graphene oxides (GOs) allows large-scale synthesis of graphenes at relatively low cost (Allen et al., 2010), which is beneficial in a number of advanced analytical applications such as ECL sensors, field-effect sensors, paper-based analytical devices, and electrochemical sensors/biosensors (Cate et al., 2015; Kimmel et al., 2012; Stine et al., 2013; Su and Lv, 2014). Nevertheless, the practical applications of the graphene sheets, usually denominated as chemically converted graphenes (CCGs), are often limited by two major challenges: how to improve the dispersibility of the CCGs in solutions for practical processability and how to tailor the properties of the CCGs further depending on applications (Su et al., 2009). Several effective methods have been developed to circumvent the obstacles via chemical functionalization of CCGs. They can be classified into two general approaches: the non-covalent functionalization of CCGs through the physisorption of aromatic organic molecules onto basal planes of CCGs, and the covalent functionalization of CCGs with reactive organic species through covalent bond formation onto CCGs (Georgakilas et al., 2012; Kim et al., 2012). Of these two
methods, the non-covalent functionalization of CCGs with pyrene derivatives is particularly attractive since it can provide versatile chemical functionalities onto CCGs without degrading the intrinsic properties of the graphenes. For example, the non-covalent functionalization of graphene with 1-pyrenecarboxylic acid was demonstrated to achieve stable aqueous dispersions of graphenes for multifunctional applications of the graphenes (An et al., 2010). Electrochemiluminescence (ECL) is a unique type of chemiluminescence in which electrogenerated species are involved to form excited states emitting light in the vicinity of electrode surfaces (Forster et al., 2009; Richter, 2004). The ECL provides the beneficial features of chemiluminescence even with better temporal and spatial controllability due to the intrinsic light-generation process of ECL on electrode surfaces. The ECL technique has been thus developed as a powerful tool in many analytical applications such as immunoassays, DNA analysis, and molecular diagnosis of environmentally or clinically relevant compounds (Dennany et al., 2004; Miao, 2008). Recently, the ECL technique has also been used with graphene-related nanomaterials because graphene sheets present many excellent properties including high surface-to-volume ratio (~2,600 m2g-1) and high electron transfer rate (up to 200,000 cm2V-1s-1) for efficient ECL signal transduction (Allen et al., 2010; Schedin et al., 2007). However, the translation of the extraordinary properties of graphenes into the practical ECL applications has some major challenges. Specifically, graphene sheets are only dispersible in a limited number of solvents, such as N-methyl pyrrolidone, N,N-dimethylacetamide, and -butyrolactone, with high surface tension (Hernandez et al., 2008), and are often re-aggregated via - stacking interactions even in these limited solvents, which makes handling and processing of graphene sheets challenging (Mali et al., 2015). The reactivity of pristine graphenes is also
relatively low compared to other carbon allotropes such as carbon nanotubes and fullerenes (Mali et al., 2015), which often requires the use of highly catalytic species for catalyzed electrochemical reactions leading to sensitive ECL-based assays (Gu et al., 2015; Huang et al., 2016; Jiang et al., 2014; Li et al., 2015; Zhao et al., 2015). For example, the use of hemin as a biomimetic catalyst was reported to facilitate the electrocatalytic reduction of oxygen on graphene sheets, resulting in ultrasensitive ECL-based immunoassays (Deng et al., 2013). Recently, our group also reported the covalent decoration of graphenes with amine-terminated dendrimers encapsulating catalytic nanoparticles, which can be utilized for enhanced stable ECL of Ru(bpy)32+/tripropylamine (Kim et al., 2012; Kim and Kim, 2014). In this context, we report the functionalization of CCGs using a water-soluble pyrene derivative 1 (3-((pyren-1-yl)methyl)imidazolium-1-propionate), and the feasibility of the functionalized CCGs to sensitive ECL-based analyses. The compound 1 consisted of a pyrene appended with a 3-(imidazolium)propionate zwitterionic arm (Scheme 1a), and served the dual purpose of improving the dispersion of the CCGs in aqueous solutions and further tailoring the catalytic activity of the CCGs with dendrimer-encapsulated catalytic nanoparticles (DENs). The aromatic pyrene moiety of 1 enables the compound 1 to be anchored onto the hydrophobic surface of the CCG sheets. The stability of the aqueous dispersion of the resulting CCGs was greatly enhanced due to the hydrophilic zwitterionic arm in 1 anchored onto the CCGs, which thus facilitates the processability for integration of the CCGs onto the ITO substrates to develop ECL signal-transduction platforms. In addition, the carboxylic group of the zwitterionic arm in 1 allows the facile secondary functionalization of the CCGs on ITOs with catalytic DENs for sensitive ECL-based analyses. Specifically, we synthesized the pyrene derivative 1 with a 3-
(imidazolium)propionate zwitterionic arm, and prepared stable aqueous dispersions of 1-functionalized CCGs. Based on this dispersion, we functionalized the ITO surfaces by spin-coating of the 1-functionalized CCGs onto ITOs and the subsequent covalent conjugation of catalytic Pt nanoparticles (diameter 1.8 ± 0.2 nm) encapsulated inside amine-terminated polyamidoamine dendrimers (Pt DENs) to the 1-functionalized CCGs on ITOs (Scheme 1b). The resulting ITOs, which we denote as Pt DEN-1CCG/ITOs, exhibited highly enhanced catalytic activity for the electrochemical redox reactions of the luminol/H2O2 ECL system, leading to significantly improved ECL emission. The Pt DEN-1-CCG/ITOs provided two orders-of-magnitude enhancement in the ECL of luminol/H2O2 compared to ECL obtained from bare ITOs, which enabled the sensitive ECL-based analysis of cholesterol.
2. Experimental 2.1. Preparation of Functionalized CCG Dispersions Functionalized CCGs were prepared by the hydrazine-based chemical reduction of GOs in the presence of the synthesized pyrene derivative 1 or 2 (Supplementary Information for details on the synthesis of 1 and 2). The GOs were synthesized according to a modified Hummer’s method as we reported previously (Supplementary Information for details) (Kim et al., 2012). The synthesized GOs were then chemically reduced by adding 0.4 µL of hydrazine (35 wt%) and 2.8 µL of NH4OH (28 wt%) to 4 mL of the GO solution (ca. 0.05 mg/mL) in the presence of the synthesized pyrene derivative 1 (2 mM) or 2 (10 mM). The chemical reduction was carried out at 80 C for 90 min. The resulting functionalized CCGs were collected, purified by centrifugation, and washed with deionized (DI) water several times. The
functionalized CCGs solutions were additionally purified by extensive dialysis up to 10 days to remove any free pyrene derivative 1 or 2. As a control experiment, the CCGs were also prepared using the same method as that used for the functionalized CCGs, but in the absence of 1 or 2.
2.2. Modification of ITO Electrodes with 1-Functionalized CCGs and Pt DENs ITO electrodes were modified by spin-coating of 1-functionalized CCGs on the surface of ITOs and subsequent conjugation of Pt DENs onto the 1-functionalized CCGs on ITOs. The resulting ITOs, i.e. Pt DEN-1-CCG/ITOs, were used as signal transduction platforms for ECL-based analyses. Briefly, ITOs were cleaned as we reported previously (Kim and Kim, 2014). The cleaned ITOs were spin-coated with 1functionalized CCG solution (ca. 0.05 mg/mL) in a spin-coater (JSP4A, JD Tech, Korea) at a rotation speed of 1000 rpm for 60 s. The spin-coated ITOs were baked on a hot plate at 40 C for 2 min. The processing steps of spin-coating and baking were repeated four times. The ITO electrodes were then dried at 40 C for more than 12 h. The resulting modified ITOs were conjugated with Pt DENs via amine - Nhydroxysuccinimide (NHS) coupling. Specifically, the modified ITOs were incubated in 10 mM 2-(N-morpholino)ethanesulfonic acid (MES) buffers (pH 5.2) containing 10 mM 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide methiodide (EDC) and 20 mM NHS at room temperature for 1 h. The ITOs were then rinsed with DI water and incubated again in 10 µM Pt DEN solution at room temperature for 12 h. The ITO electrodes were finally rinsed with DI water and blown dry with a N2 stream, which resulted in the formation of ITOs modified with 1-functionalized CCGs and Pt DENs, i.e. Pt DEN-1-CCG/ITOs. The Pt DENs were synthesized using amine-terminated
dendrimers according to the previous reports (Supplementary Information for details) (Lee et al., 2013; Zhao and Crooks, 1999).
2.3 Electrochemistry, ECL Measurements, and Characterization All electrochemical experiments were carried out with a Model 440 potentiostat (CH Instruments, USA) using a conventional three-electrode electrochemical cell. A Pt wire and a Ag/AgCl (3 M NaCl) electrode were used as a counter and a reference electrode, respectively. The three-electrode cell was connected to the slit of a spectrograph (Acton Standard SP2150, Princeton Instruments, USA) equipped with a charge-coupled device (CCD) camera (PIXIS 100B, Princeton Instruments, USA) for ECL measurements. The ECL-based analyses of cholesterol were performed as described in the Supplementary Information. All other characterization details are also described in the Supplementary Information.
3. Results and Discussion 3.1. 1-Functionalized CCG Dispersion Water-soluble pyrene derivative 1 with a zwitterionic arm was synthesized by the reaction of 1-(bromomethyl)pyrene and methyl 3-(imidazol-1-yl)propionate under N2 atmosphere followed by the hydrolysis of an ester group with NaOH. Methyl 3(imidazole-1-yl)propionate was prepared by the Michael addition of methyl acrylate to imidazole. The synthesized 1 was mixed with homogeneous GO aqueous solution prepared using a modified Hummer’s method (Kim et al., 2012). The mixture was subsequently reduced by hydrazine-based chemical reduction, and then purified via extensive centrifugation and dialysis. This process resulted in the formation of 1functionalized CCGs. Figure 1a shows UV-Vis absorption spectra obtained during
this process. The UV-Vis spectrum of 1 solution reveals well-resolved characteristic absorption bands below 350 nm, originating from the pyrene moiety of 1 (Figure 1a) (An et al., 2010; Salice et al., 2014). The UV-Vis spectrum of GO also exhibited a characteristic band at 232 nm and a shoulder at 305 nm, corresponding to the -* transitions of the aromatic C=C bond and n-* transitions of the C=O bond in GO, respectively (Figure 1a) (Kim et al., 2012). After the reduction of GO in the presence of 1 and subsequent purification, the UV-Vis spectrum shows that the characteristic absorption band of GO at 232 nm was red-shifted and the overall absorbance over the entire spectral region increased, which verifies the chemical reduction of GO to CCG (Figure 1a) (Li et al., 2008). The UV-Vis spectrum of 1-functionalized CCG also exhibited the characteristic absorption bands of 1, which were broadened and redshifted mainly due to the - interactions between the CCG sheets and the aromatic pyrene moiety of 1 (Figure 1a) (Xu et al., 2008). These UV-Vis spectroscopy data indicate the formation of 1-functionalized CCGs, where the pyrene moiety of 1 was anchored onto the surface of CCGs, after the preparation process. The formation of 1functionalized CCGs was also confirmed by fluorescence measurements. Figure 1b shows fluorescence spectra of 1 and 1-functionalized CCG. The fluorescence spectrum of 1 exhibited characteristic emission bands of 1 centered at 395 and 488 nm, which were ascribed to the monomer and excimer emissions of 1, respectively (Ehli et al., 2006). Interestingly, as shown in the fluorescence spectrum of the 1functionalized CCG, the excimer emission of 1 at 488 nm was extensively quenched in the 1-functionalized CCG, while the monomer emission of 1 remained, which indicates that 1 was anchored onto the surface of the CCG sheets as monomers (Xu et al., 2008; Zhang et al., 2010). In addition, Figure 1c shows the FT-IR spectra of GO, 1, and 1-functionalized CCG. Compared with GO, the FT-IR spectrum of 1-
functionalized CCG exhibited a decrease in the strong absorptions at 1728 cm -1, and at 1227 and 1061 cm-1 (corresponding to the stretching vibrations of the C=O and CO bonds, respectively), which indicates its deoxygenation after the hydrazine-based chemical reduction (Pu et al., 2011). The FT-IR spectrum of 1-functionalized CCG also showed the characteristic absorptions of 1. In the spectrum of 1-functionalized CCG, the strong peak at 840 cm-1 and weak peaks at 1388 and 1581 cm-1 are ascribed to the C=C vibration of the pyrene unit of 1 on the surface of CCG sheets (Li et al., 2010; Zhang et al., 2012). The moderated absorptions at 706, 744, and 1149 cm-1 are also associated with the out-of-plane vibration of C-H bonds of pyrene of 1 (Li et al., 2010; Zhang et al., 2012). These FT-IR spectral results indicate the presence of 1 anchored on the CCG sheets in the 1-functionalized CCGs. The atomic force microscope (AFM) measurements of 1-functionalized CCG visualized the functionalization of CCG with 1 (Figure 1d). As shown in Figure 1d (left figure), the height profiles of CCG sheets (prepared using the same method as that used for 1functionalized CCG but without 1) indicate the thickness of the single layered CCG, i.e. 0.99 ± 0.05 nm, which is in agreement with the results previously reported (Li et al., 2008; Su et al., 2009). Compared with the thickness of CCG, the height profiles of 1-functionalized CCG show increased thickness of the single layered 1-functionalized CCG, i.e. 1.75 ± 0.12 nm (right figure in Figure 1d), which suggests the presence of 1 on the basal plane of graphene sheets. Therefore, based on the above results, we conclude the unambiguous presence of 1 anchored on the hydrophobic surface of CCG sheets in the 1-functionalized CCGs. After characterizing 1-functionalized CCG to confirm the existence of 1 attached on the basal plane of CCG, we investigated the dispersion of 1-functionalized CCG in
aqueous solutions since it is important to address the practical processability issue of CCGs for the development of ECL signal-transduction platforms. Figure 2 shows the dynamic light scattering (DLS) data of CCG and 1-functionalized CCG, which compare their size distributions in aqueous solutions. While the aqueous CCG solution exhibited a broad and bimodal size distribution (centered at 363 and 5261 nm, respectively) of CCGs (dashed black line in Figure 2), the 1-functionalized CCG solution displayed a narrow size distribution centered only at 269 nm (solid blue line in Figure 2), which indicates the significantly enhanced aqueous dispersion of 1functionalized CCG when compared with CCG primarily due to the zwitterionic arm of 1 (Supplementary Information, Table S1). To verify the importance of the zwitterionic arm of 1 for dispersion stability, compound 2 (which had a similar structure to 1, but the methylcarboxylate moiety of 1 was removed as shown in Scheme 1a) was also synthesized and subjected to the functionalization of CCG under the same conditions as those of 1 but with the use of 10 mM of 2 rather than 2 mM of 1. Indeed, as shown in Figure 2 (dotted red line), the aqueous dispersion of 2functionalized CCG exhibited significant aggregation of CCG sheets, of which the size distribution was centered at 884 nm, even at the 5 times higher concentration of 2 used to stabilize CCG sheets than that of 1, which clearly demonstrates the importance of the zwitterionic arm of 1 for the dispersion stability of CCGs. These results are also consistent with the previous report of pyrene derivatives, demonstrating that the effectiveness of pyrene derivative for the graphene stabilization is significantly dependent on the functional group attached to the pyrene (Parviz et al., 2012). To quantify the stability of the aqueous dispersions, we also performed zeta potential measurements of CCG, 1-functionalized CCG, and 2functionalized CCG. The zeta potential value of 1-functionalized CCG was found to
be –(33.6 ± 0.7) mV at the original pH of the aqueous dispersion (pH 7), while CCG and 2-functionalized CCG displayed zeta potentials of –(26.4 ± 0.2) mV and +(15.3 ± 0.8) mV, respectively. This suggests that 1-functionalized CCG forms the most stable dispersion at the original pH of the solution displaying a zeta potential larger than ±30 mV, which is a common stability criterion (Parviz et al., 2012). 3.2. Enhanced ECL of Luminol/H2O2 System on Pt DEN-1-CCG/ITO Taking advantage of the stable 1-functionalized CCG dispersion with the improved processability in aqueous solutions, we prepared ECL signal-transduction platforms by spin-coating the aqueous dispersion of 1-functionalized CCG onto ITO surfaces. Since the ITO substrate spin-coated with 1-functionalized CCG (denoted as 1CCG/ITO) provides the carboxylic group in the zwitterionic moiety of 1 as a conjugation site, we functionalized 1-CCG/ITO further with amine-terminated dendrimers encapsulating electrocatalytic Pt nanoparticles (Pt DENs) for efficient ECL generation. Recently, we reported that the modification of ITO with Pt DENs provides enhanced stable ECL generation due to the electrocatalytic activity of Pt DENs (Kim and Kim, 2014). For the secondary functionalization of 1-functionalized CCG on ITO, we thus synthesized Pt DENs which we subsequently conjugated onto 1-CCG/ITO. The Pt nanoparticles were nearly monodispersed in size and rarely aggregated due to stabilization of the nanoparticles via their encapsulation inside dendrimers (Supplementary Information, Figure S1) (Kim et al., 2012; Kim and Kim, 2014; Lee et al., 2013; Zhao and Crooks, 1999). Quantitatively, the size distribution data indicates the fairly uniform size of the Pt nanoparticles, i.e. 1.8 ± 0.2 nm, which is in agreement with the theoretical size (1.79 nm) of the Pt nanoparticles containing 200
atoms (Ye et al., 2007). The synthesized Pt DENs were then conjugated onto 1CCG/ITO using EDC/NHS chemistry for the coupling of the terminal amine group of Pt DEN with the carboxylic group of 1-functionlaized CCG on ITO, which led to the formation of Pt DEN-1-CCG/ITO (Supplementary Information, Figure S2). We then investigated the electrochemical and ECL properties of luminol/H2O2 on the modified ITOs as ECL signal-transduction platforms. Figure 3 shows the cyclic voltammograms (CVs) and corresponding ECL responses of the luminol/H2O2 system on a bare ITO, a 1-CCG/ITO, and a Pt DEN-1-CCG/ITO. Compared with negligible redox currents of luminol/H2O2 on the bare ITO, the 1-CCG/ITO displayed significantly increased currents attributable to the facile redox reactions of luminol/H2O2 (dashed red line in Figure 3a). More importantly, the Pt DEN-1CCG/ITO exhibited even further increase in the redox currents of luminol/H2O2 (solid blue line in Figure 3a), which is attributed to the catalytic activity of Pt nanoparticles on the Pt DEN-1-CCG/ITO (Kim and Kim, 2014). The increased currents were not observed on a similar platform (but without Pt nanoparticles), i.e. dendrimer-1CCG/ITO (Supplementary Information, Figure S3). Accordingly, the Pt DEN-1CCG/ITO exhibited much enhanced ECL emission of luminol/H2O2 compared with the emissions obtained with the bare ITO and 1-CCG/ITO (Figure 3b). It is also important to note that the enhanced ECL emission starts from ca. 0.3 V, consistent with the onset potential for the oxidation of luminol on the Pt DEN-1-CCG/ITO, which is much lower than on the bare ITO (ca. 0.6 V) (Figure 3a). The low required potential for ECL emission would be beneficial in avoiding possible interferences during ECL-based analyses (Lee et al., 2015). As shown in Figure 4, we also performed spooling ECL spectroscopy of the luminol/H2O2 system on the bare ITO, 1-CCG/ITO, and Pt DEN-1-CCG/ITO electrodes in the same potential window used
in the ECL-potential curves shown in Figure 3b. This allowed us to track the ECL wavelength as well as the intensity as a function of the applied potential (i.e., time) in order to obtain further information on the ECL reactions of the luminol/H2O2 system on the electrodes (Hesari et al., 2014; Lee et al., 2015). Concurring with the ECLpotential curves in Figure 3b, similar trends of ECL evolution and devolution were obtained on the electrodes; especially, much enhanced ECL emission of luminol/H2O2 was observed on the Pt DEN-1-CCG/ITO (Figure 4a) compared with the emissions on the bare ITO (Figure 4c) and 1-CCG/ITO (Figure 4b). The inset in Figure 4a compares the ECL spectra of luminol/H2O2 on the Pt DEN-1-CCG/ITO (solid blue line), 1-CCG/ITO (dashed red line), and bare ITO (dotted black line) obtained upon the potential applied at 0.5 V (or 0.7 V for the bare ITO). This indicates that the ECL intensity integrated over the wavelength, obtained with the Pt DEN-1-CCG/ITO (i.e., (8.3 ± 0.4) x 105 a.u.), was ca. 138 times larger than that of the bare ITO (i.e., (6.0 ± 0.3) x 103 a.u.). The integrated ECL intensity obtained with the Pt DEN-1-CCG/ITO was also ca. 2 times greater than that of the 1-CCG/ITO (i.e., (4.5 ± 0.2) x 105 a.u.). 3.3. Sensitive ECL-based Analysis of Cholesterol After confirming the enhanced ECL emission of the luminol/H2O2 system on the Pt DEN-1-CCG/ITO, we demonstrated the feasibility of the Pt DEN-1-CCG/ITO as an ECL signal transduction platform by using cholesterol as a model target analyte. Cholesterol is a principal sterol biosynthesized by mammalian cells, serving as an essential structural component of mammalian cell membranes and a precursor for the synthesis of steroid hormones, bile acids, and fat-soluble vitamins (Hong et al., 2013).
The ECL-based analysis of cholesterol is feasible with the Pt DEN-1-CCG/ITO because of the enhanced ECL of luminol on the Pt DEN-1-CCG/ITO with H2O2 produced by oxidation of cholesterol in the presence of cholesterol oxidase (ChOx). Figure 5 shows a calibration curve obtained with the Pt DEN-1-CCG/ITO (solid blue line) for the ECL-based analysis of cholesterol, which represents a good linear relationship between the ECL intensity and the cholesterol concentration ranging up to 100 M with a correlation coefficient of 0.986 (inset in Figure 5). The limit of detection (LOD) of cholesterol was determined to be 4.8 M (k = 3). In particular, we also found that the sensitivity of the ECL-based analysis of cholesterol with the Pt DEN-1-CCG/ITO was 43.20 a.u./M, which is ca. 154 times higher than that obtained with the bare ITO (dotted black line in Figure 5), consistent with the ca. 138fold enhanced ECL of luminol/H2O2 on the Pt DEN-1-CCG/ITO as discussed earlier. The long-term stability of the Pt DEN-1-CCG/ITOs was also tested by comparing the ECL responses obtained with fresh and 2 months-old electrodes to cholesterol. The ECL intensities changed negligibly compared to the initial ECL even after the 2months storage of the Pt DEN-1-CCG/ITO (Supplementary Information, Table S2), which confirms the high stability of the Pt DEN-1-CCG/ITO for the ECL-based analysis of cholesterol. In addition, the reliability of the Pt DEN-1-CCG/ITO to the practical ECL-based analysis of cholesterol in serum samples was evaluated by the standard addition method for recovery measurements as previously reported (Hong et al., 2013). The recoveries of spiked 10 ~ 30 M cholesterols were found to be between 93.6 to 106.1% (Supplementary Information, Table S3), indicating the acceptable reliability of the Pt DEN-1-CCG/ITO for the ECL-based analysis of cholesterol in serums. These analytical performances of the Pt DEN-1-CCG/ITO are comparable to other state-of-the-art cholesterol sensors (Supplementary Information,
Table S4), but could be improved further by optimizing the cholesterol-based assay conditions such as enzyme concentration, reaction buffer, incubation time, and so on. Most importantly, these results clearly demonstrate the feasible tailoring of CCGs using the functional pyrene derivative 1, facilitating the integration and the functionalization of the CCGs to develop the Pt DEN-1-CCG/ITO as an efficient signal transduction platform for sensitive ECL-based analyses.
4. Conclusions In summary, we described the facile tailoring of CCGs using a water-soluble pyrene derivative 1 with a zwitterionic arm for the purpose of highly sensitive ECLbased analytical applications. Two orders-of-magnitude enhancement in the ECL emission of the luminol/H2O2 ECL system was achieved with the tailored CCGs on ITOs compared to that obtained with bare ITOs. The highly enhanced ECL of luminol/H2O2 allowed a ca. 154 times more sensitive ECL-based analysis of cholesterol as a preliminary analytical application of the tailored CCGs on ITOs compared with the use of bare ITOs. Currently, we are attempting to achieve better catalytic nanoparticles encapsulated inside a dendrimer, which can be conjugated to the 1-functionalized CCGs as demonstrated in the present study, for ECL applications. We envision that the use of the pyrene derivative 1 can be a facile and versatile way to tailor CCGs with a variety of catalytic nanoparticles, such as monometallic, bimetallic alloy, and even core-shell structured nanoparticles (Scott et al., 2005), encapsulated inside dendrimers, which provides controllable and tunable properties for sensitive ECL-based analyses.
Acknowledgment This work was supported by the National Research Foundation of Korea funded by the Ministry of Science, ICT and Future Planning (NRF-2014S1A2A2028540 and NRF-2014R1A1A2058218), and the Agency for Defense Development through Chemical and Biological Defense Research Center.
References Allen, M.J., Tung, V.C., Kaner, R.B., 2010. Chem. Rev. 110, 132-145. An, X., Simmons, T., Shah, R., Wolfe, C., Lewis, K.M., Washington, M., Nayak, S.K., Talapatra, S., Kar, S., 2010. Nano Lett. 10, 4295-4301. Cate, D.M., Adkins, J.A., Mettakoonpitak, J., Henry, C.S., 2015. Anal. Chem. 87, 1941. Deng, S., Lei, J., Huang, Y., Cheng, Y., Ju, H., 2013. Anal. Chem. 85, 5390-5396. Dennany, L., Forster, R.J., White, B., Smyth, M., Rusling, J.F., 2004. J. Am. Chem. Soc. 126, 8835-8841. Ehli, C., Rahman, G.M.A., Jux, N., Balbinot, D., Guldi, D.M., Paolucci, F., Marcaccio, M., Paolucci, D., Melle-Franco, M., Zerbetto, F., Campidelli, S., Prato, M., 2006. J. Am. Chem. Soc. 128, 11222-11231. Forster, R.J., Bertoncello, P., Keyes, T.E., 2009. Annu. Rev. Anal. Chem. 2, 359-385. Georgakilas, V., Otyepka, M., Bourlinos, A.B., Chandra, V., Kim, N., Kemp, K.C., Hobza, P., Zboril, R., Kim, K.S., 2012. Chem. Rev. 112, 6156-6214. Gu, W., Deng, X., Gu, X., Jia, X., Lou, B., Zhang, X., Li, J., Wang, E., 2015. Anal. Chem. 87, 1876-1881.
Parviz, D., Das, S., Ahmed, H.S.T., Irin, F., Bhattacharia, S., Green, M.J., 2012. ACS Nano 6, 8857-8867. Pu, J., Wan, S., Zhao, W., Mo, Y., Zhang, X., Wang, L., Xue, Q., 2011. J. Phys. Chem. C 115, 13275-13284. Richter, M.M., 2004. Chem. Rev. 104, 3003-3036. Salice, P., Gambarin, A., Daldosso, N., Mancin, F., Menna, E., 2014. J. Phys. Chem. C 118, 27028-27038. Schedin, F., Geim, A.K., Morozov, S.V., Hill, E.W., Blake, P., Katsnelson, M.I., Novoselov, K.S., 2007. Nat. Mater. 6, 652-655. Scott, R.W.J., Wilson, O.M., Crooks, R.M., 2005. J. Phys. Chem. B 109, 692-704. Stine, R., Mulvaney, S.P., Robinson, J.T., Tamanaha, C.R., Sheehan, P.E., 2013. Anal. Chem. 85, 509-521. Su, Q., Pang, S., Alijani, V., Li, C., Feng, X., Muellen, K., 2009. Adv. Mater. 21, 3191-3195. Su, Y., Lv, Y., 2014. RSC Adv. 4, 29324-29339. Xu, Y., Bai, H., Lu, G., Li, C., Shi, G., 2008. J. Am. Chem. Soc. 130, 5856-5857. Ye, H., Crooks, J.A., Crooks, R.M., 2007. Langmuir 23, 11901-11906. Zhang, L., Li, C., Liu, A., Shi, G., 2012. J. Mater. Chem. 22, 8438-8443. Zhang, M., Parajuli, R.R., Mastrogiovanni, D., Dai, B., Lo, P., Cheung, W., Brukh, R., Chiu, P.L., Zhou, T., Liu, Z., Garfunkel, E., He, H., 2010. Small 6, 1100-1107. Zhao, H.-F., Liang, R.-P., Wang, J.-W., Qiu, J.-D., 2015. Biosens. Bioelectron. 63, 458-464. Zhao, M., Crooks, R.M., 1999. Adv. Mater. 11, 217-220.
Figure Captions Scheme 1. (a) Structure formulas of pyrene derivatives (1 and 2). (b) Schematic illustration of preparation process of Pt DEN-1-CCG/ITO. Figure 1. (a) UV-Vis absorption spectra of 1-CCG (ca. 0.05 mg/mL), GO (ca. 0.05 mg/mL), and 1 (30 μM). (b) PL spectra of 1 (2 mM) and 1-CCG (ca. 0.05 mg/mL). (c) FT-IR spectra of GO, 1, and 1-CCG. (d) AFM images and height profiles of CCG (left) and 1-CCG (right). Figure 2. DLS graphs of CCG, 1-CCG, and 2-CCG. Figure 3. (a) CVs and (b) ECL curves of luminol (1 mM) and H2O2 (1 mM) in phosphate-buffered saline (PBS) solution (0.1 M, pH 8.0), obtained on Pt DEN-1CCG/ITO (solid blue), 1-CCG/ITO (dashed red), and bare ITO (dotted black). Scan rate: 50 mV·s-1. Integration time: 1 s. Figure 4. Spooling ECL spectra of luminol (1 mM) and H2O2 (1 mM) in PBS solution (0.1 M, pH 8.0) obtained on (a) Pt DEN-1-CCG/ITO, (b) 1-CCG/ITO, and (c) bare ITO. Scan rate: 50 mV·s-1. Cycled potential range: -0.8 V ~ 0.8 V (i.e., -0.8 V 0.8 V -0.8 V). Integration time: 1 s. The red colored ECL spectra represent the ECL evolution during the potential scanning. The inset in (a) shows the ECL spectra of the Pt DEN-1-CCG/ITO (solid blue), 1-CCG/ITO (dashed red), and bare ITO (dotted black) obtained upon the potential applied at 0.5 V (or 0.7 V for the bare ITO). The inset in (c) shows the expanded view of the spooling ECL spectra obtained on the bare ITO. Figure 5. Calibration curves obtained with Pt DEN-1-CCG/ITO (solid blue) and bare ITO (dotted black) for ECL-based analysis of cholesterol in phosphate buffer (0.1 M, pH 8.0) containing 1 mM of luminol during the potential scan between -0.8 and +0.8 V and between -0.8 and +1.0 V (vs. Ag/AgCl), respectively. Inset shows an expanded
view of the low-concentration region. Scan rate: 100 mV·s-1. Integration time: 32 s (or 36 s for the bare ITO).
Scheme 1
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Graphical Abstract
Highlights
A water-soluble pyrene derivative compound with a zwitterionic arm was synthesized. Tailoring of chemically converted graphene (CCG) was demonstrated using the pyrene derivative. Use of the tailored CCG allowed a 138-fold enhancement in the electrochemiluminescence of luminol. The enhanced electrochemiluminescence allowed a sensitive analysis of cholesterol.
