New Titanium Dioxide-Based Heterojunction Nanohybrid for Highly

Subscriber access provided by PEPPERDINE UNIV

Article

A New Titanium Dioxide Based Heterojunction Nanohybrid for Highly Selective Photoelectrochemical–Electrochemical Dual Mode Sensor Muthuchamy Nallal, Gopalan Anantha Iyengar, and Kwang-Pill Lee ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b10519 • Publication Date (Web): 27 Sep 2017 Downloaded from http://pubs.acs.org on September 28, 2017

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.

ACS Applied Materials & Interfaces 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 58

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

ACS Applied Materials & Interfaces

A New Titanium Dioxide Based Heterojunction Nanohybrid for Highly Selective Photoelectrochemical–Electrochemical Dual Mode Sensor Muthuchamy Nallal1,2,#, Gopalan Anantha Iyengar 2, 3,# and Kwang Pill-Lee1,2 3,*

1 2

Department of Chemistry Education, Kyungpook National University, Daegu, South Korea

Research Institute of Advanced Energy Technology, Kyungpook National University, Daegu, South Korea 3

Department of Nanoscience and Nanotechnology, Kyungpook National University, Daegu, South Korea #: Authors contributed equally to this work *Corresponding author: [email protected]

1 ACS Paragon Plus Environment


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 58

ABSTRACT: A new titanium dioxide (TiO2) based heterojunction nanohybrid (HJNH), composed of TiO2, graphene (G), poly[3-aminophenylboronic acid] (PAPBA) and gold nanoparticles (Au NPs), was synthesized and designated as TiO2(G) NW@PAPBA-Au HJNH.

The

TiO2(G)

NW@PAPBA-Au

HJNH

possesses

dual

mode

signal

photoelectrochemical (PEC) and electrochemical transduction capabilities to sense glucose and glycated hemoglobin (HbA1c) independently. The synthesis of HJNH material involved two sequential stages: i) the simple electrospinning synthesis of G-embedded TiO2 nanowires (TiO2(G) NWs) and ii) a one-step synthesis of Au NPs dispersed PAPBA nanocomposite (NC) in the presence of TiO2(G) NW. The as-synthesized TiO2(G) NW@PAPBA-Au HJNH was characterized by field emission scanning electron microscopy, field emission transmission electron microscope, X-ray diffraction, X-ray photoelectron spectroscopy, Fourier transform infrared, thermogravimetric analyses, and UV–vis diffuse reflectance spectroscopy. A PEC platform was developed with TiO2(G) NW@PAPBA-Au HJNH for the selective detection of glucose without any enzyme auxiliary. The PEC glucose sensor presents an acceptable linear range (from 0.5 mM to 28 mM), good sensitivity (549.58 µA mM–1 cm–2), and low detection limit (0.11 mM), that are suited for the diabetes glucose monitoring. Besides, the boronic acid groups in PAPBA were utilized as a host to capture the HbA1c. We fabricated the electrochemical HbA1c sensor based on monitoring the electrocatalytic reduction current of hydrogen peroxide produced by HbA1c tethered to the sensor probe. The amperometric electrochemical sensor for HbA1c exhibited linear responses to HbA1c levels from 2.0 % to 10% (with a detection limit of 0.17%). Notably, the performances of the fabricated glucose and HbA1c sensors are superior in the dual signal transduction modes as compared to the literature suggesting the significance of the newly designed bifunctional TiO2(G) NW@PAPBA-Au HJNH. KEYWORDS: Titanium dioxide; Nanohybrid; Dual mode sensor; Glucose; HbA1c 2 ACS Paragon Plus Environment

Page 3 of 58

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


1. INTRODUCTION Titanium dioxide (TiO2) has been regarded as the most promising photo- and photoelectrochemical (PEC) catalyst owing to its strong light absorption and favorable conduction band (CB) edge/positions.1,2,3

The PEC process for the TiO2 based catalysts

depend on the efficiency of separation of the hole (h+) and electron (e–) across the heterogeneous interface.4 Many effective strategies have been explored to achieve efficient h+–e- pair separation and extension of light absorption to the visible region for TiO2 based materials. The promising strategies are i) inclusion of light absorbing materials (sensitizer and quantum dots) and ii) creation of heterojunction with an another organic semiconductor conducting polymer5 or nanostructured carbon materials (graphene (G)6 or metal nanoparticle such as Au NPs7. TiO2 could also be integrated with more components to achieve high performances for the PEC processes. Particularly, the hetero junctions at the TiO/carbon nanostructure or TiO2/metal interface provides beneficial characteristics such as charge region, facile e– injection/transfer from TiO2 and extension of light absorption to the visible region. For example, the heterojunction generated between TiO2 and Au NPs extended the water splitting photoactivity from ultraviolet to the visible region.8 The Au/ TiO2 hybrid exhibited synergistic coupling effects between surface plasmon resonance (SPR) of Au NP and optical property of TiO2.9 These reports suggest that inclusion of Au NPs into photoactive TiO2 could result in electric field amplification as well in the injection of SPRgenerated e– into the CB of TiO2. Moreover, as compared to bulk TiO2, nanostructured TiO2 materials exhibit enhanced quantum efficiencies for the photocatalytic process as they have minimum distance necessary for minority carriers to diffuse towards the surface.10 In particular, one-dimensional (1D) TiO2 nanomaterials such as nanowires (NWs) are preferably used in PEC devices over the other nano geometries due to their large specific surface areas, direct e– pathways,11 ability to decouple the direction of light absorption and charge carrier 3 ACS Paragon Plus Environment


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

generation.12 Furthermore, the hybrid materials based on multicomponent heterojunction, such as semiconductor-metal-TiO2,13 exhibit much more effective h+/e– charge separation and photocatalytic properties as compared to the TiO2 based on bifunctional counter parts. As compared to pristine TiO2, TiO2-G composites exhibit enhanced photocatalytic properties due to three main reasons: efficient h+/e– pair separation, an extension of light absorption to the visible region and efficient absorption of light. The heterojunction between TiO2 and G augments the separation of the h+ and e– as G accepts the photogenerated e–s, and h+s in TiO2 are preferably utilized for photooxidation of the reactant. However, the TiO2-G NCs prepared by physical mixing of TiO2 and G exhibit limited stability and photocatalytic efficiency. In our earlier report, we synthesized G - TiO2 NWs hybrid via the combined electrospinning-hydrothermal approach and demonstrated the enhanced charge carrier separation.14–17 Conducting polymers, such polyaniline (PANI), possess extended π - e– system in their polymeric chains and behave as p-type semiconductors. PANI, with a band gap of 2.8 eV can sensitize TiO2. Thus, p-n heterojunctions were tailored between TiO2 and PANI to expand the light absorption of TiO2 into the visible region.18 The superior photocatalytic performance of TiO2-PANI under visible light irradiation was attributed to the high absorption coefficient of PANI in the visible region, efficient separation of h+-e– pair in the excited state and the alignment of the conduction band (CB) energy level of TiO2 with the unoccupied lowest molecular orbital (LUMO) of PANI.18 The electrons generated through ππ* absorption of PANI can be transferred to the CB of TiO2.19 On perusal of literature, it is understood that there is a growing interest in developing heterojunction based TiO2 materials to improve the efficiency of the photocatalytic system by including two or three or more components that can contribute to visible light absorption and h+-e– separation capabilities. In this work, we design and develop a new heterojunction nanohybrid (HJNH) to exhibit dual PEC/electrochemical catalysis to evolve a novel PEC/electrochemical dual sensing strategy 4 ACS Paragon Plus Environment

Page 4 of 58

Page 5 of 58

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


through the judicious selection of components such as TiO2 NW, G, Au NP and a PANI derivative. PEC sensors attract research because of the low background signal capability and facile bioassay.20-22 The PEC sensing mechanism involves reduction of the analyte by the photogenerated e– or oxidation by the corresponding h+.23 However, the selectivity of the PEC sensor is limited because of the intrinsic oxidation of the analyte molecules by the photogenerated h+. This problem is partially circumvented through the inclusion of certain specific biomolecules, such as enzymes,24 antibodies, cells,25 or DNA26 into the sensor probe which can specifically bind the target analyte. Although the inclusion of additional biomolecules, such as enzymes, into PEC biosensors contributes to good selectivity, it involves a complex immobilization process, stringent storage conditions, instability in the sensing environment, and high costs. Hence, there is a great demand in designing the fabrication of a highly selective PEC biosensor without including any enzyme. In this respect, there is an option of utilization of a material that mimicks the catalytic property of an enzyme. Functional polymers or biomimetic artificial receptors, that can have specific binding to the analyte, are suitable for improving PEC selectivity.27 For example, boronic acid-containing polymers or small molecule derivatives are utilized as glucose-sensing molecular probes as they can function as a receptor for 1,2-diols such as glucose via efficient reversible formation of covalent bonds to generate cyclic boronate complexes.28,29 As compared to enzyme-based (e.g., glucose oxidase) biosensors, boronic acid-based glucose sensors are more promising due to their high stability, easy fabrication, and regeneration ability. Poly[3aminophenylboronic acid] (PAPBA), a PANI derivative, has been proved to be an efficient artificial receptor for the detection of glucose based on its affinity interactions with glucose30 Au NPs exhibit distinct electrochemical and optical properties that are well suited for the



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

fabrication of electrochemical/optical biosensors.31 Also, the combination of Au NPs and PANI or PANI derivatives exhibited enhanced sensor characteristics.32 Electrochemical glucose sensor (glucometers) is widely utilized for the diagnosis and management of diabetes mellitus. However, owing to the fluctuations in blood glucose levels, because of exercise and intake of carbohydrates, the spot test of glucose at the moment does not reflect the average glucose level over a long period. Alternatively, the glycated percentage hemoglobin (%HbA1c) level (the ratio between HbA1c and total hemoglobin concentration) is formally endorsed in many counties as a diagnostic marker for diabetic patients, in addition to the spot measurement of glucose level by glucometer.33 The %HbA1c gives an integrated index of blood glucose levels over the entire period of 120 days. Unlike the spot blood glucose level determined by a glucometer, the value of HbA1c is very stable, not affected by prandial status and has no diurnal rhythms. Hence, HbA1c levels, which can be measured at any time of a day is considered as the crucial parameter for the diagnosis of diabetes. However, a rationally designed sensor capable of detecting both glucose level and HbA1c, with an ability to detect both the parameters in the diabetes monitoring levels would provide an important advance to allow delineation of the relationship between them. Development of dual mode sensors based on two simultaneous detection approaches is a promising concept and recently have been applied to detect two independent analytes (Table 1). On perusal of Table 1, it can be noticed that either simultaneous optical or electrochemical signal transduction modes are employed. In this work, an innovative PEC/electrochemical dual sensing platform has been proposed for the detection of glucose and HbA1c, independently. However, the fabrication of sensors for both glucose and HbA1c involving dual PEC/electrochemical signal transaction modes requires a genuine design of a new bifunctional (photo and electro active) material. In this work, we designed and successfully developed a new HJNH based on TiO2(G) NWs and PAPBA-Au NC (denoted as


Page 6 of 58

Page 7 of 58

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


TiO2(G) NW@PAPBA-Au HJNH) (Scheme 1) giving due consideration to the enhanced photocatalytic properties of G-TiO2 hybrid, affinity binding properties of PAPBA with glucose and electrocatalytic properties of PAPBA-Au NC.49 The newly developed TiO2(G) NW@PAPBA-Au HJNH synergistically combines the photocatalytic property of TiO2, electrocatalytic property of Au NPs, the efficient e– transport capacity of G and the glucose binding ability of PAPBA. We demonstrated the utility of the TiO2(G) NW@PAPBA-Au HJNR for PEC sensing of glucose and electrochemical sensing of HbA1c. The fabricated enzyme-free PEC glucose/electrochemical HbA1c dual sensor, based on TiO2(G) NW@PAPBA-Au HJNR, exhibited superior glucose/HbA1C sensor characteristics as compared to the sensing performances of other dual mode based sensors reported in literature (Table 1). We propose a probable PEC sensing mechanism for glucose and the mechanism for the electrochemical behavior of the TiO2(G) NW@PAPBA-Au HJNH towards HbA1c detection. This study form is expected the basis for the development of other bifunctional smart materials to suit for targeted applications.

2. EXPERIMENTAL 2.1. Chemicals G nanoplatelet sheets were obtained from E NanoTEC (Korea). HAuCl4, p-TSA, 3APBA, TiIP, glucose, and HbA1c were purchased from Sigma–Aldrich and used as received. Phosphate buffer saline solution (PBS, pH = 7) and ethanol were obtained from OCI Company Ltd., South Korea. 2.2. Preparation of TiO2(G) NW@PAPBA-Au HJNH Two main stages were involved in the synthesis of TiO2(G) NW@PAPBA-Au HJNH: i) the preparation of TiO2(G) NWs and ii) preparation of PAPBA-Au NC on TiO2(G) NWs. 7 ACS Paragon Plus Environment


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

2.2.1. Preparation of TiO2(G) NWs TiO2(G) NWs were prepared by a combined electrospinning-hydrothermal processes as previously described.15 For comparative purposes, TiO2 NWs were also prepared without using G-COOH during the electrospinning step. 2.2.2. Preparation of TiO2(G) NW@PAPBA-Au HJNH A solution (40 mL) containing 3-APBA (20 mM) and TiO2(G) NWs (10 mg) was prepared in 0.1 M p-TSA and stirred for 60 min. About 10 mL of 5 mM aqueous HAuCl4 was dropwise added to the above solution. The reaction was continued for 24 h and the obtained dark-green precipitate, TiO2(G) NW@PAPBA-Au HJNH, was washed several times with 0.1 M p-TSA. The TiO2(G) NW@PAPBA-Au HJNH was dried at 60 °C in an air oven. Two reference materials, PAPBA-Au NC and TiO2 NW@PAPBA-Au NC, were prepared for the comparison purposes. The PAPBA-Au NC was prepared by the dropwise addition of HAuCl4 (5 mM, 10 mL) to the 3-APBA (20 mM, 40 mL) solution in 0.1 M p-TSA. The TiO2 NW@PAPBA-Au NC was prepared as follows; A solution (40 mL) containing 3-APBA (20 mM in 0.1 M p-TSA) and TiO2 NWs (10 mg) was prepared and stirred for 60 min. About 10 mL of HAuCl4 (5 mM) was dropwise added to the mixture solution containing 3-APBA (20 mM, 40 mL) and

TiO2 NWs (10 mg). The obtained PAPBA-Au NC and TiO2

NW@PAPBA-Au NC were thoroughly washed with 0.1 M p-TSA and dried at 60 °C in an air oven 2.3. Fabrication of the PEC Electrode Before fabrication of the PEC electrode, the ITO substrates (surface resistance = 10 Ω) were sequentially sonicated in acetone, ethanol, and distilled water (20 min each), followed by drying under a stream of nitrogen gas. A dispersion of TiO2(G) NW@PAPBA-Au HJNH (2 mg) was prepared in a 3:2 (v/v) mixture of isopropyl alcohol and Nafion. The slurry was 8 ACS Paragon Plus Environment

Page 8 of 58

Page 9 of 58

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


sonicated for 10 min, and about 5 µL of the resulting ink was drop-cast on the surface of a pre-cleaned ITO electrode (area = 0.25 cm2) and dried at room temperature. 2.4. Characterization The morphology and elemntal compositions were characterized by field emission scanning electron microscopy (FE-SEM) with energy dispersive X-ray spectroscopy (EDX) (Hitachi S-4300, Japan) and field emission transmission electron microscopy with selected area (electron) diffraction (FE-TEM (SAED), 200 kV, DS114). X-ray photoelectron spectroscopy (XPS) measurements were conducted using a Quantera SXM, ULVAC-PHI with monochromatic Al-Kα radiation in the binding energy range 0 – 1400 eV. The crystal structure was investigated by X-ray diffraction (XRD, Quantera SXM, ULVAC-PHI, Japan) using Cu Kα radiation (λ = 1.5406 Å) in the 2θ range of 20 to 85°. Fourier transform infrared (FTIR) spectra were recorded (1486.6 eV) on a Bomem MB 100 FTIR spectrometer (ABB Bomem, QC, Canada) in the range of 400–4000 cm–1. The optical properties were characterized by a UV–vis spectrometer with a diffuse reflectance UV/vis/NIR attachment (Cary 5000, Agilent, Korea). Thermogravimetric analysis was performed at a scanning rate of 10 CO min-1 using a thermogravimetric analyzer (Q600, TA Instrument, New Castle, DE, USA) with samples placed in a nitrogen atmosphere (temperature range from 30 to 900 CO). Electrochemical measurements were performed using an Ivium-Stat (Netherland) electrochemical interface with a PC-controlled analyzer workstation. A conventional threeelectrode cell assembly was used. Electrodes modified with TiO2(G) NW@PAPBA-Au HJNH, TiO2 NW@PAPBA-Au NC, and PAPBA-Au NC were used as the working electrode. Ag/AgCl and a platinum wire were used as the reference and counter electrodes, respectively. The electrocatalytic activities of TiO2(G) NW@PAPBA-Au HJNH, TiO2 NW@PAPBA-Au NC, and PAPBA-Au NC modified electrodes were evaluated using cyclic voltammetry in 0.1 M PBS (pH 7.0). Electrochemical impedance spectroscopy (EIS) measurements were 9 ACS Paragon Plus Environment


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

performed in a background solution of 5 mM K3Fe(CN)6/K4Fe(CN)6 (1:1) and 0.1 M KCl within a frequency range of 10 mHz to 100 kHz. The amplitude of the alternating voltage was 5 mV. 2.5. PEC Glucose Sensing PEC measurements were carried out under irradiation with a light source (Viller Lourmat, France, monochromator, λ = 365 nm) placed at a distance of 10 cm from the working electrode surface. Glucose sensing measurements were performed by immersing TiO2(G) NW@PAPBA-Au HJNH in a 0.1 M PBS (pH 7.0) solution containing different concentrations of glucose in an electrochemical cell, using chronoamperometry as a measuring technique for PEC glucose detection under light irradiation. 2.6. Electrochemical HbA1c Detection The electrochemical detection of HbA1c using TiO2(G) NW@PAPBA-Au HJNH was performed by chronoamperometry. Before detection, the HbA1c sample (in 0.1 M PBS solution, pH 7.0) was diluted to different concentrations, with dilution percentages of 2, 4, 6, 8, and 10%. About 50µL of HbA1c (required %) in 0.1 M PBS was dropped onto the electrode covered with TiO2(G) NW@PAPBA-Au HJNH and incubated for 15 min to immobilize HbA1c on to TiO2(G) NW@PAPBA-Au HJNH. After a gentle rinse with deionized water, the determination of HbA1c (%) was carried out by chronoamperometry at an applied bias potential of –0.70 V in the presence of hydrogen peroxide (3 mM).

3. RESULTS AND DISCUSSION 3.1. Preparation of TiO2(G)-PAPBA-Au HJNH


Page 10 of 58

Page 11 of 58

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


TiO2(G) NW@PAPBA-Au HJNH was synthesized based on the procedure depicted in Scheme 1. Typically, the synthesis involved two major steps: i) preparation of TiO2(G) NW and ii) formation of PAPBA-Au NC in the presence of TiO2(G) NW. Briefly, TiO2(G) NW were first prepared via a combined electrospinning-hydrothermal process utilizing titanium isopropoxide (TiIP) and carboxylic acid-functionalized G (G-COOH) precursors.15 Electrospinning is a simple and cost effective technique for the mass production of metal oxide NWs. Utilizing the combined electrospinning-hydrothermal processes, we prepared TiO2(G) NW. Subsequently, PAPBA-Au NC was formed in the presence of TiO2(G) NWs by a one-step method through simultaneous generation of PAPBA and Au NPs. APBA was oxidized by HAuCl4 to produce PAPBA, and the Au ions were concomitantly reduced to Au0 to generate PAPBA-Au NC. We propose the following reaction mechanism for the deposition of PAPBA-Au NC on the surface of TiO2(G) NW during the oxidative chemical polymerization of APBA by HAuCl4 in the presence of TiO2(G) NW. During the reaction, the HAuCl4- protonates the –NH2 group of APBA to result in the formation of aniline cations, which subsequently generate an aniline radical cations with the release of e–s. AuCl4- acts as an oxidizing agent and oxidizes the anilinium cation form of APBA to PAPBA. The Au3+ ions are simultaneously reduced to Au atoms, coalesce and nucleate to Au NPs. The Au NPs are also stabilized by the PAPBA. There could possible bonding interactions between TiO2(G) NW and PAPBA. The bonding interactions include the hydrogen bonding between the –NH2 groups in PAPBA and –OH groups on the surface of TiO2(G) NW and co-ordination interaction between the N atom in –NH2 group of PAPBA and Ti atoms. Thus, we envisage the generation of the heterojunction between TiO2(G) NW and PAPBA-Au NC. 3.2. Characterization of TiO2(G) NW@PAPBA-Au HJNH 3.2.1. Morphology and Elemental Compositions



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

We demonstrated the morphology and elemental compositions of TiO2(G) NW@PAPBA-Au HJNH by combining the results derived from FESEM, HRTEM and XPS. Panels A and B in Figure 1 show FESEM and TEM images of TiO2(G) NW@PAPBA-Au HJNH. Panel A in Figure 1 reveals the presence of cluster rod like nanostructures (average diameter of ~300 nm) with rough outer surfaces. The nanorod surface consists of randomly distributed spherical particles (bright white spots in the image) with sizes in the ranges 20 – 40 nm (Figure 1B). A thorough examination of the single rod at a higher resolution TEM image (Figure 1C) reveals that the rod is comprised of two distinguishable inner and outer phases, namely, the dark (inner) and diffused (outer) phases. The inner part of the nanorods is expected to be the TiO2(G) NW with a dimeter of 270 nm. And, the outer region has a diameter of ~30 nm with randomly distributed spherical particles on the surface. Pristine TiO2(G) NWs (Figure S1) shows a wire-like morphology with average diameters of 150–180 nm with randomly dispersed G sheets within the NWs. However, upon polymerization of APBA using HAuCl4, the diameter of pristine TiO2(G) NW increased from 150-180 nm to ~300 nm because of the coating of PAPBA-Au NC on the surface of TiO2(G) NW. The details of morphology and elemental compositions of the other reference materials, TiO2 NW@PAPBA-Au NC and PAPBA-Au NC, are presented in Figure S2. Figure 1C (inset) shows the SAED pattern of the Au particles in the vertical direction of the nanorod (indicated by an arrow). One could identify three different ring patterns; i) mainly indexed (110), (220), and (200) planes of the face-centered cubic (fcc) Au lattice, ii) (101) plane of anatase TiO2, and iii) well-defined (100) and (110) diffraction spots of G. The variations in the e– densities of Au, TiO2, and G, causes differences in lattice parameters. The corresponding high-resolution TEM (HRTEM) image of TiO2(G) NW@PAPBA-Au HJNH (Figure 1D) shows three clear lattice fringes with interspacing values of 0.348, 0.241, and 0.139 nm, corresponding to the (101) plane of anatase, the (111) plane of Au, and carbon12 ACS Paragon Plus Environment

Page 12 of 58

Page 13 of 58

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


carbon spacings in G, respectively. Altogether, HRTEM (Figure 1D) and SAED (Figure 1C (inset)) confirmed the co-existence of three crystalline materials, TiO2, G, and Au in TiO2(G) NW@PAPBA-Au HJNH. EDAX analysis (Figure S3) also corroborates with the presence of Ti, C, Au, O and B as the main elements in TiO2(G) NW@PAPBA-Au HJNH. XPS survey spectrum of TiO2(G) NW@PAPBA-Au HJNH (Figure S4) shows the binding energy (BE) values corresponding to six elements, namely Ti (459.4 eV), O (533 eV), C (285 eV), N (403 eV), B (192 eV) and Au (89 eV) (Figure S4). The clear existence of Ti in TiO2(G) NW@PAPBA-Au HJNH over PAPBA-Au NC supports our synthetic approach for the preparation of TiO2(G) NW@PAPBA-Au HJNH. On the other hand, XPS survey spectrum of PAPBA-Au NC showed the BE values of only the five elements, C, O, N, B and Au (Figure S4). 3.2.2. Electronic and Microstructural Properties We characterized the electronic, structural and crystalline properties of TiO2(G) NW@PAPBA-Au HJNH, through the results obtained from XPS, Raman, FTIR and XRD analyses. The electronic state of the Ti, N and Au in the TiO2(G) NW@PAPBA-Au HJNH was investigated by the deconvolution of core level XPS spectrum of the individual constituents. The high-resolution XPS spectra of Ti2p (Figure 2A), N1s and Au4f revealed that there could be chemical interactions or charge transfer processes between them in the TiO2(G) NW@PAPBA-Au HJNH. As shown in Figure 2A, the doublet peaks of Ti2p, with BE values of 459.45 eV and 465.05 eV (Ti2p1/2), are assigned to Ti2p3/2 and Ti2p1/2 electronic state of Ti4+ and the energy separation value between them is 5.60 eV. Through our previous report, we identified an energy level split value of 5.85 eV for TiO2(G).15 The increase in Ti2p splitting energy value for TiO2(G) NW@PAPBA-Au HJNH as compared to TiO2(G) suggests that the Ti in the TiO2(G) NW@PAPBA-Au HJNH can have chemical interactions or contribute to the energy transfer process with the other components. The N1s 13 ACS Paragon Plus Environment


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

deconvoluted XPS spectrum of PAPBA-Au NC (Figure 2B(a)) reveals the presence of two BE peaks centered at 400.0 eV and 401.45 eV and a weak peak at 398.6 eV that correspond to the –NH–, amine nitrogen of protonated nitrogen (N+) and quinoid imine moiety (=N–), respectively. The N1s peak of –NH– has a relatively lower intensity as compared to the other two N (N+ and =N–) moieties. Thus, XPS N1s results suggest that PAPBA in PAPBA-Au NC exists in the conducting state with predominant quinoid structures in the polymer backbone. The BE positions of N moieties in TiO2(G) NW@PAPBA-Au HJNH showed shifts as compared to the BE values in N1s in XPS of PAPBA-Au NC. The shifts in N1s BE state for TiO2(G) NW@PAPBA-Au HJNH (Figure 2B(b)) as compared to PAPBA-Au NC inform that the surface groups in TiO2(G) interact with PAPBA through the amine or imine containing nitrogen atoms. Also, the BE peak corresponding to N1s of =N– showed lesser peak intensity as compared to the peak intensities of other N1s states. The BE of the two individual Au4f peaks are located at 84.5 eV (Au4f7/2) and 88.15 eV (Au4f5/2), respectively, for the TiO2(G) NW@PAPBA-Au HJNH (Figure 2C). These doublet peaks suggest the existence of metallic state Au in the TiO2(G) NW@PAPBA-Au HJNH. To note, the BE values of Au4f7/2 and Au4f5/2 in TiO2(G) NW@PAPBA-Au HJNH are shifted to higher values as compared to metallic Au (BE of metallic Au is 84.0 eV (Au4f7/2) and 87.8 eV (Au4f5/2)).50 The shifts in BE of Au4f suggests the possible interactions of Au with the other components. Hence, the XPS results support the formation of a layer of PAPBA-Au NC over the surface of TiO2(G) NW and existence of chemical interactions between the TiO2(G) NW, and PAPBA as well with Au. Raman spectrum of TiO2(G) NW@PAPBA-Au HJNH (Figure 3A) exhibits characteristic bands associated with the vibrations of TiO2 and PAPBA. Typically, Figure 3A shows specific vibrations at 436 cm-1 and 631 cm-1 that correspond to B1g + A1g and Eg modes of anatase crystalline form of TiO2 and bands at 1336 cm-1 and 1576.7 cm-1 that correspond 14 ACS Paragon Plus Environment

Page 14 of 58

Page 15 of 58

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 C=C and C–C stretching vibrations of quinoid and semi quinoid rings of delocalized – N=C groups. Thus, Raman spectrum of TiO2(G) NW@PAPBA-Au HJNH (Figure 3A) confirms the existence of TiO2 and PAPBA. Raman peaks that could be identified for the two photon double resonance (D band) and in plane vibration of sp2 carbon band (G band) are not be seen in the Raman spectrum of TiO2(G) NW@PAPBA-Au HJNH, because of

the

overlapping of those carbon vibration bands with the quinoid and semi quinoid vibrational bands of PAPBA.15 We compared the Raman spectrum of TiO2(G) NW@PAPBA-Au HJNH (Figure 3A) and PAPBA-Au NC (Figure S5B) to investigate the probable interactions, if any, between TiO2(G) and PAPBA-Au NC. Raman spectrum of PAPBA-Au NC exhibited the characteristic semi quinoid (C~C) vibration band at 1350 cm-1. Two differences could be identified in the Raman spectral bands of quinoid and semi quinoid vibrations between TiO2(G) NW@PAPBA-Au HJNH (Figure 3A) and PAPBA-Au NC (Figure S5B). Firstly, the peak position of semi quinoid vibrational band of TiO2(G) NW@PAPBA-Au HJNH (1336 cm-1) is shifted by 14 cm-1 from the vibrational band of PAPBA-Au NC. Secondly, the peak intensity of semi quinoid vibrational band in TiO2(G) NW@PAPBA-Au HJNH is significantly diminished as compared the peak intensity in the case of PAPBA-Au NC. These two factors in conjunction confirm that there can be possible chemical interactions between TiO2(G) and PAPBA in the TiO2(G) NW@PAPBA-Au HJNH. Further, the comparison of Raman spectrum of TiO2(G) NW@PAPBA-Au HJNH (Figure 3A) with TiO2 NW@PAPBAAu NC (Figure S5A), informs there is a clear shift in the peak position of Eg vibration mode of TiO2 between them. This confirms that inclusion of G influence the microcrystalline state of anatase TiO2. In all, the result obtained from Raman spectrum of TiO2(G) NW@PAPBAAu HJNH (Figure 3A), TiO2 NW@PAPBA-Au NC and PAPBA-Au NC (Figure S5A), suggest that there could be heterojunctions between TiO2 and G as well between TiO2(G) and PAPBA-Au NC.



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

FTIR spectroscopy was used to infer the molecular structure of PAPBA in TiO2(G) NW@PAPBA-Au HJNH and PAPBA-Au NC and to explore the molecular level interactions between TiO2(G) NW and PAPBA-Au NC. For comparative purposes, Figure S5B presents the Raman and FTIR spectra of TiO2 NW@PAPBA-Au NC and PAPBA-Au NC. We compared FTIR spectral characteristics of TiO2(G) NW@PAPBA-Au HJNH (Figure 3B) with PABA- Au NC (Figure S5B (b)). FTIR spectra of TiO2(G) NW@PAPBA-Au HJNH (Figure 3B) and PAPBA- Au NC (Figure S5B (b)) show characteristic features of quinoid (N=Q=N) and benzenoid (N–B–N) vibrations of PAPBA (where Q and B stand for quinone and benzene ring, respectively), but with shifts in their positions and changes in intensities. Typically, the N=Q=N and N–B–N vibrational bands are observed at 1650 cm-1 and 1510 cm1

for the TiO2(G) NW@PAPBA-Au HJNH (Figure 3B). FTIR spectrum of PAPBA-Au NC

(Figure S5B (b)) showed N=Q=N and N–B–N vibrational bands at 1640 cm-1 and 1520 cm-1, respectively. The interaction between TiO2(G) NW and PAPBA-Au NC, caused peak shifts possibly due to the changes in the proportion of Q=N=Q and N–B–N groups in TiO2(G) NW@PAPBA-Au HJNH. Besides, FTIR spectrum of TiO2(G) NW@PAPBA-Au HJNH shows bands corresponding to C–N stretching vibration (~1340 cm-1), the aromatic C–H out of plane determination vibrations (~1040 cm-1), C–H vibrations of 1,3,4 trisubstituted benzenoid ring vibrations (~570 cm-1) and B–O stretching vibration (~1340 cm-1), informing the co-existence of conducting emeraldine base type PANI structure and boronic acid groups. The peak observed at 1201 cm−1 for TiO2(G) NW@PAPBA-Au HJNH is assigned to Ti–O–C bonds formed due to impregnation of G into TiO2. The two weak bands around 550–700 cm−1 represent Ti–O stretching modes,51 and the broad bands around 3430 and 3250 cm–1 correspond to the OH group stretching of B–OH or the surface hydroxyl groups on TiO2. The microstructural details of TiO2(G) NW@PAPBA-Au HJNH are derived from the comparison of XRD patterns of TiO2(G) NW and PAPBA-Au NC (Figure 3C(a-c)). The 16 ACS Paragon Plus Environment

Page 16 of 58

Page 17 of 58

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


XRD pattern of PAPBA-Au NC exhibits a broad peak centered at 2θ = 20-25° that is ascribed to the periodically parrel chain of amorphous polymers of PAPBA. The sharp peak at 2θ = 44.5° corresponds to (200) Bragg reflection of Au. The Au reflection peaks at ~38° (100), ~64.5° (220) are less intense. Hence, XRD pattern of PAPBA-Au NC confirms the coexistence of amorphous PAPBA and crystalline Au. XRD pattern of TiO2(G) shows peaks corresponding to (110), (103), (104), (221), (116) and (112) reflection of anatase phase of TiO2. There are no obvious peaks of G in the XRD pattern of TiO2(G) implying that G sheets are homogeneously dispersed within TiO2 NW. XRD spectrum of TiO2(G) NW@PAPBA-Au HJNH shows reflections corresponding to anatase TiO2 (2θ = 27.54° (110) and 39.0° (004), PAPBA (2θ = 20-25° (weak and broad)) and Au (2θ = 44.3° (200), 64.4° (220) and 77.6° (311)) with obvious differences in the intensities of anatase TiO2 and Au reflection peaks (Figure 3C). Typically, the intensities of 2θ = 27.54° (100) and 39.0° (200) reflection peaks of anatase TiO2 are 6.03 and 1.82 times lesser intense as compared to the respective XRD reflections in TiO2(G) (Figure 3C). The decrease in intensities of the reflection peaks of TiO2 in the XRD pattern of TiO2(G) NW@PAPBA-Au HJNH corroborate with the presence of the well-dispersed layer of PAPBA-Au NC over the surface of TiO2(G) NW (FESEM, Figure 1B). 3.2.3. Optical Characteristics UV-visible diffuse reflectance spectra of TiO2(G) NW@PAPBA-Au HJNH, TiO2 NW@PAPBA-Au NC, and PAPBA-Au NC (Figure 4A(a-c)) were compared to determine the differences in the optical properties between them. Typically, the absorption spectrum of PAPBA-Au NC (Figure 4A(c)) exhibits a peak at 320 nm and two broad bands around 420– 430 and 600–610 nm. These spectral characteristics are ascribed to the π-π*, polaron-π*, and bipolaron transitions of the conductive form of PAPBA, respectively. The plasmonic



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 18 of 58

resonance peak of Au NPs, normally located around 520 nm, is suppressed by the strong absorption of PAPBA in the 400–600 nm region. The DRS spectrum of TiO2(G) NW shows few absorption peaks in the UV wavelength region (250–280 nm) that corresponds to the tetrahedrally co-ordinated Ti cations. However, PAPBA exhibited strong optical absorption in the visible wavelength (400–600 nm) region (curve c). The UV-vis DRS spectra of TiO2(G) NW@PAPBA-Au HJNH (Figure 4A(a)) and TiO2 NW@PAPBA-Au NC (Figure 4A(b)) showed strong absorptions in the 400–600 nm wavelength region. Otherwise, the visible light absorption is enhanced by the presence of PAPBA-Au NC layer over the TiO2(G) NW or TiO2 NW. We consider that this visible light absorption enhancement is due to the synergistic contribution from the SPR effect of Au NP and strong visible region absorption of PAPBA. We envisage PAPBA can sensitize TiO2 and extend the absorption of TiO2(G) NW@PAPBA-Au HJNH and TiO2 NW@PAPBA-Au NC into the visible range.52 The optical bandgap of any photocatalyst can be estimated by 53 (αhυ)2 = Bd(hυ – Eg)

(Eq. 1)

where α is the absorption efficiency, hυ is the incident photon energy, Eg is the optical energy bandgap, and Bd is the absorption constant. The reflectance data (Figure 4A) of the three materials were translated into optical bandgap values using plots of (αhυ)2 vs. hυ (Figure 4B). The estimated optical bandgaps for TiO2(G) NW@PAPBA-Au HJNH (curve a) and TiO2 NW@PAPBA-Au NC (curve b) were 3.15 and 3.30 eV, corresponding to absorption edges of 375 and 394 nm, respectively. While both TiO2(G) NW@PAPBA-Au HJNH and TiO2 NW@PAPBA-Au NC could be effectively used as photocatalysts in the visible region, TiO2(G) NW@PAPBA-Au HJNH possesses additional advantages due to the e–-accepting capability of G, which is needed for efficient h+/e– pair separation. Thus, the existence PAPBA-Au NC on the surface of TiO2(G) NW or TiO2 NW enhances the visible light


Page 19 of 58

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


absorptions. As a result, TiO2(G) NW@PAPBA-Au HJNH can contribute to the visible lightinduced generation of h+/e– pairs, which is beneficial for the fabrication of visible light PEC sensors. 3.2.4. Thermal Properties The thermal characterization results of TiO2(G) NW@PAPBA-Au HJNH (Figure S5(a and a’)), TiO2 NW@PAPBA-Au NC (Figure S6(b and b’)), and PAPBA-Au NC (Figure S6(c and c’)) inform that TiO2(G) NW@PAPBA-Au HJNH comprises a thin layer of PAPBA (with a relatively low content of 21.7 wt.%) on the surface of the TiO2(G) NW. 3.2.5. Electrochemical Properties The electrochemical characteristics, such as kinetics of e– transfer reactions and electrochemical interfacial properties, were investigated. Figure 5A(a-c) shows the CVs of TiO2(G) NW@PAPBA-Au HJNH, TiO2 NW@PAPBA-Au NC, and PAPBA-Au NC recorded in a 0.1 M KCl solution containing 5 mM K3Fe(CN)6/K4Fe(CN)6 (each) at a scan rate of 50 mV s–1. All electrodes exhibited the reversible oxidation-reduction features of K3Fe(CN)6/K4Fe(CN)6, suggesting their good electroactivity. However, the oxidation peak current of the TiO2(G) NW@PAPBA-Au HJNH electrode was 56 µA/cm2, about 1.5 and 3.6 times higher than those of TiO2 NW@PAPBA-Au NC and PAPBA-Au NC electrodes, respectively. The larger peak current of TiO2(G) NW@PAPBA-Au HJNH suggested that the e– transfer process is greatly facilitated at the electrode. Moreover, the highest peak current observed for the TiO2(G) NW@PAPBA-Au HJNH electrode among the three electrodes was ascribed to the additional electrical conductivity provided by the embedded G. EIS is a powerful tool for probing the surface electrochemical properties at the semiconductorelectrolyte interface. Figure 5B(a-c) shows typical Nyquist plots of the fabricated electrodes obtained in 0.1 M KCl containing 5 mM K3Fe(CN)6/K4Fe(CN)6. Using the semicircle 19 ACS Paragon Plus Environment


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

diameter portion at high frequencies in the Nyquist plots, we estimated the charge transfer resistance (Rct) values like 5800, 7350, and 7850 Ω for TiO2(G) NW@PAPBA-Au HJNH, TiO2 NW@PAPBA-Au NC, and PAPBA-Au NC, respectively. The inclusion of G in TiO2 NWs contributed to the increase in conductivity. Based on the electrochemical characteristics, we concluded that the TiO2(G) NW@PAPBA-Au HJNH electrode is superior for the construction of an electrochemical/PEC glucose sensor. To understand the electrode kinetics of e– transfer, the corresponding CV curves were recorded in 5 mM K3Fe(CN)6/K4Fe(CN)6 (each) solution at different scan rates (υ), from 20 to 500 mV s–1 (Figure 5A-C). The results showed that the K3Fe(CN)6/K4Fe(CN)6 e– transfer reaction at the modified electrodes is a quasi-reversible surface-confined process. For all three electrodes, the separation between the anodic and cathodic peak positions increased with increasing υ, and the anodic/cathodic peak positions themselves were shifted. Both the anodic (Ipa) and cathodic peak currents (Ipc) steadily increased with increasing υ. The electroactive electrode surface area (A) was determined from the plots of peak currents vs. υ1/2 (Figure 6D) and was found to be 5.816 ± 0.05, 4.692 ± 0.10, and 1.808 ± 0.054 cm2 for the TiO2(G) NW@PAPBA-Au HJNH, TiO2 NW@PAPBA-Au NC, and PAPBA-Au NC electrodes, respectively. These results suggested that TiO2(G) NW@PAPBA-Au HJNH possess the highest A compared to other electrodes, signifying that the G embedded within TiO2 NWs contributed to the increased surface area. After determining the electrochemical parameters, we investigated the PEC characteristics of TiO2(G) NW@PAPBA-Au HJNH and other electrodes (for comparison). 3.3. PEC/Electrochemical Dual Mode Sensing The morphological and XPS characterization results of TiO2(G) NW@PAPBA-Au HJNH (Figure 1-2) revealed the presence of a layer PAPBA-Au NC on TiO2(G) NW, and


Page 20 of 58

Page 21 of 58

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


there are chemical interactions between TiO2(G) NW and PAPBA-Au NC. As a consequence of the presence of PAPBA-Au NC on TiO2(G) NW, metal (Ti) – semiconductor (PAPBA) and metal (Ti) – metal (Au) heterojunctions are established and caused enhancement of visible light absorption, photocatalytic and electrocatalytic properties. We have specifically selected PAPBA for glucose sensing via boronate-glucose complexation.54 The synergistic contribution of visible-light-driven photogeneration of charge carriers h+/e– by TiO2(G) NWs, the glucose binding ability of boronate-containing PAPBA, the SPR effect of Au NPs, and the electroconductivity of PAPBA provided the basis for PEC and electrochemical sensing of glucose and HbA1c, respectively. Considering the suitability of TiO2(G) NW@PAPBA-Au HJNH for photocatalytic and electrocatalytic applications, we explored the utilization of TiO2(G) NW@PAPBA-Au HJNH in enzyme-free PEC glucose sensors and electrochemical HbA1c sensors in two independent signal transduction modes. 3.3.1. Capability and Mechanism We envisage a probable mechanism for the photoexcitation and e– transfer of TiO2(G) NW@PAPBA-Au HJNH under visible light irradiation of TiO2(G) NW@PAPBA-Au HJNH (Scheme 2). It is important to note that the generation of h+/e– pair under visible light radiation is not feasible for TiO2(G) NW because of its wider band gap (~3.20 eV). However, TiO2(G) NW@PAPBA-Au HJNH can exhibit visible light photocatalytic activity due to the photon absorptions by the SPR of Au NP and visible light absorption by PAPBA.55,56 As a result, an enhanced photocatalytic activity is expected for TiO2(G) NW@PAPBA-Au HJNH under visible light irradiation. When TiO2(G) NW@PAPBA-Au HJNH is irradiated with visible light, the plasmonic Au NPs can interact with light through SPR effects.57 The conjugated PANI type PAPBA in TiO2(G) NW@PAPBA-Au HJNH has π-orbitals as the highest occupied molecular orbital (HOMO) and π* as the lowest unoccupied molecular orbital (LUMO). Under visible light irradiation, PAPBA also absorbs photons to induce π-π* 21 ACS Paragon Plus Environment


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

(HOMO-LUMO) transitions. Hence, PAPBA can absorb visible light and generate h+ - e– pair and shuttle the e– to the CB of Au and subsequently CB of TiO2 (Scheme 2). The LUMO of PAPBA is energetically higher than the CB edge of anatase (3.20 eV) and Au NPs. The plasmonic Au NPs function as the e– reservoir and extend the life time of the photogenerated charge carriers. The plasmonic Au NPs in TiO2(G) NW@PAPBA-Au HJNH absorbs visible light and shuttle e–via ‘hot electron’ mechanism.57 Thus, multiple roles for the Au NPs in TiO2(G) NW@PAPBA-Au HJNH such as a sensitizer, visible light photon absorber and reservoir for e– and interface to transfer the excited higher energy electron (hot electron) to TiO2(G) NW (Scheme 2) are expected.58 The photogenerated e-, from both Au NP and PAPBA, can be shuttled to the to the electrode via the photocatalyst (TiO2) and e– accepting G. The scavenged e– are transferred to oxygen molecules and yield highly reactive oxidative species, superoxide radicals (O2•‒) and hydrogen radicals (OH•).59 Further, the PANI matrix in PAPBA is a good h+ transporting material.5 Hence, the photogenerated h+ left in both Au NPs and PAPBA can be transferred to the surface of the photocatalyst (TiO2(G) NW@PAPBA-Au HJNH), react with any of the analyte (x) and oxidize, rather than to produce OH•. The more positive potential of OH•/OH– process than both the HOMO energy level of PANI type π conjugation in PAPBA and Fermi level of TiO2.60 favours the oxidation of the analyte. The synergistic effects of the SPR phenomenon of Au NPs, visible light absorbing and h+ transporting capability of PAPBA, e– accepting/transferring properties of G and the band alignment of TiO2, cause a decrease in recombination rate of the photogenerated h+ and e–. As a result, a large number of h+ and e– are separated. This special feature of TiO2(G) NW@PAPBA-Au HJNH can result in enhanced photocatalytic activities. Generally, for the development of a selective glucose PEC sensor, enzymes, such as glucose oxidase (GOx), are immobilized in the biosensing matrix to provide high selectivity for the glucose molecules. However, the enzyme (such as GOx) immobilized in a glucose 22 ACS Paragon Plus Environment

Page 22 of 58

Page 23 of 58

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


biosensor has several drawbacks that include poor reproducibility, stability problems, long term activity decay and operation temperature. In recent years, the sensitivity of nonenzymatic glucose sensors has enhanced significantly21,61 using nanostructured materials or affinity polymers such as boronic acid containing polymers. Among the many possible nonenzymatic glucose sensor designs, much attention has been devoted to a boronic acid group containing polymers as affinity polymers for glucose detection. The ability of boronic acids to bind with 1,2-diols such as glucose and undergo ionization transitions or complexation make boronic acid containing polymers as the ideal materials for diabetes related sensor applications.54,62 Real time molecular recognition of glucose has been demonstrated based on the ability boronic acid groups in polymers such as PAPBA to form reversible complexation with glucose.63 We introduced PAPBA within TiO2(G) NW@PAPBA-Au HJNH to import selectivity for glucose in PEC detection. The PEC interactions between TiO2(G) NW@PAPBA-Au HJNH and glucose, without any auxiliary enzyme, is due to the strong photooxidation of glucose molecules that are bound to the sensor probe because of the affinity binding property of PAPBA. At an appropriate applied potential differences and under visible light irradiation, the photogenerated h+ react with glucose. + 2ℎ → +

(Eq. 2)

3.3.2. Strategy and Mechanism of Electrochemical Sensing of HbA1c The TiO2(G) NW@PAPBA-Au HJNH was utilized for the selective amperometric electrochemical HbA1c sensor utilizing the HbA1c chemical binding ability of the TiO2(G) NW@PAPBA-Au HJNH as the sensor probe via interactions between the glucose moieties in HbA1c and PAPBA) and the catalytic activity of the heme groups in HbA1c for the reduction of H2O2 (Scheme 3). The hemoglobin (Hb) part of HbA1c contains four iron groups that effectively catalyze the reduction of H2O2. Particularly, Hb exhibits peroxidase (enzyme) like



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

activity towards H2O2 and hence be used to develop non-enzyme based H2O2 sensor. HbA1c can specifically bind to the sensing probe via glycolate part of HbA1c and boronic acid part of PAPBA (Scheme 3). The Hb in the bound HbA1c catalyzes the reduction of H2O2 at an applied potential. The H2O2 reduction current can be correlated to the concentrations of bound Hb to the sensor probe and in turn detections the level of HbA1c. The principle of electrocatalysis of H2O2 by the bound HbA1c to the sensor probe (TiO2(G) NW@PAPBA-Au HJNH) is explained by the following equations: 1 … . − @ !" # + $ %&&'()* &+,)-,(.'

/0000000000001 "2 333 # … . − @ !" # (Eq. 3) "2 333 # … . − @ !" # + 2 + !

/000000000001 2 "2 333 # + −@ !" # (Eq. 4) The overall electrocatalytic reaction would be 45$6(7 8"9#

! + 2 $ +2 /000000000001 2 ! (Eq. 5) The presence of PANI type π-conjugated structural in PAPBA and Au NPs in TiO2(G) NW@PAPBA-Au HJNH facilitate the e– transfer to the electrode.64,65 3.3.3. PEC Glucose Sensor Performances 3.3.3.1. Optimization of Experimental Conditions To acquire an optimal analytical performances and photocurrent values, few possible factors that influence the photoresponses should be studied. We performed the experiments to optimize experimental parameters such as applied bias potential (Figure S7) and wavelength of light (Figure S8). The sensor electrode showed deplorably low photocurrent levels when irradiated with a wavelength of 254 nm over a wide applied bias potentials. However, the 24 ACS Paragon Plus Environment

Page 24 of 58

Page 25 of 58

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


sensor electrode showed a significantly enhanced photoactivity of TiO2(G) NW@PAPBA-Au HJNH with a wavelength 365 nm. This result corroborate with the extended visible light absorption of TiO2(G) NW@PAPBA-Au HJNH as compared to TiO2(G). Apart from the optimization of the wavelength of light source, the photocurrent was maximum at an applied potential of +0.48 V. Based on our results; we concluded that the optimized experimental conditions for PEC sensor experiments: applied potential = +0.48 V and irradiation wavelength = 365 nm. We compared the PEC performances of TiO2(G) NW@PAPBA-Au HJNH, TiO2 NW@PAPBA-Au NC,

and PAPBA-Au NC in 0.1 M PBS (pH 7.0) using cyclic

voltammetry (Figure 7), linear sweep voltammetry (Figure 8), and chronoamperometry (Figure 9). Results obtained under illumination of the fabricated electrodes with intermittently visible light irradiation (“ON”) and in the dark (“OFF’) in the presence/absence of 3 mM glucose are compared. Figure 7A(a-c) presents the CV curves of the three electrodes recorded under “OFF” conditions in 0.1 M PBS in the absence of glucose. Two conclusions were drawn from the CV data. First, the three electrodes did not show any prominent redox characteristics in the potential region from –0.50 to +0.80 V. The CV patterns were almost identical or overlapped with each other under “ON” and “OFF” conditions. Figure 7B(a-c) depicts the CVs recorded in the presence of glucose under “ON” (Figure 7B(a-c)) and “OFF” (Figure 7B(a’-c’), inset) conditions. The TiO2(G) NW@PAPBA-Au HJNH electrode exhibited a prominent oxidation peak at +0.48 V (Figure 7B(a)) under “ON” conditions, ascribed to the photooxidation of glucose. However, the other two electrodes did not show any prominent peaks around +0.48 V (Figure 7B(b and c)). CVs of the TiO2(G) NW@PAPBA-Au HJNH electrode, recorded under “OFF’ conditions (Figure 7B, inset) in the presence of 3.0 mM glucose, also exhibited an oxidation peak at +0.48 V (Figure 7B, inset(a’)), albeit with a very low peak current, which is ascribed to the electrooxidation of 25 ACS Paragon Plus Environment


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

glucose at the electrode. The other two electrodes did not show any oxidation peaks in the presence of 3 mM glucose under either “ON” or “OFF” conditions. On comparison of the glucose oxidation peak currents, the “ON” photocurrent was found to be 2.77 times higher than “OFF” (electrochemical) current at +0.48 V. Thus; we concluded that the TiO2(G) NW@PAPBA-Au HJNH electrode is more suited for PEC sensing of glucose. We recorded the LSV of the three electrodes under light irradiation in the presence of 3 mM glucose at a low potential scan rate to understand the electrode kinetics (Figure 8). The LSV curves make it clear that the TiO2(G) NW@PAPBA-Au HJNH electrode showed a significant photocurrent rise above 0.30 V, as compared to the very slow PEC processes at the other two electrodes. For a comparison, the linear sweep voltammogram of G modified electrode in 0.1 M PBS, 0.1 M PBS containing 3 mM glucose (“OFF” condition) and 0.1 M PBS containing 3 mM glucose (“ON” condition) are presented (Figure S9). On comparing the current-potential features in the LSVs recorded for G modified electrode (Figure S9), we infer that G modified electrode did not show any electrooxidation peak for glucose in both “OFF” and “ON” conditions. A very weak photocurrent response was noticed at the G modified electrode for 3 mM glucose under ‘ON’ condition. Considering the work function of G as -4.5 eV, we believe that G is weak excited under 365 nm irradiation. However, for the G-semiconductor composite photocatalysts, there seems to be a consensus about photoactivity enhancement of the respective semiconductor after coupling with G, primarily due to two important factors. The first factor is that the e– energy level on CB of semiconductors is much higher than that of G, which makes the transfer of photogenerated e– from the CB of the semiconductor to G as an energetically favorable process.66 The second factor is the outstanding e– conductivity of G which enables G to function as the 2D network of e– reservoir for accepting e– from the semiconductor. Due to these two factors, G included semiconductors exhibited enhanced PEC properties. Our results revealed that the photo e– transfer is faster for TiO2(G)


Page 26 of 58

Page 27 of 58

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


NW@PAPBA-Au HJNH under light irradiation. Thus, the photooxidation of glucose predominantly occurred at TiO2(G) NW@PAPBA-Au HJNH. We recorded the transient photocurrent (j(photo)-time) responses to 3.0 mM glucose at an applied bias potential of +0.48 V under intermittent light switch “ON–OFF” conditions. Figure 9A(a-c) shows the photocurrent responses to the light “ON–OFF” switches for TiO2(G) NW@PAPBA-Au HJNH, TiO2 NW@PAPBA-Au NC, and PAPBA-Au NC electrodes in the presence of 3 mM glucose (0.1 M PBS, pH 7.0). For the TiO2(G) NW@PAPBA-Au HJNH electrode under the “ON” conditions, the photocurrents reached a steady value within three seconds. Moreover, the photocurrent density of TiO2(G) NW@PAPBA-Au HJNH (Figure 9(a)) is 23 and 203 times higher than that of TiO2 NW@PAPBA-Au NC (Figure 9A(b)) and PAPBA-Au NC (Figure 9A(c)) electrodes, respectively. The latter (Figure 9A(c)) electrode showed a very low current, which is ascribed to the specific binding of glucose to the boronic groups of PAPBA and the subsequent PEC processes due to the current-generated catalytic and plasmonic effect of Au NPs. The photocurrent generated at PAPBA-Au NC is attributed to the following reasons. First, PAPBA features a PANI like backbone structure and is an organic semiconductor whose absorption is extended into the visible region. Second, Au NPs enhance light absorption in the visible region due to plasmon resonance and provide a conductive pathway for e– transfer. However, the higher photocurrents at TiO2(G) NW@PAPBA-Au HJNH and TiO2 NW@PAPBA-Au NC revealed that the inclusion of TiO2(G) and TiO2 augmented the photocurrents. Specifically, the photocurrent increment of TiO2(G) NW@PAPBA-Au HJNH (27.861 µA, 3 mM glucose) was ~ 63 times higher than that of the TiO2 NW@PAPBA-Au NC (0.441 µA, 3 mM glucose) electrode. The highest photocurrent witnessed for TiO2(G) NW@PAPBA-Au HJNH is due to the synergistic effect of PAPBA, TiO2, and G towards the enhancement of visible light absorption, photocurrent generation, and e– transport. The components of TiO2(G) NW@PAPBA-Au HJNH exhibit 27 ACS Paragon Plus Environment


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

individual roles such as the good electric conductivity and e– accepting capability of G, which facilitated the transport of photoexcited e– from TiO2 to the ITO electrode and the oxidation of glucose by the photogenerated h+. The stability of photocurrent generation was investigated by performing repeated light “ON” (Figure 9B(a-c)) and “OFF” (Figure 9B (a’c’)) cycles over a period of 150 min. A small decrease in the photocurrent of TiO2(G) NW@PAPBA-Au HJNH (Figure 9B(a)) was observed after 150 min. Specifically, the latter electrode retained ~99.3% of the original photocurrent values after 150 min (Figure 9B(a)). On the contrary, the photocurrents dropped to ~18 and 7.40% for TiO2 NW@PAPBA-Au NC (Figure 9B(b)) and PAPBA-Au NC (Figure 9B(c)) after 150 min. Thus, we demonstrated that TiO2(G) NW@PAPBA-Au HJNH exhibit stable “ON” (Figure 9B(a)) and “OFF” (Figure 9B(a’)) cycling over a longer period than other electrodes. The excellent photocurrent generation in the presence of glucose for TiO2(G) NW@PAPBA-Au HJNH is by the mechanism proposed in Scheme 2. The immersion of the TiO2(G) NW@PAPBA-Au, HJNH electrode into a solution of glucose, caused the chemical binding of the boronic acid groups in PAPBA with glucose. Glucose binding by complexation with boronates has been well studied.54 The photocurrent was generated at TiO2(G) NW@PAPBA-Au HJNH in the absence of the generally used enzyme (glucose oxidase) due to the synergistic glucose binding ability of the boronic groups of PAPBA and the photooxidation capability of TiO2(G) NWs and Au NPs. Thus, under light irradiation at a positive applied potential (+0.48 V), the PEC process at the TiO2(G) NW@PAPBA-Au HJNH electrode plays an important role in the significant enhancement of the photo amperometric response to glucose (Eq. 2). 3.3.3.2. Photoamperometry and Sensor Calibration


Page 28 of 58

Page 29 of 58

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


We presented more straightforward glucose sensing characteristics by the timedependent “ON” photocurrent measurements at an applied bias potential of 0.48 V. Figure 10(a)) presents the photocurrent vs. time (j–t) responses of the TiO2(G) NW@PAPBA-Au HJNH electrode for successive addition of glucose (0.5 mM to 28 mM) to 0.1 M PBS. Each addition caused a well-defined amperometric response, and the current reached a maximum value within 3 s. The photocurrent was steady after 3 s with glucose concentration in the range from 2 to 15 mM. However, the photocurrent reached a maximum by 3 s and dropped steadily after beyond 3 s, when the glucose concentration was more than 15 mM. In our sensor design, the boronic acid – glucose (diol) interaction is considered important. Boronic acid groups are weak Lewis bases and transform into tetrahedral boronate ester (boronic acid - glucose complex) upon interaction with glucose. The boronic acid – glucose complexation process is accompanied by the release of protons. The glucose binding process and the stability of gluconate ester are influenced by the number of factors such as pH, dihedral angle of the diol etc. We presume that higher concentration of glucose (>15 mM) leads to modification in pH (due to the release of protons) results in the decrease of the stability of boronic acid – glucose complex. This may be the reason for the decrease in peak current at higher glucose concentration after reaching the maximum current. For comparison, amperometric experiments were also performed for the successive addition of glucose at an applied potential +0.48 V under “OFF” conditions. The TiO2(G) NW@PAPBA-Au HJNH electrode showed much smaller currents upon each glucose addition (Figure 10(a’)) under the “OFF” condition, with a small current increase after successive additions of glucose (up to 15 mM). Figure 10(inset) shows the calibration plots of photocurrent vs. time for various glucose concentrations under light “ON” and “OFF” conditions. The calibration plots (Figure 10(inset)) show that the photocurrents were linear concerning glucose concentrations over a wider range, 0.5 to 28 mM, under “ON” conditions. The TiO2(G) NW@PAPBA-Au HJNH



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

electrode exhibited a relatively narrow linear concentration range (from 0.5 to 15 mM) with lower sensitivity under the “OFF”

conditions as compared to the “ON” ones. The

photocurrent responses of TiO2 NW@PAPBA-Au NC (Figure 11(a, a’)) and PAPBA-Au NC (Figure 11(b, b’)) were also determined under conditions similar to those used for TiO2(G) NW@PAPBA-Au HJNH. Typically, the photocurrent witnessed at TiO2(G) NW@PAPBAAu HJNH for 5 mM (49.93 µA, Figure 10(a)) was much higher than the photocurrent noticed at TiO2 NW@PAPBA-Au NC (3.1 µA, Figure 11(a)) and PAPBA-Au NC (0.22 µA, Figure 11(b)). This trend was also evident for wider glucose concentrations. The sensitivity of glucose sensing was deplorably low for TiO2 NW@PAPBA-Au NC. The PAPBA-Au NC (Figure 11(b’)) electrode showed marginal current changes or poor current responses to glucose concentration, either in the “ON” (Figure 11b) or “OFF” (Figure 11b’) conditions, with very low sensitivity. On close perusal of the calibration plot for TiO2(G) NW@PAPBAAu HJNH (Figure 10(inset)), the photocurrents were proportional to the concentration of glucose in two linear ranges, 0.5–20 mM and 20–28 mM, with correlation coefficients of 0.991 and 0.992, respectively. The detection limit of glucose accomplished on the TiO2(G) NW@PAPBA-Au HJNH electrode, was 0.11 mM (at a signal to noise (S/N) ratio of 3). Glucose sensors can be designed and fabricated for a vast range of applications such as diabetes management, bioprocessing, beverage industry and environmental monitoring. The limit of glucose detection for the designed sensor is dependent on the targeted application. We focused our attention on fabricating a glucose sensor for diabetes glucose monitoring. The physiological blood glucose concentration of non-diabetes is usually in the range 4-8 mM for diabetes; glucose concentration range is much wider like 0.5-28 mM. However, sensing platforms described in large part of scientific publications have little or no relevance for practical Diabate monitoring applications as their linear range is often outside the blood glucose range of diabetes. Further, the new ISO 15197:2013 standard (as applicable from


Page 30 of 58

Page 31 of 58

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


May 2016 onwards) requires the accuracy of blood glucose test strips within ±20% of laboratory results for glucose concentrations above 5.6 mM and within ±0.83% mM for lower concentrations (

New Titanium Dioxide-Based Heterojunction Nanohybrid for Highly

Recommend Documents