Introduction

Hemoglobin (Hb) is a tetrameric heme protein comprised of two α chains and two β chains [1]. Hb can be used as an ideal model molecule for the research of the redox of proteins and biological sensing due to its relatively stable structure. However, it is difficult to obtain effective electrochemical signals and study the electrochemical behavior of Hb at the bare electrode surface due to its relatively large spatial structure and electroactive center is hard to expose. In addition, the adsorption of Hb on the bare electrode may lead to the passivation of electrode and the restriction of electron transfer rates [2]. Therefore, researchers have tried to modify electrodes using different kinds of nanomaterials such as graphene [3], quantum dots [4], metal nanoparticles [5], etc. to achieve its direct electrochemistry.

Among various shapes of gold nanoparticles (GNPs), gold nanorods (GNRs) attracted substantial attention [6, 7]. In addition to the unique optical properties [8], GNRs possess general properties similar to GNPs such as better conductivity, and have several advantages over the spherical GNPs, including fast electron transfer rate and high surface area [9, 10], which inspire researchers to explore the application of GNRs in electrochemical biosensors [11, 12]. Nevertheless, GNRs are prone to agglomeration which causes a loss of its own features. Therefore, the core-shell Au@Ag nanorods (Ag@GNRs) have attracted much attention due to their tunability in nanometer-scale structure and tunable longitudinal SPR peaks as well as chemical stabilities [13, 14] Combined with the superior catalytic activity of Au and Ag, Ag@GNRs have been synthesized and utilized in dopamine detection [15], genes analysis [16] and electrochemical immunoassay [17].

Trichloroacetic acid (TCA) is a class of environmental contaminants which has certain carcinogenic and caused serious impact on human life [18]. It can be found not only in the industrial waste water but also in drinking water as disinfection by-product [19]. Therefore, it is of great significance for developing a simple, rapid, selective and sensitive method to detect TCA. There are many methods for the detection of TCA such as gas chromatography [20], high performance liquid chromatography [21] and mass spectrometry [22]. These methods with adequate sensitivity are difficult for in-situ or online monitoring and suffered from expensive equipments, time-consuming for procedures of derivation and extraction [23]. On the contrary, electrochemical sensor shows its advantages over these drawbacks owning to their intrinsic advantages such as simple operation, high sensitivity, selectivity, low cost, and good portability [24, 25]. Meanwhile, bioelectrochemically reductive TCA by Hb has been considered as a promising way for the TCA determination because it exhibits a relatively high current efficiency. Since electron transfer occurs between Hb and TCA, the activity of active centers in Hb immobilized on the electrode surface plays the key factor during the reaction process. Different kinds of Hb modified electrodes have been fabricated to investigate the electrocatalytic effect toward TCA [26–29].

In this work, Ag@GNRs have been prepared and characterized by UV-vis spectroscopy, transmission electron microscopy. Then it was used to immobilize Hb on the modified electrode surface by self-assembly technique. The synthesized Ag@GNRs can provide large surface area, superior conductivity and favorable microenvironment for retaining the biological activity of the adsorbed Hb and facilitate the direct electron transfer between the Hb and the electrode. The electrocatalytic reduction of TCA has been investigated, showing possible potential application of Ag@GNRs for the construction of biosensors.

Experimental

Reagents and materials

Bovine Hb (Hb ~ 90%, MW = 66,000), poly (diallyldimethylammonium chloride) (PDDA, 20% (m:m)), polystyrene sulfonate (PSS, MW = 70,000), sodium citrate and trichloroacetic acid (TCA) were purchased from Sigma-Aldrich (http://www.sigmaaldrich.com/). Chloroauric acid (AuCl3·HCl·4H2O) and H2O2 (30% (m:m)) were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China, http://www.sinoreagent.com/). Cetyltrimethyl ammonium bromide (CTAB), silver nitrate (AgNO3), sodium borohydride (NaBH4), ascorbic acid (AA) and sodium oleate (NaOL) were all purchased from Nantong Chengji Chemical Trade Co., Ltd. (Nantong, China, http://cjhgmy.chinapyp.com/). Solutions of H2O2 were prepared by dilution of H2O2 appropriately. Phosphate buffer (0.1 M) with different pH values were prepared by mixing stock solutions of Na2HPO4 and NaH2PO4 and adjusting the pH value with 0.1 M H3PO4 or NaOH solution. All other chemicals used were of analytical reagent grade. All the solutions were prepared using double distilled water.

Apparatus and methods

All electrochemical experiments were carried out on a CHI660C electrochemical workstation employing a conventional three-electrode cell. The working electrode was the modified glassy carbon electrode (GCE) (3 mm in diameter). A saturated calomel electrode (SCE) and a platinum wire electrode served as the reference electrode and the counter electrode, respectively. All electrochemical measurements were carried out under a high-purified nitrogen atmosphere. The electrolyte was 0.1 M phosphate buffer which must be purged with nitrogen for at least 20 min. The GNRs and Ag@GNRs were characterized by transmission electron microscopy (TEM) (Tecnai-12, Netherlands). UV–vis spectrum was recorded on a UV–2450 spectrophotometer (Shimadzu, Japan). Electrochemical impedance spectroscopy (EIS) was performed with an AUTOLAB PGSTAT 302 N electrochemical workstation (Metrohm, Swiss) in 0.1 M KNO3 solution containing 1.0 mM Fe(CN)6 3−/Fe(CN)6 4−(1:1) across the frequency range from 10−3 to 105 Hz.

Preparation of Ag nanoparticles (AgNPs), GNRs and Ag@GNRs

GNRs were chemically synthesized using seed mediated growth method [30]. Seed solution was prepared by firstly mixing 5 mL HAuCl4 (0.5 mM) with 5 mL CTAB (0.2 M) and then adding 0.6 mL ice-cold NaBH4 (0.01 M) under vigorous stirring. After 2 min of reaction, the prepared seeds were stored at room temperature for 30 min before use.

For the preparation of growth solution [6, 31], CTAB (9.0 g) and NaOL (1.234 g) were added to an Erlenmeyer flask and dissolved in doubly distilled water (250 mL) at 50 °C. AgNO3 (12 mL, 4 mM) and HAuCl4 (250 mL, 1 mM) were added to this solution when it was cooled to 30 °C. After 15 min, the mixture was stirred at 700 rpm for 90 min until it gradually became colorless and the pH was adjusted by adding 2.1 mL of 12.1 M HCl (37% (m:m)). After stirring gently at 400 rpm for 15 min, AA (1.25 mL, 0.064 M) was added to the colorless solution under stirring for 30 s. Thereafter, 0.8 mL of the seed solution was quickly added to the above solution. The resultant mixture was stirred for 30 s and left undisturbed at 30 °C for 12 h. Finally, the excess CTAB in the solution was removed by centrifugation thrice at 7000 rpm (7669 rcf) for 30 min. The GNR-solution was obtained by diluting the sediment to 10 mL with doubly distilled water.

The Ag@GNR-nanocomposite was obtained by reduction of Ag+ on the surface of GNRs [32, 33]. AgNO3 (200 μL, 10 mM) was added into the prepared GNR-solution (5 mL). The mixture was mixed with AA (200 μL, 100 mM) and stirred for 10 min, after that, the reduction processes was finished and Ag@GNR-solution can be obtained.

For the preparation of AgNPs, a solution of 1 mM AgNO3 (100 mL) was added and heated with vigorous stirring. Upon boiling, 1 mL of 10 mg⋅mL−1 sodium citrate solution was added rapidly. Then the solution was heated under reflux for approximately 1 h. Before use, it was stored at 4 °C.

Preparation of the modified electrode

Prior to modification, the GCE was firstly polished with abrasive paper and then with alumina slurry to a mirror. After being thoroughly rinsed with doubly distilled water, the well-polished GCE was cleaned in absolute ethanol by sonication. The electrode was washed with doubly distilled water and then dried in air. The GCE was dipped in turn into PDDA solution (1% (m:m)), PSS solution (2.0 mg⋅mL−1) and Ag@GNR stock solution at 4 °C for 30 min, respectively. At last the electrode was dipped into 3.0 mg⋅mL−1 Hb solution (pH 9.0 phosphate buffer) at 4 °C for 30 min. The electrode is denoted as Hb-Ag@GNR-PSS-PDDA/GCE in abbreviation. For comparison, Hb-AgNP-PSS-PDDA/GCE and Hb-GNR-PSS-PDDA/GCE were fabricated with similar procedure.

Results and discussion

Choice of materials

With the development of nanotechnology, numerous novel nanomaterials especially noble metal nanomaterials have been extensively applied in the field of biosensor. Pure silver and gold nanoparticles, with the advantages of large specific surface area, excellent stability as well as the ability of modulating the electron transfer, have been utilized to realize the direct electron transfer of redox proteins and enzymes [34, 35]. Although the individual nanoparticles show unique shape- and size-dependent physicochemical properties, the shell-core structures, assembled from them, may give rise to some unexpected mechanical, electrical, and catalytic performances, which can be stimulated a variety of potential applications in photoelectricity, catalysis and biosensor [36, 37]. The preparation of bimetallic Ag@GNRs has attracted much attention due to their characteristic properties differ from those of the GNRs [38, 39]. The protective layer of Ag shell can not only prevent the GNRs from aggregation but also provide a better platform for further functionalization. Accordingly, considering the excellent catalytic activity of GNRs and Ag, Ag@GNRs were synthesized in this work and further applied as a function interface for Hb immobilization.

Characterization of GNRs and Ag@GNRs

TEM and UV–vis absorption spectra were used to characterize the synthesized GNRs and Ag@GNRs (Fig. 1). The average length and the aspect ratio of the GNRs are ~50 nm and ~5.1, respectively (Fig. 1a). The TEM image (Fig. 1b) reveals the morphology of Ag@GNRs, where each GNR is encircled distinctly with a layer of Ag halo. GNRs and Ag@GNRs were further confirmed by the UV–vis absorption spectra (Fig. 1c). The UV–vis absorption spectrum of GNRs exhibit two distinct absorption peaks at 509 nm and 960 nm, which are assigned to the typical transverse and longitudinal absorption peaks of GNRs, respectively. As the Ag is coated on the surface of GNRs, the longitudinal absorption peak blue shifts to 797 nm and the transverse absorption peak splits into two absorption peak bands which are the typical absorption spectrum bands of Ag@GNRs and located at 406 nm (a) and 476 nm (b), respectively. These results reveal that Ag has been wrapped on the surface of GNRs successfully.

Fig. 1
figure 1

TEM spectra of a GNRs and b Ag@GNRs. cHAA72175 UV–vis absorption spectra of GNRs and Ag@GNRs

Full size image

Characterization of the modified electrode

EIS is often used to monitor the assembly process. In EIS, the semicircle part at higher frequencies corresponds to the electron-transfer limited process. Its diameter equals the electron transfer resistance, R et, which exhibits the electron transfer kinetics of the redox probe at the electrode interface. Fig. S1 displays the EIS observed upon the changes of surface-modified process. The result discussed in Electronic Supplementary Material (ESM) shows that Hb has been successfully immobilized on the modified electrode.

Direct electrochemistry of Hb assembled on Ag@GNR-PSS-PDDA/GCE

Electrochemical behaviors of Hb on the modified electrodes were investigated by cyclic voltammetry (CV) with the results shown in Fig. 2a. No obvious redox peak can be observed at bare GCE (curve a), PDDA/GCE (curve b), PSS-PDDA/GCE (curve c) and Ag@GNR-PSS-PDDA/GCE (curve d), indicating no electroactive substances exist on the electrode surface. However, a couple of stable and well-defined redox peaks at −0.32 V and −0.40 V can be observed on Hb-Ag@GNR-PSS-PDDA/GCE (curve e), which is correspond with the characteristic of Hb-heme Fe(III)/Fe(II) reported by literature [40]. These results reveal that Ag@GNRs can provide a good microenvironment for Hb, so as to realize the direct electrochemical response and maintain the nature form of Hb.

Fig. 2
figure 2

a Cyclic voltammograms of (a) GCE, (b) PDDA/GCE, (c) PSS-PDDA/GCE, (d) Ag@GNR-PSS-PDDA/GCE and (e) Hb-Ag@GNR-PSS-PDDA/GCE in 0.1 M pH 7.0 phosphate buffer. Scan rate: 100 mV⋅s−1. b Cyclic voltammograms on Hb-Ag@GNR-PSS-PDDA/GCE in pH 7.0 phosphate buffer at scan rates of 50, 100, 150, 200, 250, 300, 350, 400, 500 and 600 mV⋅s−1 (from a to j). Inset: The linear relationship between the redox current (I p) and the scan rate (v)

Full size image

Fig. 2b displays a series of cyclic voltammograms of Hb immobilized on Ag@GNR-PSS-PDDA/GCE in pH 7.0 phosphate buffer at the different scan rates. With an increasing scan rate ranging from 50 mV⋅s−1 to 600 mV⋅s−1, the anodic and cathodic peak potentials of the Hb show a small shift and the redox peak currents increase linearly (inset in Fig. 2b), indicating a surface-controlled electrode process. According to Laviron eq. [41]:

$$ {I}_{\mathrm{p}}=\frac{n^2{F}^2 A\Gamma \nu}{4 RT}=\frac{ n FQ\nu}{4 RT} $$

Where Γ represents the surface coverage of Hb (mol⋅cm−2), Q is the total charge. I p, T, F and R have the usual physical meaning. From the slope of the I p∝v, n is calculated to be 1.22, which demonstrates that the redox reaction of Hb at Ag@GNR-PSS-PDDA/GCE is a one-electron transfer process.

When nΔE p ≤ 200 mV, electron transfer rate constant k s can be calculated by the following formula [42]:

$$ \log {k}_s=\alpha \log \left(1-\alpha \right)+\left(1-\alpha \right) \log \alpha - \log \frac{RT}{nFv}-\frac{\alpha \left(1-\alpha \right) nF\Delta {E}_p}{2.3 RT} $$

Where α represents electron transfer constant, ΔE p is peak-to-peak separation. When the scan rate sets as 100 mV⋅s−1, k s can be estimated to be 2.32 s−1. The estimated value is in the controlled range of surface-controlled quasi-reversible process, and is larger than that of Hb assembled on AuNPs [43]. Such results reveal a reasonably fast electron transfer between the immobilized Hb and the underlying electrode.

The electrochemical behavior of Hb is significantly dependent on the solution pH. Therefore, the effect of solution pH was investigated by CV at different pH values of phosphate buffer and the results are shown in Fig. S2A and the corresponding plot of formal peak potential (E 0’) against pH is shown in Fig. S2B. The E 0’ was found to be linear with pH values over the range of 4.0–9.0 with a linear regression equation of E 0’(V) = 0.082–0.040 pH (R2 = 0.984). The slope value of −40 mV⋅ pH −1 is smaller than the theoretically expected value of −57.6 mV⋅ pH −1 for a single-proton coupled to the reversible one-electron transfer at 18 °C [44]. It may be due to the influence of the protonization of ligands to the heme iron and amino acids around the heme [45]. The influence of temperature of phosphate buffer on the cyclic voltammograms of Hb-Ag@GNR-PSS-PDDA/GCE is shown in Fig. S3A. The details discussed in ESM indicate that Hb still maintain good biological activity with the temperature in the range of 15–45 °C.

Electrocatalytic behavior of Hb-Ag@GNR-PSS-PDDA/GCE

To explore the applicability of the Hb-Ag@GNR-PSS-PDDA/GCE for TCA biosensing, CVs were carried out to demonstrate the electrochemical reduction of TCA at the modified electrode in pH 4.0 phosphate buffer. The electrochemical reduction of TCA can be promoted in acid medium due to a large number of hydrogen ions on the surface of modified electrode. Meanwhile, because the isoelectric point of Hb is about 6.8 [46], the Hb-Ag@GNR-PSS-PDDA/GCE would be positively charged at pH 4.0. Due to the anionic property of TCA, the acidic Hb-Ag@GNR-PSS-PDDA/GCE surface seems profitable for the adsorption of TCA. Therefore, the acid medium can promote the electrocatalytic activity of TCA.

As shown in Fig. 3a, the electrochemical response of TCA at Hb-Ag@GNR-PSS-PDDA/GCE was investigated using CV. When 2.0 μM TCA was added into the phosphate buffer, a great increase of the reduction peak was observed at about −0.4 V with the simultaneous disappearance of the oxidation peak. In addition, compared with the single nanoparticles modified electrode, it can be clearly seen that Hb-Ag@GNR-PSS-PDDA/GCE exhibits more sensitive response to TCA. It reveals that Ag@GNRs can accelerate the electron transfer between TCA and the modified electrode owing to the synergistic effect of AgNPs and GNRs together. Therefore, the modified electrode based on Ag@GNRs can be used as a good electrochemical platform for determining TCA.

Fig. 3
figure 3

a Cyclic voltammograms of (a) Hb-GNR-PSS-PDDA/GCE, (b) Hb-AgNP-PSS-PDDA/GCE and (c) Hb-Ag@GNR-PSS-PDDA/GCE in pH 4.0 phosphate buffer containing 2.0 μM TCA. Scan rate: 100 mV⋅s−1. b Cyclic voltammograms of Hb-Ag@GNR-PSS-PDDA/GCE in pH 4.0 phosphate buffer containing 0, 1.3, 2.0, 2.7, 3.4, 5.3 and 8.7 μM TCA (curves a–g)

Full size image

Fig. 3b shows the CVs of Hb-Ag@GNR-PSS-PDDA/GCE in pH 4.0 phosphate buffer containing different concentrations of TCA. When the concentration of TCA increased, the reduction peak current at −0.40 V increased gradually with the disappearance of the oxidation peak current which demonstrates the typical characteristic of the electrocatalytic reaction. Electrocatalytic reduction mechanism of TCA by the Hb modified electrode can be expressed with the following Eqs. [47]:

$$ \mathrm{HbhemeFe}\left(\mathrm{II}\mathrm{I}\right)+\mathrm{e}\to \mathrm{HbhemeFe}\left(\mathrm{II}\right) $$
$$ 2\mathrm{HbhemeFe}\left(\mathrm{II}\right){+\mathrm{CCl}}_3{\mathrm{COOH}+\mathrm{H}}^{+}+\mathrm{e}\to 2\mathrm{HbhemeFe}\left(\mathrm{II}\mathrm{I}\right){+\mathrm{Cl}}_2{\mathrm{CHCOOH}+\mathrm{Cl}}^{\_} $$

Fig. 4 shows the typical amperometric response of TCA on the Hb modified electrode at an applied potential of −0.4 V. The current increases with each increment of TCA within a linear range from 0.16 μM to 1.7 μM. The linear regression equation is i(A) =0.00655 [TCA] (M) + 8.77 × 10−8 (R2 = 0.9975). The detection limit is 0.12 μM (S/N = 3). Electrocatalytic activity of Hb-Ag@GNR-PSS-PDDA/GCE and other sensors toward TCA are compared in Table 1. From Table 1, we can see that the detection potential is relatively low and thus the oxidation of common interferents can hardly happen at such negative potential. Meanwhile, compared with other reports, the detection limit is lower than that of some reported values such as CTS/Hb/Ag/GR/CILE (42 mM) and CTS/Hb-Fe3O4/GCE (33 μM). However, the linear range of this biosensor is narrower than other reported methods [48, 49].

Fig. 4
figure 4

a The amperometric response of HbAg@GNR-PSS-PDDA/GCE upon successive addition of 10 μL TCA to pH 4.0 phosphate buffer with stirring. Working potential: −0.4 V (vs. SCE) b Linear relationship of electrocatalytic current vs. concentration of TCA

Full size image
Table 1 Comparison of the analytical performance for TCA with others
Full size table

The apparent Michaelis-Menten constant (K m), which is an indication of the enzyme-substrate reaction kinetics, is calculated by the Lineweaver-Burk eq. [50]:

$$ \frac{1}{I_{\mathrm{SS}}}=\frac{K\mathrm{m}}{I_{\max }}\cdot \frac{1}{C}+\frac{1}{I_{\max }} $$

Where, I ss is the steady-state current after the addition of substrate, C is the bulk concentration of substrate, and I max is the maximum current under saturated substrate conditions and measured as 0.102 μA. K m value for Hb-Ag@GNR-PSS-PDDA modified electrode was estimated to be 0.996 μM. The lower value of K m was smaller than that of Hb immobilized on CTS/NG/Hb film modified CILE (4.3 mM) [51] and CTS/Hb/Ag/GR modified CILE (10.2 mM) [52].

Interference studies

The potential interferences to the determination of 1.0 μM TCA were investigated on Hb-Ag@GNR-PSS-PDDA/GCE under the optimized conditions. The changes of the peak current were calculated with the results summarized in Table 2. It can be seen that 100 times of the TCA comcentration of Ba2+, Ca2+, Mg2+, Zn2+, Na+, K+, Cu2+, SO4 2−, NO3 −and Cl− did not interfere with the electrochemical responses of TCA. Other species such as dichloroacetic acid and monochloroacetic acid can be present at less than 30 times of the TCA concentration. In addition, the same concentration of Pb2+ can be tolerated with the changes less than ±5%. The only serious interference was H2O2. The same effect has been observed earlier and has been the subject of many studies [53, 55].

Table 2 Influence of coexisting substances on the detection of 1.0 μM TCA (n = 3)
Full size table

Determination of TCA in laboratory water samples

This electrochemical method was used to detect the TCA content in laboratory water samples and the results were listed in Table 3. The accuracy of the method was evaluated by the recovery test which was implemented by the standard addition method. The recoveries are ranged from 92.71% to 107.50%, which indicate that this sensor exhibits an acceptable accuracy for the analysis of TCA in laboratory water samples.

Table 3 Analysis of TCA in laboratory water samples (n = 3)
Full size table

Stability and reproducibility of Hb-Ag@GNR-PSS-PDDA/GCE

We studied the stability of the biosensor by storing it at 4 °C in pH 4.0 phosphate buffer and measured intermittently. A week later the current response of the biosensor still retained 95% of the initial value. The good stability of the biosensor can be ascribed to the Ag@GNR-nanocomposite, which provides a favorable microenviroment for immobilizing Hb effectively and keep its bioactivity on the electrode surface. The repeatability was also investigated for 6 successive measurements at TCA concentration of 1.0 μM and it exhibited a good result with the RSD of 4.2%.

Conclusions

In summary, the Ag@GNRs with the shell-core structure was successfully synthesized and used as an effective electrocatalyst for fabrication of the Hb modified electrode. Hb on the Ag@GNR-film keeps its natural structure and bioactivity and the fast direct electron transfer is realized. The electrochemical studies suggest that the resulting electrode exhibits good electrocatalytic activity to the reduction of TCA. The amperometric detection of TCA exhibits a wide linear concentration range and a low detection limit. Therefore, the application of Ag@GNR-composite can afford an effective platform for the immobilization of Hb and the construction of biosensors. The biosensor possesses good stability and reproducibility, which would be capable of being a potential biosensing platform in the electroanalysis and electrocatalysis.