Optical fiber sensor for the detection of tetracycline using surface plasmon resonance and molecular imprinting

Optical fiber sensor for the detection of tetracycline using surface plasmon resonance and molecular imprinting
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Optical  fi ber sensor for the detection of tetracyclineusing surface plasmon resonance and molecular imprinting Roli Verma and Banshi D. Gupta* We present a simple and highly selective optical  fi ber sensor for the detection of tetracycline in foodstu ff sby using the combination of surface plasmon resonance (SPR) and a molecular imprinted polymer (MIP)matrix. The sensor is fabricated  fi rst by coating a thin  fi lm of silver metal over the unclad portion of anoptical  fi ber, and then preparing a molecular imprinting of the target molecules over the metal coatedregion. The MIP creates several binding sites/nano-cavities which have the complementary shape andfunctional groups of the target molecules on its surface. The sensor works on the wavelengthmodulation scheme in which the shift in resonance wavelength is measured with respect to the changein tetracycline concentration. Two kinds of tetracycline, tetracycline hydrochloride (TC) and oxy-tetracycline hydrochloride (OTC) have been selected as the imprinting molecules because these possesssimilar structures. The presence of tetracycline samples in the vicinity of the sensing region causes theinteraction between binding sites and target molecules which results in the change in the dielectricproperties of the sensing surface, causing the shift in the SPR spectrum towards the red region. Theselectivity of the sensor for dissimilar compounds, and cross-selectivity between TC and OTC, has beentested. The reusability of the sensor has also been checked. The present sensor is suitable forcommercialization for the detection of tetracycline in food as it has several advantageous features suchas low cost, ease of handling, a miniaturized probe, fast response, high selectivity, reusability and thepossibility of online monitoring and remote sensing. Introduction Biosensors have become a very relevant topic for researchersregarding health and environmental issues relating to humanhealth. There are several pesticides and antibiotics which areused to protect food and vegetables, 1 but when the quantity of pesticides/antibiotics is in excess of their recommended level, it can cause many problems for human health. Tetracycline is oneof the antibiotics which is used to treat bacterial infectionsincluding pneumonia, acne and skin infections. 2,3 Tetracyclinefalls into a class of medication called tetracycline antibiotics.The consumption of antibiotics by humans and in aquaculture,plants and the farming industry in large amounts becomes apoint of concern because large quantities of antibiotics such astetracycline may create di ffi culties associated with allergicsymptoms and skin diseases. Further, the intake of largequantities of tetracycline with food is hazardous for humanbeings. 4 In some countries the maximum and safe residue limit of tetracycline in di ff  erent foods is kept    xed, for example, 0.1mg kg   1 in milk and 0.01 mg kg   1 in honey. 5 Therefore, theconcentration level of residue of tetracycline in foodstu ff  sbecomes an intense issue of concern and its monitoring is very important to control food safety. There are several methodsreported in the literature for the detection of tetracycline using chromatography, electrochemical detection, ultraviolet absor-bance and microbiological methods. 6 – 9  All of these methods arecumbersome, expensive and time consuming. Hence, thedevelopment of a simple, quick and selective method for thesensing of tetracycline is a challenge for researchers.In recent years, the surface plasmon resonance techniquehas become very popular for the sensing of various physical,chemical and biological entities. 10 – 12 Thisis because it is simple,has a low response time and is inexpensive. Surface plasmonsare the charge density oscillations of conduction electrons at the metal – dielectric interface. These are transverse magneti-cally (TM) polarized waves which traverse along the interface,and the associated   eld decays exponentially in both the metaland the dielectric media. Surface plasmons are excited reso-nantly by   p -polarized light incident to the metal – dielectricinterface when the wave vector and frequency of the incident light matches with that of the surface plasmon. The equality condition is called surface plasmon resonance (SPR), and whenthis condition is satis  ed, a sharp dip is obtained in the output   Physics Department, Indian Institute of Technology Delhi, New Delhi  – 110016, India. E-mail:; Tel: +91-11-26591355 Cite this:  Analyst  , 2013,  138 , 7254Received 3rd June 2013Accepted 11th September 2013DOI: 10.1039/c3an01098h 7254  |  Analyst  , 2013,  138 , 7254 – 7263 This journal is  ª  The Royal Society of Chemistry 2013 Analyst  PAPER   transmitted power because the energy of the incident light iscoupled to the surface plasmon. The resonance conditions arehighly sensitive to the variation in the nature of the dielectricmedium in the vicinity of the sensing surface (metal layer). Theslight variation in the nature of the dielectric medium changesthe position of the resonance dip, and hence the SPR methodcan be used for sensing purposes. The SPR technique is basedon the Kretschmann con  guration, where the base of a highindex prism is coated with a thin   lm of metal. In the case of SPR-based optical   ber sensors, a small unclad portion in themiddle of the optical   ber is coated with a thin   lm of theplasmonic metal. The light is launched at one end of the optical  ber and is guided by the total internal re  ection phenomenonthrough the core of the optical   ber. The surface plasmons areexcited by the evanescent    eld at the metal and   ber coreinterface. The importance of this approach lies in the advan-tages o ff  ered by optical   bers and surface plasmons as they areimmunetoelectromagneticinterference,smallinsize, low-cost,can be used in remote sensing and online monitoring, show resistance to chemicals and are biocompatible.In some studies, the surface plasmon resonance (SPR)technique has been used along with molecular imprinting. 13 Molecular imprinting has been extensively used over the last decade due to its promising characteristics in chemical andbiological sensing applications, as it is an easy technique, haspredetermined selectivity for the target molecule, and possesseshigh a ffi nity and robustness. Molecular imprinted polymers(MIP) require the self assembly of the target molecule with thepolymerizable monomers and cross-linker. A    er the comple-tion of the polymerization process, target molecules areremoved and   nally a molecularly imprinted sensing surface isobtained. MIPs are organic structures containing molecularrecognition sites which are complementary in shape, size andfunctional groups to the target molecule. The molecularly imprinted surface results in tailor-made high receptor cavities which can easily rebind with template molecules from closely related compound mixtures. MIPs have several promising properties, such as high stability, strong reusability, low cost and easy synthesis. 14 Molecular imprinting is a good alternativeto the use of enzymes and immunoassays in the   eld of bio-sensing. MIPs have been used for several biological sensing applications involving proteins, cocaine, dopamine  etc. 15 – 17 In the present study we have carried out the fabrication andcharacterization of a   ber optic tetracycline sensor using theSPR technique along with molecular imprinting. The SPR isused as the transducer, while the MIP technique is used for thepreparation of the sensing surface. The MIP generates speci  cbinding sites with functional groups that are complementary tothe target molecules. These are mechanically stable and with-stand manipulation. The MIP structures are micro-porous toallow the binding and extraction of the target molecules. Whenthe analytes bind on the sensing surface, the interfacial archi-tecture changes and the surface plasmons become excited,generating a readable optical signal  i.e.  the shi    in the reso-nance dip. Tetracycline and oxy-tetracycline have been chosenas imprinting molecules to check the stability and selectivity of the MIP-produced sensing surfaces. MIP is synthesised by thepolymerization method by using the monomers acrylamide(AM) and acrylic acid (AA) and the cross-linker BIS acrylamide.The sensor is characterized by the testing of various concen-trations of tetracycline (TC) and oxy-tetracycline (OTC) ,whichhave similar structures, as shown in Fig. 1. In addition, theselectivity of the MIP probes towards dissimilar analytes, ureaand vitamin B 3 , has also been checked and found non-operativetowards these analytes. Experimental Materials Multimode plastic clad silica (PCS) optical   bers of 0.37numerical aperture (NA) and 600  m m core diameter werepurchased from Fiberguide Industries. Tetracycline hydrochlo-ride (TC) and oxy-tetracycline hydrochloride (OTC) werepurchased from CDH Bioscience Pvt. Ltd. Acrylamide (AM),sodium dihydrogen phosphate dihydrate (NaH 2 PO 4 $ 2H 2 O),disodium hydrogen phosphate dihydrate (Na 2 HPO 4 $ 2H 2 O),sodium lauryl sulphate (SDS) and urea were procured fromMerck India.  N  ,  N  -Methylenebisacrylamide (BIS acrylamide),acrylic acid (AA) and glacial acetic acid were purchased fromFisher Scienti  c. Water from Millipore   System was used formaking the bu ff  er and aqueous solutions of tetracycline (TC)and oxy-tetracycline (OTC). Vitamin B 3  was purchased fromSigma Aldrich. Ammonium persulfate (APS) and  N  -tetramethy-lethylenediamine (TEMED) were obtained from Sisco ResearchLaboratory Pvt. Ltd. All of these chemicals were used without any further puri  cation. Silver (Ag) (99.99% pure) wire wasprocured from a local vendor. Acrylamide and acrylic acid wereused as monomers while BIS acrylamide was used as the cross-linking reagent. TEMED worked as a catalyst and APS as aninitiator of the reaction. Fig. 1  Molecular structure of the two tetracyclines. This journal is  ª  The Royal Society of Chemistry 2013  Analyst  , 2013,  138 , 7254 – 7263 |  7255 Paper Analyst  Preparation of the sensing probe The sensing probe was prepared in two steps. In the   rst step,the metal-coated   ber was prepared while in the second step,the MIP was formed over the metal-coated sensing region of theoptical  ber. For the  rst step, the cladding was removed from a1 cm length of the 17 cm long PCS   ber from the middleportion. The unclad portion was cleaned twice with acetonebefore performing the coating. The unclad portion was coated with a 40 nm thick    lm of silver metal by the thermal evapo-ration method. A 40 nm coating of silver was applied as it isreported to be the optimum thickness for sensing by the SPR technique. The complete coating procedure is given in theliterature. 18 In brief, the target silver metal was kept in amolybdenum spiral boat and was evaporated in a vacuumchamber kept at a pressure of 5    10  6 mbar while coating. Water was running continuously during the whole process of deposition for the cooling of the vacuum chamber. The thick-ness of the metal   lm was monitored online by a quartz crystalmicrobalance (QCM) digital thickness monitor, which was  xedin the vacuum chamber. The metal-coated region of the   ber was further cleaned with a copious amount of de-ionized water.In the second step, molecular imprinting was performed overthe metal-coated probe. A phosphate bu ff  er of 0.1 M and pH 7 was prepared using Na 2 HPO 4 $ 2H 2 O and NaH 2 PO 4 $ 2H 2 O inMillipore   water. The phosphate bu ff  er was used to make thepolymerization precursor. For the preparation of the polymeri-zation precursor, both the AM and AA monomers were used with the BIS acrylamide cross-linker. The sensing capability of MIPs depends on the ratio of monomers and cross-linker, andhence on the swelling and shrinkage properties of the polymermatrix. However, the MIP network becomes hard when thecross-linking density in the polymer is high, resulting in theslowing of the sensing capabilities. Therefore, various permu-tations and combinations of di ff  erent ratios of monomers andcross-linker were tried before   nalizing the ratio (19 : 1) in thepresent study. The master solution was prepared by mixing  AM/BIS (19 : 1) and TC molecules (40% of AM) in Millipore   water. It was stirred for 10 minutes in a nitrogen atmosphere.The polymerization medium was prepared by mixing appro-priate amounts of the master solution (2.5 mL), phosphatebu ff  er (4.5 mL), AA (30  m L), APS and TEMED in a   ask anddiluting up to 10 mL, and then poured in a cylindrical vessel. Inthis preparation of the solution, APS was added as the reactioninitiator while TEMED was used as the catalyst. Immediately a   er this process, the metal-coated cleaned  ber probe was dip-coated with the polymerization medium and kept in a vacuumovenat60  Cfor4hoursforpolymerization.Thepolymerization was carried out using the standard mechanism of polymeriza-tion. During the process of polymerization, the polymer matrix became attached over the silver-coated   ber core. The adhesionof the polymer matrix over the silver surface mainly depends onthe interaction of metal ions with the polymer matrix. Theinteraction occurs due to electrostatic forces and the formationof coordinating bonds. 19  Amino groups form stable complexesthrough the lone pair of the nitrogen atom present in themonomers; this is the main cause of the adhesion of thepolymer on the silver-coated surface. Moreover, polymer – metalion interactions may be interchain and/or intrachain. Oncemetal ions interact and form bonds with the polymer, thepolymer matrix becomes stably    xed on the silver surface. A    erthe completion of this polymerization process, the   ber wasremoved from the oven. Then it was dipped in a 10% (v/v) aceticacid solution containing 10% (w/v) SDS for 2 hours at roomtemperature for the removal of the TC target molecules. It wasthen washed thoroughly with a copious amount of de-ionized water. It may be noted that the polymer matrix stably attachedon the silver surface remains there even a   er the removal of thetarget molecules and the creation of nano-cavities on the MIPmatrix. Finally, the MIP probe was dried for 3 hours beforecharacterization. This TC-MIP probe has several binding sites/nano-cavities which are complementary in shape and func-tional group to the target molecules. In addition to the TC-imprinted probe (TC-MIP), the OTC imprinted (OTC-MIP) probe was also prepared by the same method, replacing TC with OTCas the imprinting molecule. Moreover, a non-imprinted (NIP)probe was also prepared by the same procedure when the target molecules were not added to the master solution. The stepwiseprocedure of the preparation of the sensing probe is shown inFig. 2. The low density cross-linker polymer for MIP preparationpossesses high sensing capabilities. Preparation of the samples The stock solutions of TC and OTC were prepared by adding 0.002 g of TC and OTC in 20 mL de-ionized water in di ff  erent  vessels, and stirring well for 5 minutes. Next, aqueous samplesof tetracycline and oxy-tetracycline of di ff  erent concentrationsrangingfrom0.0 m M to0.96 m Mwere preparedbythe dilution of their stock solutions in 10 mL de-ionized water. All of thesolutions were stirred well to ensure the homogeneity of thesolutions. The aqueous samples of urea and vitamin B 3  of  various concentrations ranging from 0.0  m M to 0.96  m M werealso prepared by the same procedure as stated above. Thesesamples were prepared to check the selectivity of the probe fordissimilar analytes. The refractive indices of all of the samples were measured in white light by Abbe's refractometer, with aresolution of 0.001. All of the samples were found to have thesame refractive index as water, within the accuracy of therefractometer.  Apparatus and experimental set-up The experimental setup used for the characterization of theTC-MIP and OTC-MIP sensing probes is shown in Fig. 3. The  ber optic sensing probe was cleaved at both ends by a tungstencutter for launching maximum light. Further, the probe was  xed in a  ow cell. Polychromatic light from a tungsten halogenlamp (THL) (Avalight-Hal) was used for the launching of light into the   ber. Its output power is stabilized by a stabilizing circuit, which generates a regulated current. A fan regulates theair  ow near the heat sink to maintain a constant temperature.Theincidentlightguidedthrough theoptical  berwascollectedat the other end of the optical   ber by a spectrometer (AvaSpec-3648) interfaced with a computer. For the calibration of the 7256  |  Analyst  , 2013,  138 , 7254 – 7263 This journal is  ª  The Royal Society of Chemistry 2013 Analyst Paper   sensor,SPRspectrawererecordedforaqueoussample solutionsof TCs, OTCs and dissimilar compounds urea and vitamin B 3  of di ff  erent concentrations ranging from 0.0  m M to 0.96  m M. Thesensing surface was washed with Millipore   water before add-ing any fresh sample in the   ow cell, in order to remove any residue of the previous sample solution and avoid interferencefrom le   -over aqueous solutions. Results and discussion The TC and OTC molecules, featuring similar structures, wereused for imprinting. As mentioned above, we prepared TC-MIP,OTC-MIP and NIP probes. Below, we present the results of theircharacterization, selectivity, speci  city and sensitivity. Characterization of the TC-MIP probe To characterize the TC-MIP probe, di ff  erent concentrationsamples of tetracycline were used in the   ow cell having theTC-MIP probe, and their SPR spectra were recorded. Fig. 4(a)shows the SPR spectra for the TC samples of di ff  erent concen-trations ranging from 0.0  m M to 0.96  m M. All of the spectra wererecorded within 40 seconds a   er pouring the sample in the  ow cell because it was observed that within this time, the opticalsignal becomes a permanent signal. In all, 12 SPR spectra wererecorded but only 8 spectra have been shown to avoid theoverlap between the spectra. It is observed that the SPR curvesshi    towards higher wavelengths with the increasing concen-tration of TC from 0.0  m M to 0.96  m M. The resonance dip cor-responding to the minimum transmitted power in the SPR curve obtained at a particular wavelength, called the resonance wavelength, is the characteristic of the sensor for a particularconcentration of the sample. The sensing surface was washed with Millipore   water a   er the removal of each sample for the Fig. 2  Stepwise preparation of the sensing probe. Fig. 3  Schematic of the experimental set-up for the characterization of thetetracycline sensors. Fig. 4  Surface plasmon resonance spectra for di ff erent concentrations of (a)tetracycline samples over the TC-MIP probe, (b) oxy-tetracycline samples over theTC-MIP probe and (c) tetracycline samples over the NIP probe. This journal is  ª  The Royal Society of Chemistry 2013  Analyst  , 2013,  138 , 7254 – 7263 |  7257 Paper Analyst  unbinding of TC molecules from the TC-MIP probe, to avoid theinterference of le   -over samples and for the renewal of theprobe for the next sample. The shi    in the SPR spectra occursdue to the binding of TC molecules in the nano-cavities on theTC-MIP matrix. When the TC sample enters the vicinity of thesensing surface, it causes a change in the volume of the MIPmatrix, which results in the change in the dielectric propertiesof the sensing surface. Hence the change in the e ff  ectiverefractive index of the sensing surface causes the red shi   of theSPR curves with the increasing concentration of the TCsamples. Another reason for the shi    of the SPR curves is theinteraction of TC molecules with the MIP matrix, and theformation of hydrogen bonds between the polymer matrix andTC. There are several hydroxyl groups in TC molecules whichallow the formation of hydrogen bonds between carboxylic acidgroups of the polymer matrix and the hydroxyl group in TC.Hence, owing to the shape, size and functional groups of thebinding sites in the MIP matrix, only speci  c target moleculescan bind to the nano-cavities present in the MIP matrix, andthis causes the change in the volume and morphology of thesensing surface which results in the red shi   of the SPR curves.To check the selectivity of the TC-MIP probe, we recordedSPR spectra for the OTC sample solutions of the same concen-tration range (0.0  m M to 0.96  m M) over the TC-MIP probe. TheSPR spectra of all 12 OTC samples are shown in Fig. 4(b). It canbe seen that all of the SPR curves almost overlap with eachother, and no shi    in the resonance dip occurs. This impliesthat OTC molecules are not compatible with the nano-cavitieson the TC-MIP probe. In other words, no binding occursbetween the polymer matrix of TC-MIP and the OTC samples.Hence the TC-MIP probe has a high a ffi nity particularly for TCsamples, as it has binding sites with complementary shape, sizeand interaction sites only to TC molecules. Additionally, theperformance of the non-molecularly imprinted (NIP) (no TCadded to the master solution while preparing the polymeriza-tion matrix) probe was also investigated for the di ff  erent concentration samples of TC. Fig. 4(c) shows the SPR spectra forall 12 TC samples over the NIP probe. It can be seen fromFig. 4(c) that, similar to Fig. 4(b), a negligible shi    in the SPR curves occurs. Therefore, it can be concluded that there is nointeraction taking place between the TC molecules and thepolymerization matrix because there are no binding sites/nano-cavities formed in the NIP probe, as the target molecule was not added in the polymerization medium. Hence there should not be a shi    in the SPR curves, and we obtained the expectedresults in the control experiments.Fig. 5 shows the variation of the shi    in the resonance wavelength with the concentration of the TC and OTC samples.Solid squares, circles and triangles represent the resonance wavelength shi    obtained from the SPR curves shown inFig. 4(a) for TC samples over the TC-MIP probe, in Fig. 4(b) forOTC samples over the TC-MIP probe and in Fig. 4(c) for TCsamples over the NIP probe, respectively. The solid lines drawnfor each set of experimental points are the best    ts. In the caseof the TC-MIP probe and TC samples, an exponential curve   tsthe data points well. No modelling has been carried out for theexponential  t to the data points. The error bars were calculatedby taking into account the measurement accuracies of theapparatus (spectrometer, weighing balance, volumetriccylinder, pipette) used for the characterization of the sensor. It is observedthat the shi   inthe resonance wavelength fortheTCsamples over the TC-MIP probe increases rapidly for the lowerconcentration TC samples, but a   erwards it increases slowly and almost saturates at higher concentrations of tetracycline.Further, it is found that the total shi    in the resonance wave-length is 35.892 nm when the TC-MIP probe is dipped in the TCsamples over the 0.0  m M to 0.96  m M concentration range.Moreover, in the data represented by the circles in Fig. 5,obtained when the TC-MIP probe was dipped in OTC samples with the same concentrations, some   uctuations and random variations in the shi    in resonance wavelength are observed.Similar is the case for the data represented by the triangles inFig. 5, when the NIP probewas soakedinsamples withthe sameconcentrations of TC. From all the  gures and the discussion, it is concluded that the nano-cavities in the TC-MIP probe operateonly for TC and not for OTC, while the NIP probe does not respond to TC. This is good support for the selectivity of themolecular imprinting probe (TC-MIP) and the recognitioncapabilities of the target molecules.To extend the study on the selectivity of the TC-MIP probe, weperformed control experiments on the TC-MIP probe fordissimilar analytes. For this we chose urea and vitamin B 3  as thedissimilar analytes because they do not belong to the tetracyclinegroup. The SPR spectra of 12 samples of urea over the concen-tration range of 0.0  m M to 0.96 over the TC-MIP probe are plottedin Fig. 6(a). It can be seen from the   gure that the SPR spectraalmost overlap and there appears to be no shi    of the SPR dips.This implies that the TC-MIP probe does not give any signi  cant signal for urea samples, as urea is a foreign molecule for thenano-cavities/binding sites on the TC-MIP probe. Similar exper-iments were performed using vitamin B 3  samples of varying concentrations over the TC-MIP probe. The corresponding SPR spectra for 12 samples of vitamin B 3  of di ff  erent concentrationsovertherange0.0 m Mto0.96 m MareshowninFig.6(b).SimilartoFig. 6(a), there appears to be no shi   of the SPR dips in Fig. 6(b), Fig.5  Variation of the shift in resonance wavelength with the concentrations ofthe tetracycline and oxy-tetracycline samples. 7258  |  Analyst  , 2013,  138 , 7254 – 7263 This journal is  ª  The Royal Society of Chemistry 2013 Analyst Paper 
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!