Layer-by-Layer Quantum Dot Constructs Using Self-Assembly Methods

Oct 11, 2010 - Sakandar Rauf,*,†,‡ Andrew Glidle,† and Jonathan M. Cooper*,†. †Department of Electronics and Electrical Engineering, Oakfiel...
0 downloads 0 Views 2MB Size
Download PDF
pubs.acs.org/Langmuir © 2010 American Chemical Society

Layer-by-Layer Quantum Dot Constructs Using Self-Assembly Methods Sakandar Rauf,*,†,‡ Andrew Glidle,† and Jonathan M. Cooper*,† †

Department of Electronics and Electrical Engineering, Oakfield Avenue, University of Glasgow, UK G12 8LT, and ‡Australian Institute for Bioengineering and Nanotechnology (AIBN), Corner College and Cooper Roads (Bldg 75), The University of Queensland, Brisbane Qld 4072, Australia Received August 25, 2010. Revised Manuscript Received September 23, 2010

We describe the creation of CdSe/ZnS quantum dot assemblies using layer-by-layer construction strategies, using selfassembly. In the first approach, a dithiol linker was used to make multilayers of CdSe/ZnS quantum dots, while in the second biotin- and streptavidin-conjugated CdSe/ZnS quantum dots were used to make multilayer constructs. Both the chemical bonding nature and fluorescence spectroscopic properties of quantum dot films were characterized using X-ray photoelectron spectroscopy (XPS) and fluorescence spectroscopy.

Introduction Quantum dots are highly luminescent and monodisperse nanocrystals (CdSe/ZnS or CdTe, for example) that are currently of great interest especially due to their use as labels in bioanalytical applications.1 Quantum dots are often composed of atoms from group II-VI or III-V elements in the periodic table, and are defined as particles with physical dimensions smaller than the exciton Bohr radius.2 In general, they offer good photostability, high fluorescence intensity, and broad tunability that make these an excellent choice as a chromophore.3 Different methods of synthesis of quantum dot nanocrystals have been reported, but the formation of quantum dots in colloidal form via solution chemistry is an easy and successful route to realize quantum dots.4,5 Quantum dot spectral responses and emission wavelengths are dependent on the size of the quantum dots.6,7 The immobilization of these semiconductor nanoparticles is of importance in the fabrication of a variety of optical, bioelectronic, and biosensing devices. To date, several techniques to make nanocomposite ultrathin films of quantum dots or nanoparticles have been described, which include spin-casting,8 Langmuir-Blodgett deposition,9 and layer-by-layer assembly.3,10,11 The effect of different substrates (glass, silicon, PDMS, and metals) on the photoluminescence properties of CdSe/ZnS quantum dots has also been explored.3 For example, it has been shown that encapsulation of quantum dots into layer-by-layer polymer films on the substrate surfaces resulted in the enhanced photoemission of quantum dots.3 Similarly, it has been shown using silane chemistries *To whom correspondence should be addressed. E-mail: biosensor_nibge@ yahoo.com (S.R.); [email protected]. (J.M.C.). (1) Nann, T. Chem. Commun. 2005, 1735–1736. (2) Chan, W. C. W.; Maxwell, D. J.; Gao, X. H.; Bailey, R. E.; Han, M. Y.; Nie, S. M. Curr. Opin. Biotechnol. 2002, 13, 40–46. (3) Zimnitsky, D.; Jiang, C. Y.; Xu, J.; Lin, Z. Q.; Tsukruk, V. V. Langmuir 2007, 23, 4509–4515. (4) Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc. 1993, 115, 8706–8715. (5) Hines, M. A.; Guyot-Sionnest, P. J. Phys. Chem. 1996, 100, 468–471. (6) Gill, R.; Zayats, M.; Willner, I. Angew. Chem., Int. Ed. 2008, 47, 7602–7625. (7) Alivisatos, A. P. Science 1996, 271, 933–937. (8) Konstantatos, G.; Sargent, E. H. Appl. Phys. Lett. 2007, 91. (9) Gole, A.; Jana, N. R.; Selvan, S. T.; Ying, J. Y. Langmuir 2008, 24, 8181– 8186. (10) Halaoui, L. I. Langmuir 2001, 17, 7130–7136. (11) Zimnitsky, D.; Jiang, C. Y.; Xu, J.; Lin, Z. Q.; Zhang, L.; Tsukruk, V. V. Langmuir 2007, 23, 10176–10183.

16934 DOI: 10.1021/la103385s

with both Fmoc- and t-butyl protection groups on aluminum oxide membranes provided a method to simultaneously immobilize enzymes or antibodies with the quantum dots.12 Multicolor-multilayer films of CdSe/CdS and CdTe quantum dots have also been prepared on indium tin oxide (ITO) substrates using layer-by-layer assembly of poly(diallyldimethylammonium chloride) (PDDA), citrate stabilized CdSe/CdS and mercaptopropionic acid stabilized CdTe quantum dots.13 It was found that the fluorescence intensities and quantum yield can be tuned by selecting different color quantum dots and mainly by controlling the fluorescence resonance energy transfer (FRET) between different size quantum dots.13 Different color quantum dot films were also fabricated using layer-by-layer assembly of positively charged polyelectrolytes (PDDA, polyallylamine hydrochloride (PAH)), quaternary ammonium ion cationic starch (QAICS), and negatively charged quantum dots.14 Other strategies have been used to prepare layer-by-layer assemblies of quantum dots using different polyelectrolytes and quantum dots.15-17 These multilayer films were characterized using fluorescence spectroscopy, atomic force microscopy (AFM), X-ray diffraction, and surface plasmon resonance (SPR).15-17 In other studies, monolayers of CdTe quantum dots have been covalently immobilized using poly(acrylic acid) brushes and characterized using atomic force microscopy, fluorescence spectroscopy, and X-ray photoelectron spectroscopy (XPS).18 However, it is not obvious that the relatively harsh reaction conditions of the method described lends itself to the preparation of welldefined multilayer assemblies, as we describe below. While other different strategies have been used to make multilayers of quantum dots to date, these investigations have not included detailed characterization of the chemical bonding (12) Hobler, C.; Bakowsky, U.; Keusgen, M. Phys. Status Solidi A 2010, 207, 872–877. (13) Lin, Y. W.; Tseng, W. L.; Chang, H. T. Adv. Mater. 2006, 18, 1381. (14) Zhang, J.; Li, Q.; Di, X. W.; Liu, Z. L.; Xu, G. Nanotechnology 2008, 19. (15) Zucolotto, V.; Gattas-Asfura, K. M.; Tumolo, T.; Perinotto, A. C.; Antunes, P. A.; Constantino, C. J. L.; Baptista, M. S.; Leblanc, R. M.; Oliveira, O. N. Appl. Surf. Sci. 2005, 246, 397–402. (16) Kotov, N. A.; Dekany, I.; Fendler, J. H. J. Phys. Chem. 1995, 99, 13065– 13069. (17) Zhou, D. J.; Bruckbauer, A.; Abell, C.; Klenerman, D.; Kang, D. J. Adv. Mater. 2005, 17, 1243. (18) Gupta, S.; Uhlmann, P.; Agrawal, M.; Lesnyak, V.; Gaponik, N.; Simon, F.; Stamm, M.; Eychmuller, A. J. Mater. Chem. 2008, 18, 214–220.

Published on Web 10/11/2010

Langmuir 2010, 26(22), 16934–16940

Rauf et al.

Article

interactions between quantum dots and the functional groups that bind them into the multilayer assembly. As stated, here we describe the immobilization of CdSe/ZnS quantum dots using two different strategies and mild reaction conditions in a layerby-layer configuration. We then go on to characterize these assemblies using XPS and fluorescence spectroscopy, as a prelude to using these methods to construct barcoded quantum dot structures. To the best of our knowledge, these are the first XPS and fluorescence studies that describe the chemical linkage of quantum dots in a layer-by-layer assembly prepared using dithiol assembly and biological self-assembly.

Scheme 1. Schematic Layout of the Steps Followed for the Preparation of Quantum Dots 626 nm (QD626) Using Layer-by-Layer Assembly onto the Glass Substratea

Materials and Methods Materials. Quantum dots (CdSe/ZnS) dispersed in toluene, having emission maximum wavelengths at 626 nm (QD626) and 555 nm (QD555), were purchased from Evident Technologies. Streptavidin- and biotin-conjugated CdSe/ZnS quantum dots (QD655-biotin, QD655-streptavidin, and QD525-streptavidin) were purchased from Invitrogen, U.K. (comprising a semiconductor nanocrystal of CdSe around which a semiconductor shell of ZnS is coated, to improve the optical properties). These core-shell quantum dots are coated with a polymer shell and directly coupled to streptavidin or biotin molecules using a poly(ethylene glycol) (PEG) linker. The size of the quantum dot streptavidin conjugate is ∼15-20 nm, and that of biotin conjugate is ∼1012 nm. The loading of streptavidin onto the quantum dots is estimated as 5-10 streptavidin/QD conjugate, while in the case of biotin conjugate it is 5-7 biotin molecules/QD conjugate.19,20 (3-Mercaptopropyl)trimethoxysilane (MPTS), (3-aminopropyl)triethoxysilane (APTES), biotinamidohexanoic acid 3-sulfo-Nhydroxysuccinimide ester sodium salt (NHS-biotin), a dithiol linker 1,9-nonanedithiol, and toluene were purchased from SigmaAldrich. Absolute ethanol was purchased from Fisher Scientific. Plain microscope glass slides, used for the preparation of layer-bylayer assembly of quantum dots, were purchased from MenzelGlaser and Thermo Scientific. Decon 90, a surface active cleaning agent, was purchased from Decon laboratories limited, U.K. Methods. Dithiol Assembly. Glass slides were first cut into 1  1 cm2 pieces and cleaned by sonication in 5% vol/vol solution of Decon 90 in deionized water for 15 min. They were then rinsed with plenty of deionized water and dried with nitrogen gas. These glass slides were modified with MPTS to introduce thiol groups on the glass surface.21 Briefly, cleaned glass slides were placed in a wide-neck round-bottom flask equipped with a condenser, and 400 mL of isopropyl alcohol (IPA) along with 10 mL of water were added. To this mixture, 10 mL of MPTS was added and refluxed for 10 min in the boiling silane solution. After this, the flask was removed and allowed to cool. Glass slides were then rinsed with IPA and dried with nitrogen gas (Scheme 1). The same procedure was repeated three times. After modification with MPTS, the glass slides were immediately incubated in a 2 μM solution of quantum dots 626 nm (stock solution of quantum dots was 14.0 μM) in toluene for 1 h at room temperature. After incubation, the slides were then rinsed exhaustively with toluene and dried with nitrogen gas. For the preparation of multilayers of quantum dots, the glass slides were subsequently incubated in a solution of dithiol linker22 as shown in Scheme 1. The glass slides were incubated in a 10 mM ethanolic solution of 1,9-nonanedithiol for 1 h. After incubation, the glass slides were rinsed with ethanol and dried with nitrogen gas. These slides were then incubated in a quantum dot solution (2 μM) for 1 h to make the second layer of quantum dots. (19) Qdot Streptavidin conjugates. http://probes.invitrogen.com/media/pis/ mp19000.pdf, September 23, 2010. (20) Qdot biotin conjugates. http://probes.invitrogen.com/media/pis/mp19003. pdf, September 23, 2010. (21) Goss, C. A.; Charych, D. H.; Majda, M. Anal. Chem. 1991, 63, 85–88. (22) Pacifico, J.; Jasieniak, J.; Gomez, D. E.; Mulvaney, P. Small 2006, 2, 199– 203.

Langmuir 2010, 26(22), 16934–16940

a For the preparation of mixed multilayers of quantum dots, QD626 and QD555 were immobilized in an alternate fashion. Scheme is not to scale. Shown is an idealized scheme for the assembly process.

Scheme 2. Schematic Representation of Immobilization of QD525-Streptavidin (S525) Conjugate and QD655-Biotin Conjugate (B655) Using Layer-by-Layer Biological Self-Assemblya

a Scheme is not to scale. Shown is an idealized scheme for the assembly process.

By repeating the incubation in 1,9-nonanedithiol and quantum dot solutions, multilayers of quantum dots were formed. In these studies, 10 layers of quantum dots were formed on the glass slides using QD626. Alternatively, two different color quantum dots QD626 and QD555 were also used to make sandwich (layer-bylayer) assemblies using the same method as described above. Biological Self-Assembly. Streptavidin and biotin interactions were used to immobilize quantum dots onto a planar glass surface. A layer-by-layer assembly approach was used to construct an assembly of the quantum dots on to the glass substrate surface (Scheme 2). The glass slides were first cleaned by sonication in 5% vol/vol solution of Decon 90 for 15 min and then rinsed exhaustively with water and finally dried with nitrogen gas. To form an amino-terminated layer on the glass slides (Scheme 2), the substrates were immersed in a 3 vol% solution of 3-(aminopropyl)triethoxysilane (APTES) in 95% ethanol for 2 h and thoroughly rinsed with ethanol and dried with nitrogen gas. The samples were further dried in an oven at 120 °C for 0.5 h. After this, the samples were incubated in a solution of NHS-biotin (1 mg/mL in 10 mM phosphate buffer pH 7.4) for 2 h. After incubation in NHS-biotin, samples were rinsed with buffer and incubated in 0.15% solution of bovine serum albumin (BSA) for 0.5 h. This step was used to block the nonspecific adsorption of quantum dot conjugates. Samples were then incubated in 10 nM solution of QD525-streptavidin conjugate (S525) or QD655-streptavidin for 1 h. After washing with buffer, samples DOI: 10.1021/la103385s

16935

Article

Rauf et al.

Figure 1. Comparison of S(2p) high resolution XPS spectra of (a) a layer of MPTS on glass; (b) immobilization of the first layer of quantum dots onto MPTS modified glass substrate; (c) immobilization of dithiol linker on the monolayer of quantum dots immobilized by using MPTS; and (d) immobilization of the second layer of quantum dots onto dithiol modified substrate. were incubated in 10 nM solution of QD655-biotin (B655) solution for 1 h to make two layers of quantum dots as shown in Scheme 2. In this way, 2, 4, 6, 8, and 10 layers of QD-conjugates were prepared. X-ray Photoelectron Spectroscopy. XPS or electron spectroscopy for chemical analysis (ESCA) was used to characterize the immobilization of CdSe/ZnS quantum dots onto the glass substrate. All the samples for XPS analysis were prepared on ∼1  1 cm2 glass slides and stored in a vacuum desiccator prior to analysis. All the XPS spectra were taken using the high resolution Scienta ESCA 300 spectrometer at Daresbury laboratories, U.K. The Al KR radiation is provided with a rotating anode source, and it was focused on the samples at a take-off angle (TOA) of 45° after passing through the monochromators.23,24 To study the immobilization of CdSe/ZnS quantum dots on the glass substrates using dithiol chemistry, in addition to the survey spectra for each sample, C(1s), N(1s), S(2p), Cd(3d), and Zn(2p) high resolution XPS spectra were also measured. Similarly, for the immobilization of CdSe/ZnS quantum dots using biotin-streptvidin interaction, C(1s), N(1s), Cd(3d), and Zn(2p) high resolution XPS spectra were measured in addition to the survey scan for each sample. The slit width (0.8 mm) and channel resolution (0.05 eV) were kept constant in all the samples. All spectra were adjusted for an offset due to the flood gun by assigning 285 eV to the main C-H peak in the C(1s) spectrum as a reference.24 The deconvolution of the S(2p) peaks was performed using CaSaXPS software, Casa Software Limited. When deconvoluting the S(2p) peaks, different sulfur species are present as doublets. The doublet appears as a consequence of the known 2:1 ratio of S(2p1/2) and S(2P3/2) spin orbit states.25 The known binding energy difference between these spin orbit states is 1.18 eV.25 These values are used as constrained input parameters to the software during the deconvolution of the S(2p) peaks. (23) Jiang, L.; Glidle, A.; Griffith, A.; McNeil, C. J.; Cooper, J. M. Bioelectrochem. Bioenerg. 1997, 42, 15–23. (24) Glidle, A.; Yasukawa, T.; Hadyoon, C. S.; Anicet, N.; Matsue, T.; Nomura, M.; Cooper, J. M. Anal. Chem. 2003, 75, 2559–2570. (25) Romano, E. J.; Schulz, K. H. Appl. Surf. Sci. 2005, 246, 262–270.

16936 DOI: 10.1021/la103385s

Fluorescence Spectroscopy Measurements. For fluorescence measurements, a Nikon microscope (Microphot-SA) (NA = 0.85, 40 objective) was coupled with a Triax 320 spectrometer (Jobin Yvon) equipped with a symphony CCD detector (Figure S1, Supporting Information). A UV-2A filter set was used to excite the quantum dots in the UV-region using a mercury lamp and to collect the fluorescent signal. The UV-2A filter set consists of an excitation filter (330-380 nm), a dichroic mirror (cutoff wavelength 400 nm), and a 420 nm long-pass emission filter. Figure S1 in the Supporting Information shows the experimental setup with a variable aperture. When the aperture is fully closed, it produces a spot size of ∼50 μm diameter of excitation light. For the measurement of fluorescence signal from the quantum dots immobilized on the glass substrates, the aperture was closed and quantum dots were excited using filter set UV-2A. Spectra were collected with 0.1 s acquisition times. In the case of each layer assembly, five measurements were taken (where n is the number of measurements taken on one sample at different sampling sites).

Results and Discussion Quantum dots were immobilized onto glass slides using two different strategies. The first strategy involves dithiol chemistry to immobilize quantum dots as shown in Scheme 1. The immobilization schemes were characterized using XPS and fluorescence spectroscopy as explained below. Layer-by-Layer Assembly of Quantum Dots onto Glass Substrates using Dithiol Chemistry. For the attachment of quantum dots with the thiol (-SH) group from MPTS, high resolution XPS scans of S(2p) were collected as described in the Methods section. Results are presented in Figure 1. It can be seen that two S(2p) peaks are present (Figure 1a). The main peak at 163.6 eV showed the presence of sulfur species from MPTS, while the lesser peak at 168.5 eV showed the presence of oxidation of thiol (-SH) groups.26 The deconvolution of these peaks revealed (26) Laibinis, P. E.; Whitesides, G. M.; Allara, D. L.; Tao, Y. T.; Parikh, A. N.; Nuzzo, R. G. J. Am. Chem. Soc. 1991, 113, 7152–7167.

Langmuir 2010, 26(22), 16934–16940

Rauf et al.

the presence of two different species in the form of a pair of doublets at 162.3 and 163.6 eV. Each doublet was assigned after the deconvolution due to the known 2:1 ratio of S(2p1/2) and S(2p3/2) spin orbit states and by providing the value of their energy difference of 1.18 eV25 in the software. The doublet at 162.3 eV corresponds to S-H, and the doublet at 163.6 eV indicates the presence of the (S-S) group.27 After the immobilization of first layer of quantum dots (Figure 1b), the deconvolution of the S(2p) peaks indicated the presence of a doublet at 161.6 eV that can be assigned to the presence of S-Zn bonds28 in addition to the S-S doublet 163.6 eV and oxidized sulfur at 168.5 eV. Similar results have been reported in the case of silver deposition to the mercaptosilane layer.26,29 After the deposition of dithiol linker on the monolayer of quantum dots, the same peaks were apparent with a greater contribution of the S-S doublet (compared to S-Zn), clearly indicating the attachment of (-SH) to the quantum dots and the presence of free -SH groups (Figure 1c). When the second layer of quantum dots was immobilized, the deconvolution of the S(2p) peak revealed the same set of doublets with an increased contribution from the S-Zn doublet (Figure 1d), indicating the bonding of the second layer of quantum dots (S-Zn bonds). These studies support the proposed Scheme 1 and show that the CdSe/ZnS quantum dots immobilized via layer-by-layer assembly using thiol chemistry. Moreover, the survey scans of all four samples confirmed the immobilization of CdSe/ZnS quantum dots (Figure S2, Supporting Information). To confirm the immobilization of quantum dots due to the presence of thiol groups on the glass substrate, control experiments were carried out where the glass substrate, without prior modification using mercaptosilane, was incubated in a solution of quantum dots. Figure S3a in the Supporting Information shows the high resolution XPS spectra of Zn(2p) for a monolayer of quantum dots immobilized using MPTS and for the control, without using MPTS (involving adsorption only). It can be seen that the signal from physioadsorbed quantum dots is 5 times smaller when compared to that obtained when immobilization is carried out using MPTS. Similar results were found when high resolution XPS spectra of Cd(3d) were compared for both samples (Figure S3b, Supporting Information). Fluorescence spectroscopy was performed to characterize the optical properties of thin films of quantum dots prepared via layer-by-layer assembly of quantum dots using a dithiol linker. Figure 2 shows the fluorescence intensity plotted against the number of layers of quantum dots. It can be seen that as the number of layers of quantum dots increases, as expected the fluorescence intensity also increases (second order polynomial, R = 0.989). However, the increase in fluorescence does not follow a straight line, a phenomena which may be explained by the so-called inner filter effect,30 resulting in the absorption of the incident or excitation light before it reaches the point in the sample at which luminescence is observed and/or reabsorption of some of the emitted light, resulting in decrease in fluorescence intensity. It was also noted that, as the number of layers increased, the standard deviation of fluorescence measurements also increased, indicating that the quantum dots may cluster and the (27) Siegbahn, K.; Nordling, C.; Fahlman, A.; Nordbert, R.; Hamrin, K.; Hedman, J.; Johansson, G.; Bergmark, T.; Karlsson, S. E.; Lindgren, I.; Linberg, B. ESCA-Atomic, Molecular and solid state structure studied by means of electron spectroscopy; Almqvist & Wiksells: Uppsala, Sweden, 1967. (28) Lang, P.; Nogues, C. Surf. Sci. 2008, 602, 2137–2147. (29) Liu, Z. C.; He, Q. G.; Xiao, P. F.; Liang, B.; Tan, J. X.; He, N. Y.; Lu, Z. H. Mater. Chem. Phys. 2003, 82, 301–305. (30) Kubista, M.; Sjoback, R.; Eriksson, S.; Albinsson, B. Analyst 1994, 119, 417–419.

Langmuir 2010, 26(22), 16934–16940

Article

Figure 2. Layer-by-layer assemblies of CdSe/ZnS quantum dots (QD626) onto glass substrate: (a) Fluorescence spectra of different layers of quantum dots onto the glass substrate and (b) layer number vs fluorescence intensity. The data points are the mean measurements, and the error bars show the standard deviation for n = 5 measurements in case of each layer assembly and R = 0.989, where n represents the number of measurements from a single sample in the case of each layer assembly.

surface roughness increases as the number of layers increased. This is a well-known effect observed during the layer-by-layer assembly of quantum dots which was characterized by other groups using AFM studies.14,15,31,32 Multicolor quantum dot multilayers were also prepared using the same experimental procedure by immobilizing quantum dot 626 nm (QD626) and quantum dot 555 nm (QD555) in an alternate fashion (Figure 3). Note that QD555 was always the upper quantum dot layer in these assemblies and two layers refer to a first QD626 layer and second QD555 layer. Fluorescence emission spectra of these multilayers showed that signals from both classes of quantum dots were increased as the number of layers increases. However, again, and as expected, the increase in fluorescence is not linear, which we again attribute to the inner filter effect. Particular to the layer of smaller quantum dots QD555, the increase in fluorescence intensity is not linear when compared to the case of the larger quantum dots (Figure 3b), which may be due to the FRET from the smaller QD555 to the larger QD626 as studied in more detail by a group examining layer-by-layer assembly of smaller and larger quantum dots.33 Layer-by-Layer Assembly of Quantum Dot-Biotin and Quantum Dot-Streptavidin Conjugates onto Glass Substrates. The second strategy involved the immobilization of quantum dots using biotin-streptavidin interaction as shown in Scheme 2. In order to overcome the problem of the inner filter effect, symptomatic of high density packing of quantum dots afforded in Scheme 1, quantum dot-streptavidin and quantum (31) Wang, X. H.; Du, Y. M.; Ding, S.; Wang, Q. Q.; Xiong, G. G.; Xie, M.; Shen, X. C.; Pang, D. W. J. Phys. Chem. B 2006, 110, 1566–1570. (32) Neves, M. C.; Pereira, A. S.; Peres, M.; Kholkin, A.; Monteiro, T.; Trindade, T. Mater. Sci. Forum 2006, 514-516, 1111–1115. (33) Franzl, T.; Koktysh, D. S.; Klar, T. A.; Rogach, A. L.; Feldmann, J.; Gaponik, N. Appl. Phys. Lett. 2004, 84, 2904–2906.

DOI: 10.1021/la103385s

16937

Article

dot-biotin conjugates were investigated. We wanted to investigate whether these quantum dot-biomolecular conjugates can provide sufficient spacing to overcome the inner filter effect.

Figure 3. Layer-by-layer assemblies of QD626 and QD555 onto glass substrates: (a) Fluorescence spectra of different layers of quantum dots onto the glass substrate. Note that QD555 is the outer layer in all the layer assemblies. (b) Layer number vs fluorescence intensity. The data points are the mean measurements, and the error bars show the standard deviation for n = 5 measurements in the case of each layer assembly, where n represents the number of measurements from a single sample in the case of each layer assembly.

Rauf et al.

The bioconjugates were immobilized on the glass substrate using layer-by-layer assembly of biotin-streptavidin interaction and characterized using XPS and fluorescence spectroscopy. Figure 4A shows the N(1s) high resolution XPS spectrum of aminosilane on the glass substrate. The N(1s) spectrum confirmed the presence of two different nitrogen species occurring at (399.5 and 401.3 eV) two different binding energies separated by 1.8-2 eV,34 confirming the immobilization of aminosilane on the glass substrate. Those components of the spectra recorded at lower binding energies can be assigned to free aliphatic amino groups, while those at higher binding energies can be assigned to the protonated aliphatic amino groups.34 For the immobilization of streptavidin-conjugated quantum dots, glass slides were incubated in a solution of NHS-Biotin for one hour at room temperature, as described in the Methods section. Figure 4 shows the N(1s) and C(1s) spectra both before (a) and after (b) immobilization of biotin onto the aminosilane modified glass slides. Before immobilization, the deconvolution of the C(1s) spectrum shows the presence of two types of carbon species (C-C and C-N), while, following the immobilization of biotin, one additional peak appeared at 288 eV which corresponded to the CdO group from biotin (thereby confirming the attachment of biotin on the glass substrate).35,36 To immobilize streptavidin-conjugated quantum dots, samples were then incubated in a streptavidin-conjugated quantum dot (S525) solution in PBS pH 7.4 containing 0.15% BSA for 1 h to block nonspecific adsorption of quantum dot conjugates. These substrates were then washed with quantum dot free PBS pH 7.4 containing 0.15% BSA. After the immobilization of streptavidinconjugated quantum dots, the substrate was incubated in a biotinconjugated quantum dot (B655) solution for 1 h. The sample was then washed exhaustively to remove unbound biotin-quantum dots. A control sample was prepared by the incubation of aminosilane modified glass substrate in PBS pH 7.4 containing 0.15% BSA for 0.5 h, and then it was incubated in streptavidin-conjugated

Figure 4. (A) N(1s) high resolution XPS spectra of aminosilane film on the glass substrate (a) before and (b) after the immobilization of NHSbiotin. (B) C(1s) high resolution XPS spectra (a) before and (b) after immobilization of NHS-biotin. 16938 DOI: 10.1021/la103385s

Langmuir 2010, 26(22), 16934–16940

Rauf et al.

Figure 5. (a) Comparison of Cd(3d) high resolution XPS spectra of quantum dots immobilized with or without using biotinstreptavidin interaction. (b) Comparison of Zn(2p) high resolution XPS spectra of quantum dots immobilized with or without using biotin-streptavidin interaction.

quantum dot solution (QD525-streptavidin) for 1 h. After washing, the samples were dried under nitrogen gas. To confirm that the immobilization of quantum dot conjugates is via biotin-streptavidin interaction, high resolution XPS spectra of Cd(3d) and Zn(2p) were measured for the first layer and a control sample. Figure 5 shows that streptavidin quantum dot conjugates were not immobilized on the control sample, confirming that the immobilization of quantum dot conjugates was via biotin-streptavidin interaction rather than the adsorption of quantum dot conjugates and multilayers of quantum dot conjugates can be formed on the glass substrate using this layer-by-layer assembly approach. Also the survey spectra of four different samples prepared for the immobilization of quantum dots using streptavidin and biotin conjugates confirmed the presence of CdSe/ZnS quantum dots (Figure S4, Supporting Information). Optical properties of thin films of quantum dot-streptavidin and quantum dot-biotin conjugates were also studied using fluorescence spectroscopy. In the first experiment, 10 layers of quantum dot-streptavidin 655 nm (S655) and quantum dot-biotin 655 nm (B655) were prepared using layer-by-layer biological selfassembly of these conjugates on the glass substrate. Figure 6 shows the fluorescence spectra of these thin films up to 10 layers. It can be seen that the fluorescence intensity increased with the increase in number of layers of quantum dot conjugates (R = 0.988). However, and in contrast to studies involving quantum dots immobilized using thiol chemistry, Figure 6 indicates that as (34) Bierbaum, K.; Kinzler, M.; Woll, C.; Grunze, M.; Hahner, G.; Heid, S.; Effenberger, F. Langmuir 1995, 11, 512–518. (35) Zeng, D. W.; Yung, K. C. Appl. Surf. Sci. 2001, 180, 280–285. (36) Yam, C. M.; Pradier, C. M.; Salmain, M.; Marcus, P.; Jaouen, G. J. Colloid Interface Sci. 2001, 235, 183–189.

Langmuir 2010, 26(22), 16934–16940

Article

Figure 6. Layer-by-layer assemblies of quantum dot-streptavidin 655 nm (S655) and quantum-dot biotin 655 nm (B655) conjugates onto glass substrate: (a) Fluorescence spectra of different layers of quantum dots onto the glass substrate; (b) layer number vs fuorescence intensity. The data points are the mean measurements, and the error bars show the standard deviation for n = 5 measurements in the case of each layer assembly and R = 0.988, where n represents the number of measurements from a single sample in the case of each layer assembly.

Figure 7. Layer-by-layer assemblies of quantum dot-streptavidin 525 nm (S525) and quantum-dot biotin 655 nm (B655) conjugates onto glass substrate: (a) fluorescence spectra of different layers of quantum dots onto the glass substrate; (b) layer number vs fluorescence intensity. The data points are the mean measurements, and the error bars show the standard deviation for n = 5 measurements in case of each layer assembly, where n represents the number of measurements from a single sample in the case of each layer assembly. The inset shows more detail of the data from S525. DOI: 10.1021/la103385s

16939

Article

the number of layers increased the standard deviation also increased, which may be due to the quantum dot clustering and an increase in surface roughness as the number of layers increased.14,15,31,32 Similarly, multicolor multilayers of quantum dot-streptavidin 525 nm (S525) and quantum dot-biotin 655 nm (B655) conjugates were prepared. Figure 7 shows that the fluorescence intensity of both the quantum dots increases with the increase in number of layers. However, the fluorescence intensity of B655 is higher than that of the S525 quantum dots. The inset of Figure 7b shows that the standard deviation in fluorescence intensity is higher and increases with the increase in number of layers. Again in these multilayer quantum dot assemblies, the standard deviation is higher than the multicolor multilayers of quantum dots produced using a dithiol linker (Figure 3). This increase in standard deviation of fluorescence measurements may be due to the fact that streptavidin or biotin conjugates have between 5 and 10 streptavidin or biotin molecules on each conjugate, and this number varies from conjugate to conjugate. As the streptavidin and biotin conjugate thin films are not as uniform as the thin films of quantum dots formed using a dithiol linker, this may result in quantum dot clustering and an increase in surface roughness as the number of layers increases compared to the multilayers of quantum dots produced using dithiol linker. Despite the fact that the standard deviation for these quantum dot assemblies is higher, the XPS and fluorescence measurements of these multilayer assemblies show that the streptavidin and biotin conjugates of quantum dots can be used to produce multilayers of quantum dots. Formation of multilayers of quantum dots using biotin-streptavidin interaction minimizes the inner filter effect which is observed in the case of immobilization using thiol chemistry. However, the standard deviation in the fluorescence measurements in the the case of biotin-streptavidin (37) Rauf, S.; Glidle, A.; Cooper, J. M. Adv. Mater. 2009, 21, 4020. (38) Rauf, S.; Glidle, A.; Cooper, J. M. Chem. Commun. 2010, 46, 2814–2816.

16940 DOI: 10.1021/la103385s

Rauf et al.

interaction is larger when compared to the quantum dot assemblies formed using thiol chemistry.

Conclusions We characterize the mechanism of a layer-by-layer assembly of quantum dots onto glass substrate using XPS for two different strategies based upon chemical (Scheme 1) and biological (Scheme 2) self-assembly. We also studied the fluorescence properties of these films explaining the advantages and disadvantages associated with both strategies. These multilayer assemblies can be used to prepare different multifunctional fluorescence biosensing interfaces. For example, we have reported on the production of barcoded beads based on multilayer assemblies of biotin- and streptavidin-conjugated quantum dots using layer-by-layer biological self-assembly.37,38 We have also studied the layer-by-layer assembly of quantum dots on magnetic beads using a dithiol linker. It was found that, in the case of dithiol assembly, the inner layer signal was blocked by the upper layer due to the inner filter effect or FRET (unpublished data) and it was not possible to prepare different barcoded beads using a dithiol linker. However, the inner filter effect and FRET are significantly less in the case of biological self-assembly, and different barcoded beads were prepared using layer-by-layer assembly.37 The characterization of the constructs, described in this paper, has underpinned this work, enabling us to developed multiplexed bead based fluorescence biosensing for immunosensing.38 Acknowledgment. We are thankful for the financial support of Dorothy Hodgkin’s Postgraduate Award (DHPA) scheme for funding Ph.D. studies of S.R. We thank NCESS, Daresbury Laboratory, U.K. for XPS time. Supporting Information Available: Fluorescence measurement setup; XPS data for thiol and biological self-assembly methods. This material is available free of charge via the Internet at http://pubs.acs.org.

Langmuir 2010, 26(22), 16934–16940