Novel pH-responsive nanoplasmonic sensor: controlling polymer structural change to modulate localized surface plasmon resonance response

Abstract

The detection of chemical or biological analytes in physiological media remains a great challenge and current methods suffer from low sensitivity, reproducibility, and require expensive instruments. Here we report the design of a simple, pH-responsive nanoplasmonic sensor utilizing polymer structural changes to induce localized surface plasmon resonance (LSPR) shifts. The sensors were fabricated by chemical attachment of poly(allylamine) onto ∼28 nm gold nanoprisms bound to a silanized glass surface. The reversible change of polymer structure upon protonation and deprotonation of its amine groups alters the nanoprisms' LSPR properties. A spectral shift of the nanoprisms' dipole peak was observed because of changes in thickness of local dielectric environment, which are caused by shrinking and swelling of the pH-responsive polymer. The pH-induced shrinking and the swelling transition provided the opportunity to design ultrasensitive glucose sensors. The pH change from oxidation of β-glucose by glucose oxidase, resulted in up to a 17 nm LSPR peak shift because of the 3.7 nm change in polymer thickness measured by in situ atomic force microscopy. The lowest concentration of glucose that can be repeatedly detected in bovine plasma with this sensor was 25 μM. This nanoplasmonic sensor exhibited simplicity of operation and excellent reproducibility. The polymer-functionalized sensor provided a powerful avenue for simple, ultrasensitive, and cost effective detection of target analytes, which can be translated to clinical application.

---

1. Introduction

Noble metal nanostructures display distinct localized surface plasmon resonance (LSPR) properties, which occur because of the interaction between incident photons and surface conduction electrons of the metal. These LSPR properties are strongly dependent on the size,^1–4 shape,^5–18 and composition of nanostructures.^19,20 Importantly, the dielectric constant of the surrounding environment controls the LSPR peak position^21–24 and nanoplasmonic sensors have been developed to detect biological or chemical constituents by monitoring LSPR changes induced by their presence.^{10,12,21,25,26} In every case, these changes are caused by varying the bulk or local dielectric environments of nanostructures and are monitored by optical spectroscopy.^3,21,24,27

Currently nanoplasmonic sensors are designed by attaching metallic nanostructures onto solid substrates,^12,23,24 which provide versatility and stability to the sensors. In most solid-state sensors the surface of the plasmonic nanostructure is functionalized with receptor molecules and their interaction with analytes changes the local dielectric environments and thus alters the LSPR properties. In this context, attachment of stimuli-responsive polymers onto metallic nanostructures provides many advantages because of their unique chemical and physical properties that can be modulated by temperature, pH, and solvent polarity.^22,28,29 All these external stimuli can induce a reversible swelling and shrinking polymer phase transition, which alters the local dielectric environment of nanostructures. Therefore, highly efficient nanoplasmonic sensors^30,31 can be constructed utilizing conformational changes induced in the polymer chains. Here we present a new nanoplasmonic sensing platform constructed with pH-responsive, polymer-functionalized gold nanoprisms as ultrasensitive sensors that can detect 25 μM of glucose in bovine plasma with high specificity and reproducibility. Our sensing platform displayed an ∼17 nm LSPR dipole peak red-shift of gold nanoprisms upon a change of solution pH from 7.1 to 3.5. Additionally, we have demonstrated that the surface ligand chemistry plays an important role to achieve the highest sensitivity.

Recently, significant effort has been focused on the determination of glucose concentration based on electrochemical^32–36 and colorimetric³⁷ techniques, as well as on optical,^38–40 fluorescent,^39,41 and surface enhanced Raman spectroscopies (SERS).^42–45 Accurate determination of the glucose level in physiological samples is extremely important because it can be used as a monitor of diabetes. However, the above-mentioned techniques suffer from selectivity, sensitivity and reproducibility problems and from cytotoxicity of the sensors. Additionally, aerobic oxidation and degradation of the sensing platform, irreversible aggregation of the nanoparticles, and the requirement of high salt concentration during the measurement restrict their practical application in clinical diagnosis.^26,46–48 Therefore, fabrication of simple and cost effective sensors for detection of glucose in physiological samples remains a great challenge. Furthermore, if an optical-based sensor could be developed it would circumvent the complications of electron or energy transfer in electrochemical- and fluorescent-based methods and simplify the intricate instrumentation in SERS by employing a simple UV-vis spectrophotometer.

Fig. 1 illustrates the fabrication of the platform where release of protons from the enzymatic catalysis of glucose by glucose oxidase (GOx) causes swelling and shrinking of the pH responsive polymer poly(allylamine) (PAA), which is covalently bound to gold nanoprisms attached to a glass substrate. The pH-responsive polymer undergoes a swelling and shrinking transition with change in pH because of protonation or deprotonation of the polymer.⁴⁹ Specifically, PAA undergoes a similar transition upon protonation and deprotonation of –NH₂ groups⁵⁰ due to electrostatic repulsion inside the polymeric chains. Using this principle, we have developed a nanoplasmonic sensor where swelling and shrinking transitions alter the local dielectric environment of the nanoprisms causing changes in the LSPR peak that can be detected by UV-visible spectroscopy.

Fig. 1 Schematic diagram representing the fabrication of the polymer-functionalized gold nanoprisms for glucose detection. Gold nanoprisms were attached onto a silanized glass surface. A self-assembled monolayer (SAM) of 11-mercaptoundecanoic acid was prepared on the surface of the nanoprisms. Poly(allylamine) was then covalently bound to the surface of the nanoprisms via amide coupling between the acid group on the SAM and the amine group on the polymer. The resulting pH-based nanoplasmonic sensor was used for glucose detection in solution in the presence of dissolved GOx.

This sensing platform provides several advantages over those reported in the literature: (1) gold nanoprisms are more sensitive to changes in the local refractive index versus other anisotropic metallic nanostructures due to the high EM-field enhancement near the sharp tips and edges.¹³ (2) Gold is very stable under ambient conditions, which provides additional stability to the sensing platforms. (3) No salt is required to control the swelling and shrinking of the polymer, which leads to accurate measurements. (4) Importantly, the covalent attachment of the polymers onto the surface of the nanoprisms prevents polymer detachment during the change of solution pH, which is the major drawback of the hydrogel-based method of glucose detection. To the best of our knowledge, this is the first example where a UV-vis spectroscopy-based assay has been used to detect physiologically important species in bovine plasma. Finally and most importantly, our nanoplasmonic pH sensor would be more superior than the conventional pH sensor, which not only requires large volume of sample for measurement but also could display inaccurate pH value due to potential interference from various biological molecules present in the physiological samples.

2. Materials and methods

2.1. Materials

Chloro(triethylphosphine) gold(I) (Et₃PAuCl), poly(methylhydrosiloxane) (PMHS, M_n = 1700–3300), trioctylamine (TOA), (3-mercaptopropyl)-triethoxysilane (MPTES), poly(allylamine hydrochloride) (PAA), 11-mercaptoundecanoic acid (MUA), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC), N-hydroxysuccinimide (NHS), glucose, glucose oxidase (GOx) from Aspergillus niger, bovine plasma, uric acid, ascorbic acid, anhydrous acetonitrile (CH₃CN), methanol, and ethanol were purchased from Sigma Aldrich and were used as received. Hydrochloric acid was obtained from Acros Organics and sodium hydroxide (NaOH) was purchased from Fisher Chemicals and used without additional purification. RBS 35 Detergent was obtained from Thermo Fisher Scientific and used as received. Water was purified using a Thermo Scientific Barnstead Nanopure system (pH = 7.1). Glass coverslips were purchased from Electron Microscopy Sciences. Prior to use bovine plasma was filtered through 0.45 μM diameter pore size filter. Gold nanoprisms with average edge length of 28 ± 2.8 nm were synthesized via a chemical reduction method using our previously published procedure.^51,52 Detail experimental procedures are provided in the ESI.†

2.2. Fabrication of pH-responsive nanoplasmonic sensors

Supporting substrate-bound gold nanoprisms were cleaned with tape to remove non-prismatic nanostructures. This step was performed according to our previous report and critical to enhance the performance of the sensors.⁵¹ The nanoprisms were then functionalized with MUA by incubating in a 1.0 mM ethanolic solution of MUA for 4 h. After 4 h of incubation they were rinsed with copious amount of ethanol to remove any loosely bound MUA and finally dried under nitrogen flow. The MUA functionalized nanoprisms were submersed in a 1 : 1 (v/v) aqueous solution of EDC and NHS (0.2 M each) for 1 h and then thoroughly rinsed with nanopure water and incubated in an aqueous solution of PAA (2 mg mL⁻¹, pH = 7.1) for 12 h. The PAA functionalized nanoprisms (sensing platform) were rinsed with nanopure water and dried under N₂ flow and used for glucose sensing. For glucose sensing, different concentration of glucose, i.e., 10 mM, 5 mM, 1 mM, 100 μM, 75 μM, 50 μM, and 25 μM, where prepared in water by serial dilution. The limit of detection (LOD) for glucose sensing both in water and bovine plasma were determined by measuring the λ_LSPR shift of the sensing platforms in glucose solutions (without GOx) six times and then subtracting three times the standard deviation from the average of the relative response of the sensing platform in presence of 25 μM of glucose and 15 units of GOx. The resulting Δλ_LSPR shift then converted into a relative concentration by using the linear relationship between the Δλ_LSPR shift and glucose concentration in the range of 25 μM to 100 μM.

3. Results and discussion

Recently, we have reported that chemically synthesized gold nanoprisms are much more sensitive to the changes of both bulk and local dielectric environment compared to other anisotropic metallic nanostructures.^51–53 The gold nanoprisms were synthesized and functionalized onto supporting substrates according to our previously published method.⁵⁴ The gold nanoprisms displayed the dipole peak (λ_LSPR) at 694 nm (Fig. 2A) in air when attached onto a silanized glass surface. These nanoprisms have an edge length of ∼28 nm (Fig. 2B) and a height of ∼8.3 nm (Fig. 2C and D) as determined from scanning electron microscopy (SEM) and atomic force microscopy (AFM), respectively.

Fig. 2 (A) UV-visible extinction spectra of substrate-attached nanoprisms in air after tape cleaning (λ_LSPR = 694 nm). (B) SEM image of gold nanoprisms onto silanized glass coverslip after tape cleaning. The average edge length of gold nanoprisms were 28 ± 2.8 nm. Scale bar is 100 nm. (C) AFM image of gold nanoprisms on silanized glass surface. (D) Height profile of an individual nanoprism. Average height was 8.3 ± 0.2 nm.

During the initial investigation the substrate-attached nanoprisms were incubated in an aqueous PAA (2 mg mL⁻¹) solution at neutral pH (7.1) for 12 h. These PAA-adsorbed gold nanoprisms were rinsed with a copious amount of nanopure water, dried, and an extinction spectrum was collected in air. An ∼8 nm λ_LSPR red shift was observed, which is due to the adsorption of PAA onto the nanoprism surface. The surface analysis by AFM showed an increase of 1.9 nm in height from the polymer (ESI-Fig. 1A†). Furthermore, the nanoprisms lost their sharp features, which is also an indication of polymer adsorption onto their surface. The 1.9 nm change in thickness due to PAA adsorption is in agreement with previous literature that the thickness of single monolayer of PAA is ∼2.0 nm.^6,7 The extinction spectra of PAA-adsorbed gold nanoprisms were measured in different pH solutions ranging from 7.1 to 3.5. A maximum ∼4 nm λ_LSPR red shift was observed (see Fig. 3A, red dots), which was irreversible. This small Δλ_LSPR could be due to (1) non-specific adsorption of PAA onto the nanoprisms, which resulted in low PAA coverage or (2) detachment of PAA from the nanoprisms during the pH change due to a weak gold–amine interaction. The AFM analysis (ESI-Fig. 1A†) proved that nanoprisms were completely covered with polymer. Therefore, detachment of PAA from nanoprisms at pH 3.5 is the reason for small and irreversible λ_LSPR shifts. The PAA-adsorbed gold nanoprisms at pH 3.5 were then rinsed with nanopure water and analyzed by AFM (ESI-Fig. 1B†), which showed that nanoprisms nearly regained their sharp features as observed before the PAA adsorption. Moreover, the height was also found to be 8.6 nm (ESI-Fig. 1C†), very close to the height of the nanoprisms before the PAA adsorption, which together indicated detachment of PAA.

Fig. 3 (A) The LSPR peak shift as a function of the solution pH without buffer. (B) AFM image and height profile (C) of pH-responsive nanoplasmonic sensor. Average height was 13.1 ± 1.1 nm. (D) UV-vis extinction spectra of sensing platforms at different solution pH. The sensor was fabricated according to Fig. 1.

In order to overcome the detachment of PAA from the nanoprisms, we chemically attached the polymers onto the nanoprism surface as shown in Fig. 1. The substrate-bound gold nanoprisms were first functionalized with MUA to form a self-assembled monolayer (SAM) that served as an anchor. Amide coupling was then carried out between the acid group of MUA and the amine group of PAA in the presence of EDC/NHS (see Experimental section). This process prevented the detachment of PAA from the nanoprism surface as it becomes protonated. After attachment of PAA by amide coupling, an ∼15 nm Δλ_LSPR was observed, which is in agreement with our previous report⁵¹ that addition of a 2 nm thick monolayer of PAA^6,7 would result in an LSPR shift of this magnitude (see ESI-Fig. 2†). The AFM analysis (see Fig. 3B and C) suggests that the total ∼4.7 nm increase in thickness was due to 1.7 nm from MUA SAM (Chem Draw 3D) and ∼3 nm from PAA. Furthermore, PAA did not form a perfectly homogeneous monolayer on the nanoprisms but rather some of its “arms” were extended outwards.

The extinction spectra of our pH-responsive nanoplasmonic sensor covalently attached with PAA were measured after incubating them in different pH solutions without buffer or salts ranging from 7.1 to 3.5 for 10 to 15 minutes until a stable λ_LSPR was observed. Fig. 3D displays the UV-vis extinction spectra at different solution pH. The λ_LSPR exhibited a red shift as the solution pH increased (Fig. 3A, blue diamond). An ∼1 nm λ_LSPR red shift was observed from pH 7 to 6.0 and another ∼4.0 nm shift was observed from pH 6.0 to 5.5. A sharp ∼11 nm Δλ_LSPR was observed from pH 5.5 to 4.5. Finally, another ∼1 nm shift from pH 4.5 to 4.0 was detected. No further shift was observed from pH 4.0 to 2.5. Therefore, a total ∼17 nm shift was observed from pH 7.1 to 4.0. According to our previous report,⁵¹ an ∼17 nm Δλ_LSPR should result a change of ∼3.6 nm thickness of PAA. To confirm the PAA thickness change upon pH changes, in situ AFM analyses were conducted to measure the height of the polymers while incubating the nanoplasmonic sensors in two different pH solutions, 7.1 and 3.5. Fig. 4 illustrates the AFM images of our pH-responsive nanoplasmonic sensors in pH 7.1 and 3.5 solution, respectively. At pH 7.1 the average height was 13.1 ± 2.1 nm and when the same samples were analyzed at pH 3.5 the height increased to 16.6 ± 2.1 nm. Therefore, a 3.5 ± 2.1 nm increased of height was detected from pH 7.1 to 3.5, which resulted in a Δλ_LSPR of ∼17 nm. At lower pH, the nanoprisms looked fuzzier and had blooms, which was apparently due to swelling of the polymer because of the electrostatic repulsion between ammonium groups present in the polymer backbone.⁵⁵ Interestingly, the maximum Δλ_LSPR was observed between pH 6 and 4 (Fig. 3A, blue diamond). Our finding is in agreement with the literature⁵⁶ which shows that at pH 6 only 20% of the amine groups in the PAA are protonated whereas the percentage of protonation increased to 60% at pH 4 and then remained nearly constant. Therefore in our system as the pH of the solution containing nanoplasmonic sensor gradually changed from 7.1 to 2.5, the amine groups gradually protonated resulting in increased positive charge in the polymer layer. Thus, the larger concentration of positive charge (–NH₃⁺) increased the charge repulsion inside the PAA layer, which caused an expansion of the layer and simultaneously increased the local dielectric shell thickness and red-shift of the LSPR peak. Therefore, the working principle of our pH responsive nanoplasmonic sensors is different than previously reported hydrogel-based methods,^57,58 where an increase of solution pH resulted in the swelling of the hydrogel due to absorption of water. Water has lower refractive index than hydrogel and water absorption reduces the local refractive index, which causes the λ_LSPR blue shift with the increase of solution pH. In this present investigation, the swelling or shrinking of PAA caused changes in thickness of dielectric shell, which influenced the λ_LSPR. As the solution pH become more acidic the protonation of amine groups in the PAA backbone is apparent and it swells, causing an increase in the dielectric shell. This process resembled to layer-by-layer assembly of polyelectrolytes onto plasmonic nanostructures, which induces the λ_LSPR red shift.^6,51,59 The detailed structural properties of the polymer film in the swelling and shrinking states is under investigation.

Fig. 4 In situ AFM images of pH-responsive nanoplasmonic sensor in pH 7.1 (A) and 3.5 (B) solution. AFM images were recorded 20 min after immersion in pH solution.

Our pH-responsive nanoplasmonic sensor was further characterized by optical spectroscopy to determine the change of solution pH and Δλ_LSPR over time in the presence of different GOx concentration. The observed Δλ_LSPR for our sensor proved to be higher than silver nanoparticles containing hydrogel-based plasmonic transducers.⁴⁶ The enzymatic reaction of GOx with glucose produces gluconic acid resulting in a pH change of the solution due to release of protons. In all measurements, the pH of the initial solution was ∼7.1. Initially, we simultaneously measured the Δλ_LSPR of our sensors and solution pH as a function of added GOx for a fixed concentration (5 mM) of glucose until both the pH and the Δλ_LSPR became stable. Fig. 5A represents the pH change and total Δλ_LSPR observed in the presence of different amounts of GOx ranging from 5 to 30 units (also see ESI-Fig. 3†). It was observed that in the presence of 5, 10, 12.5, 15, 20, 25, and 30 units of GOx, the pH of the solutions was 5.5, 4.8, 4.5, 4.2, 3.9, 3.7, and 3.6, respectively, and the corresponding λ_LSPR shifts were 2.5, 7.0, 9.8, 11.2, 14.6, 16.9, and 17.1 nm. Fig. 5A illustrates that as the units of GOx increased from 5 to 20, the Δλ_LSPR changed rapidly and then plateau at 25 and 30 units of GOx. Our pH-responsive nanoplasmonic sensors displayed the highest pH sensitivity at pH ∼4.0 (Fig. 5A), therefore, we decided to use 15 units of GOx for the remaining glucose detection experiments, which provides a pH of 4.2.

Fig. 5 (A) The changes of solution pH and LSPR peak shift as a function of the GOx concentration. The sensing platform was immersed in a 5 mM glucose solution. (B) The changes of solution pH and LSPR peak shift as a function of the incubation time. The sensing platform was immersed in a 5 mM glucose solution containing 15 units of GOx.

To determine the total incubation time needed for the response of the sensor to stabilize, a kinetic study was performed using a 5 mM glucose solution in the presence of 15 units of GOx. We simultaneously monitored the Δλ_LSPR and pH of the solution containing our nanoplasmonic sensor. The Δλ_LSPR of the sensing platform and the pH change of the solution were measured every 2 minutes directly after the injection of GOx into the glucose solution. Fig. 5B shows the Δλ_LSPR and pH change of the solution and that both the components became stable. The graph also illustrates that the Δλ_LSPR and pH changes occur rapidly in the beginning, slow down after ∼16 minutes, and stabilize after ∼25 minutes.

Several control experiments were also performed to confirm that the response of the pH-responsive nanoplasmonic sensor only occurred because of the catalytic enzyme reaction between glucose and GOx, and that there were no non-specific interactions during the experiments. In the first, exposure to uric and ascorbic acids were chosen because they are present in biological/physiological fluids and could interfere with the Δλ_LSPR of the sensor. The sensor was incubated in a solution containing 1.0 mM ascorbic acid and 0.2 mM uric acid for 15 min and then 15 units of GOx was added but no glucose. The solution pH and Δλ_LSPR of the sensor was monitored for 30 minutes. A 1.0 nm Δλ_LSPR was observed for the above acid solutions, however no pH change was detected, see Fig. 6A.

Fig. 6 (A) UV-visible extinction spectra of pH-responsive nanoplasmonic sensor in presence of a solution containing 1.0 mM ascorbic acid and 0.2 mM of uric acid before (blue, λ_LSPR = 764 nm) and after (red, λ_LSPR = 765 nm) addition of 15 units of GOx. (B) Reversibility study of pH-responsive polymer functionalized nanoplasmonic sensor. Blue square: incubation of sensor in 5 mM aqueous solution of glucose (pH ∼ 7.1). Red diamonds: after addition of 15 units of GOx. The extinction spectrum (Δλ_LSPR = 11.4 nm) was collected 30 min after addition of GOx. Green dots: the sensor at red diamond was rinsed with 1 mM of NaOH and then copious amount of nanopure water and incubated in 5 mM aqueous solution of glucose.

In the second control experiment, 15 units of GOx was added to pH 4.0 solution containing the sensing platform without any glucose. The purpose of this experiment was to determine whether GOx decomposes at pH < 5.0 and non-specifically adsorbs onto the sensing platform. The extinction spectra were measured before and after the addition of GOx. An ∼0.5 nm shift was observed for pH 4.0 (ESI-Fig. 4†). Thus the interference in both cases with sensor response is negligible and the Δλ_LSPR observed could be due to instrumental noise. Results from the control experiments indicate that the pH-responsive nanoplasmonic sensors are highly specific to glucose concentration.

Our nanoplasmonic sensor response depends upon changes in the local dielectric environment of the nanoprisms due to the modulation of PAA thickness upon protonation and deprotonation of the polymers. Therefore, it would be expected that our sensing platform is reversible with the pH. To test reversibility, the sensing platform was incubated in a 5 mM glucose solution (pH = 7.1) and then 15 units of GOx were added and allowed to react for 30 min. An 11.4 nm Δλ_LSPR shift was observed and the solution pH was ∼4.0. The sensing platform was then removed and rinsed with 1 mM NaOH to deprotonate the ammonium groups, followed by rinsing with nanopure water to remove any salts. The cycle was repeated 4 additional times and at this point a 10.7 nm Δλ_LSPR was detected, See Fig. 6B. Therefore, a total 0.7 nm loss of Δλ_LSPR was observed after five cycles of the measurement using a reagent as harsh as NaOH.

The transition from in vitro to in vivo analysis was investigated by determining the quantitative response of the sensing platform in the more complex medium bovine plasma. Sensor performance determines in water, which does not contain any biological components and is not quantitative. The plasma contains lipids, proteins, and several other biological components, which could interfere with sensor performance. Therefore, the limit of detection (LOD) of our pH-responsive nanoplasmonic sensor was determined both in water and bovine plasma and the sensing efficiency was compared. The sensors were incubated in aqueous glucose solutions ranging from 25 μM to 10 mM with 15 units of GOx for 30 min, following which the extinction spectra were recorded. Fig. 7 (blue squares) displays the observed total Δλ_LSPR with respect to different aqueous glucose concentrations. It shows that the lowest concentration of glucose that can be detected by the pH-responsive nanoplasmonic sensors was 25 μM, which displays a Δλ_LSPR of ∼2.6 nm. The LOD was determined based on our published literature⁵¹ and was found to be 19 μM. These experiments were repeated five times and similar results were obtained. The LOD was also determined in bovine plasma under the same experimental condition. Fig. 7 (red squares) displays the total Δλ_LSPR observed for different glucose concentrations ranging from 25 μM to 10 mM.

Fig. 7 The Δλ_LSPR as a function of glucose concentration for the pH-responsive nanoplasmonic sensor in different glucose solutions containing 15 units of GOx in water (blue) and in bovine plasma (red).

The LOD calculated for glucose in bovine plasma was 22 μM, which is significantly lower than the previously reported LOD (i) in urine (10 mM) using a colorimetric probe with gold nanoparticles,⁴⁷ (ii) in bovine plasma (4.62 mM) using SERS,⁴² (iii) of 0.1 mM by responsive thin hydrogel films loaded with silver nanoparticles in water,⁴⁶ and (iv and v) of 50 μM and 2.4 mM by colorimetric method using gold nanoparticles functionalized with charged polymers⁶⁰ and DNA.³⁷ A few electrochemical methods are also reported in the literature, however the LODs were around 100 mM.^33,36 A lower LOD of 10 μM for glucose detection was reported⁶¹ using core–shell semiconductor quantum dots in buffer, however decomposition of quantum dots at ambient laboratory conditions and long analysis times restrict their practical application. Furthermore, glucose detection of this particular sensor in biological fluids such as serum, blood, or plasma has not been investigated. All these drawbacks are overcome in this present study with a LOD of 22 μM in bovine plasma. A detailed comparison of LODs from different methods for glucose detection using nanostructures is provided in Table 1.

Table 1 The comparison of detection limit of glucose by different glucose sensors

Medium	Method of detection	Sensing platform	LOD	Ref.
Water	UV-visible optical spectroscopy	Thin hydrogel films loaded with silver nanoparticles	0.1 mM	46
Water	Colorimetric and UV-visible optical spectroscopy	Charge generation polymer with gold nanoparticles	50 μM	60
Water	UV-visible spectroscopy	pH-responsive polymer functionalized gold nanoprisms	19 μM	Present study
Buffer	Fluorescence	CdSe/ZnS quantum dots	10 μM	61
Buffer	Electrochemical	Gold nanoparticles functionalized with GOx	100 mM	36
Blood	UV-visible spectroscopy and colorimetric detection	ss-DNA functionalized gold nanoparticles	2.4 mM	37
Human serum	Colorimetric and UV-visible spectroscopy	Ceria-based bioactive paper	3.71 mM	62
Urine	Colorimetric and UV-visible absorption spectra	16-MHDA functionalized gold nanoparticles	10 mM	47
Bovine plasma	Surface enhanced Raman spectroscopy	Mixed decanethiol/mercapto hexanol partition layer	4.62 mM	42
Bovine plasma	UV-visible spectroscopy	pH-responsive polymer functionalized gold nanoprisms	22 μM	Present study

---

4. Conclusion

In summary, we have demonstrated the fabrication of a simple pH-responsive nanoplasmonic sensor for ultra-sensitive glucose detection. The swelling and shrinking transitions of the pH-responsive polymers modulated the local dielectric environment of gold nanoprisms and induced the reversible Δλ_LSPR. The sensors were highly specific and reversible. The LOD for glucose was found to be 19 μM and 22 μM in water and bovine plasma, respectively. The sensor we have fabricated has a number of advantages over the literature reports, specifically simplicity of operation, reversibility, and low LOD. Based on these experimental results, this pH-responsive nanoplasmonic sensor is a promising bio-analytical tool for simple, selective, ultra-sensitive, and cost effective detection of target analytes that can be linked to a solution pH change even in physiological samples.

Acknowledgements

This work was supported by start-up funds provided by IUPUI. We would like to thank Prof. B. Muhoberac (IUPUI) for helpful discussion.

References

1. M. M. Alvarez, J. T. Khoury, T. G. Schaaff, M. N. Shafigullin, I. Vezmar and R. L. Whetten, J. Phys. Chem. B, 1997, 101, 3706–3712.
2. Y. Bao, C. Zhong, D. M. Vu, J. P. Temirov, R. B. Dyer and J. S. Martinez, J. Phys. Chem. C, 2007, 111, 12194–12198.
3. R. Sardar and J. S. Shumaker-Parry, J. Am. Chem. Soc., 2011, 133, 8179–8190.
4. H. Chen, X. Kou, Z. Yang, W. Ni and J. Wang, Langmuir, 2008, 24, 5233–5237.
5. S. R. Beeram and F. P. Zamborini, ACS Nano, 2010, 4, 3633–3646.
6. G. J. Nusz, A. C. Curry, S. M. Marinakos, A. Wax and A. Chilkoti, ACS Nano, 2009, 3, 795–806.
7. M. A. Otte, B. Sepulveda, W. Ni, J. P. Juste, L. M. Liz-Marzan and L. M. Lechuga, ACS Nano, 2009, 4, 349–357.
8. E. Carbó-Argibay, B. Rodríguez-González, S. Gómez-Graña, A. Guerrero-Martínez, I. Pastoriza-Santos, J. Pérez-Juste and L. M. Liz-Marzán, Angew. Chem., Int. Ed., 2010, 49, 9397–9400.
9. A. Guerrero-Martínez, B. Auguié, J. L. Alonso-Gómez, Z. Džolić, S. Gómez-Graña, M. Žinić, M. M. Cid and L. M. Liz-Marzán, Angew. Chem., Int. Ed., 2011, 50, 5499–5503.
10. K. A. Willets and R. P. Van Duyne, Annu. Rev. Phys. Chem., 2007, 58, 267–297.
11. M. E. Stewart, C. R. Anderton, L. B. Thompson, J. Maria, S. K. Gray, J. A. Rogers and R. G. Nuzzo, Chem. Rev., 2008, 108, 494–521.
12. K. M. Mayer and J. H. Hafner, Chem. Rev., 2011, 111, 3828–3857.
13. L. J. Sherry, R. Jin, C. A. Mirkin, G. C. Schatz and R. P. Van Duyne, Nano Lett., 2006, 6, 2060–2065.
14. L. Rodriguez-Lorenzo, R. de la Rica, R. A. Alvarez-Puebla, L. M. Liz-Marzan and M. M. Stevens, Nat. Mater., 2012, 11, 604–607.
15. C. Novo, A. M. Funston and P. Mulvaney, Nat. Mater., 2008, 3, 598–602.
16. R. Jin, Y. Charles Cao, E. Hao, G. S. Metraux, G. C. Schatz and C. A. Mirkin, Nature, 2003, 425, 487–490.
17. R. Jin, Y. Cao, C. A. Mirkin, K. L. Kelly, G. C. Schatz and J. G. Zheng, Science, 2001, 294, 1901–1903.
18. Y. Sun and Y. Xia, Science, 2002, 298, 2176–2179.
19. K. L. Kelly, E. Coronado, L. L. Zhao and G. C. Schatz, J. Phys. Chem. B, 2002, 107, 668–677.
20. C. Novo, A. M. Funston, I. Pastoriza-Santos, L. M. Liz-Marzan and P. Mulvaney, J. Phys. Chem. C, 2007, 112, 3–7.
21. P. K. Jain, X. Huang, I. H. El-Sayed and M. A. El-Sayed, Acc. Chem. Res., 2008, 41, 1578–1586.
22. M. S. Strozyk, M. Chanana, I. Pastoriza-Santos, J. Pérez-Juste and L. M. Liz-Marzán, Adv. Funct. Mater., 2012, 22, 1436–1444.
23. N. Nath and A. Chilkoti, Anal. Chem., 2004, 76, 5370–5378.
24. J. N. Anker, W. P. Hall, O. Lyandres, N. C. Shah, J. Zhao and R. P. Van Duyne, Nat. Mater., 2008, 7, 442–453.
25. F. P. Zamborini, L. Bao and R. Dasari, Anal. Chem., 2011, 84, 541–576.
26. Q. Wu, L. Wang, H. Yu, J. Wang and Z. Chen, Chem. Rev., 2011, 111, 7855–7875.
27. A. J. Haes, S. Zou, J. Zhao, G. C. Schatz and R. P. Van Duyne, J. Am. Chem. Soc., 2006, 128, 10905–10914.
28. Q. Li, J. He, E. Glogowski, X. Li, J. Wang, T. Emrick and T. P. Russell, Adv. Mater., 2008, 20, 1462–1466.
29. S. Burkert, E. Bittrich, M. Kuntzsch, M. Müller, K.-J. Eichhorn, C. Bellmann, P. Uhlmann and M. Stamm, Langmuir, 2009, 26, 1786–1795.
30. M. A. C. Stuart, W. T. S. Huck, J. Genzer, M. Muller, C. Ober, M. Stamm, G. B. Sukhorukov, I. Szleifer, V. V. Tsukruk, M. Urban, F. Winnik, S. Zauscher, I. Luzinov and S. Minko, Nat. Mater., 2010, 9, 101–113.
31. I. Luzinov, S. Minko and V. V. Tsukruk, Soft Matter, 2008, 4, 714–725.
32. X. Jing-Juan and C. Hong-Yuan, Anal. Biochem., 2000, 280, 221–226.
33. L. Bahshi, M. Frasconi, R. Tel-Vered, O. Yehezkeli and I. Willner, Anal. Chem., 2008, 80, 8253–8259.
34. Q. Yan, B. Peng, G. Su, B. E. Cohan, T. C. Major and M. E. Meyerhoff, Anal. Chem., 2011, 83, 8341–8346.
35. J.-J. Xu and H.-Y. Chen, Anal. Chim. Acta, 2000, 423, 101–106.
36. Y. Xiao, F. Patolsky, E. Katz, J. F. Hainfeld and I. Willner, Science, 2003, 299, 1877–1881.
37. Y. Jiang, H. Zhao, Y. Lin, N. Zhu, Y. Ma and L. Mao, Angew. Chem., Int. Ed., 2010, 49, 4800–4804.
38. A. Chaudhary and R. Srivastava, Sens. Lett., 2008, 6, 253–260.
39. O. S. Wolfbeis, I. Oehme, N. Papkovskaya and I. Klimant, Biosens. Bioelectron., 2000, 15, 69–76.
40. X. J. Wu and M. M. F. Choi, Anal. Chim. Acta, 2004, 514, 219–226.
41. X. Li, Y. Zhou, Z. Zheng, X. Yue, Z. Dai, S. Liu and Z. Tang, Langmuir, 2009, 25, 6580–6586.
42. O. Lyandres, N. C. Shah, C. R. Yonzon, J. T. Walsh, M. R. Glucksberg and R. P. Van Duyne, Anal. Chem., 2005, 77, 6134–6139.
43. C. R. Yonzon, C. L. Haynes, X. Zhang, J. T. Walsh and R. P. Van Duyne, Anal. Chem., 2003, 76, 78–85.
44. D. A. Stuart, C. R. Yonzon, X. Zhang, O. Lyandres, N. C. Shah, M. R. Glucksberg, J. T. Walsh and R. P. Van Duyne, Anal. Chem., 2005, 77, 4013–4019.
45. K. E. Shafer-Peltier, C. L. Haynes, M. R. Glucksberg and R. P. Van Duyne, J. Am. Chem. Soc., 2002, 125, 588–593.
46. I. Tokarev, I. Tokareva, V. Gopishetty, E. Katz and S. Minko, Adv. Mater., 2010, 22, 1412–1416.
47. C. Radhakumary and K. Sreenivasan, Anal. Chem., 2011, 83, 2829–2833.
48. T. Endo, R. Ikeda, Y. Yanagida and T. Hatsuzawa, Anal. Chim. Acta, 2008, 611, 205–211.
49. S. S. Shiratori and M. F. Rubner, Macromolecules, 2000, 33, 4213–4219.
50. R. Sardar, N. S. Bjorge and J. S. Shumaker-Parry, Macromolecules, 2008, 41, 4347–4352.
51. G. K. Joshi, P. J. McClory, B. B. Muhoberac, A. Kumbhar, K. A. Smith and R. Sardar, J. Phys. Chem. C, 2012, 116, 20990–21000.
52. G. K. Joshi, K. A. Smith, M. A. Johnson and R. Sardar, J. Phys. Chem. C, 2013, 117, 26228–26237.
53. G. K. Joshi, K. N. Blodgett, B. B. Muhoberac, M. A. Johnson, K. A. Smith and R. Sardar, Nano Lett., 2014, 14, 532–540.
54. G. K. Joshi, P. J. McClory, S. Dolai and R. Sardar, J. Mater. Chem., 2012, 22, 923–931.
55. K. Itano, J. Choi and M. F. Rubner, Macromolecules, 2005, 38, 3450–3460.
56. J. Hodak, R. Etchenique, E. J. Calvo, K. Singhal and P. N. Bartlett, Langmuir, 1997, 13, 2708–2716.
57. N. H. Mack, J. W. Wackerly, V. Malyarchuk, J. A. Rogers, J. S. Moore and R. G. Nuzzo, Nano Lett., 2007, 7, 733–737.
58. H. Jiang, J. Markowski and J. Sabarinathan, Opt. Express, 2009, 17, 21802–21807.
59. A. J. Haes, S. Zou, G. C. Schatz and R. P. Van Duyne, J. Phys. Chem. B, 2004, 108, 6961–6968.
60. C. Li, J. Hu, T. Liu and S. Liu, Macromolecules, 2011, 44, 429–431.
61. L. Bahshi, R. Freeman, R. Gill and I. Willner, Small, 2009, 5, 676–680.
62. M. Ornatska, E. Sharpe, D. Andreescu and S. Andreescu, Anal. Chem., 2011, 83, 4273–4280.

---

Footnote

† Electronic supplementary information (ESI) available: Detailed synthetic procedure of nanoprisms, glass coverslips functionalization, and additional spectroscopic and microscopic characterization. See DOI: 10.1039/c4ra00117f

---