Journal of Colloid and Interface Science 426 (2014) 117–123
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Facile routes of manufacturing silicon quantum dots on a silicon wafer and their surface activation by esters of N-hydroxysuccinimide Xiang Liu ⇑, Heming Cheng, Tiantian Zhao, Changchang Zhang School of Chemistry and Chemical Engineering, Anhui University of Technology, Maanshan 243002, China
a r t i c l e
i n f o
Article history: Received 4 January 2014 Accepted 2 April 2014 Available online 12 April 2014 Keywords: Silicon quantum dots Porous silicon Chemical modifications N-hydroxysuccinimide Fluorescence
a b s t r a c t Fluorescent silicon quantum dots (SiQDs) could be prepared by reduction of hydrogen silsesquioxane, etching of silicon powers with wetting chemistry techniques or electrolysis of a wafer catalyzed by polyoxometalates. Chemical modifications are indispensable for the stability of the SiQDs photoluminescence and wider applications of SiQDs. Facile routes of manufacturing SiQDs derived from a silicon wafer and its surface functionalization by N-hydroxysuccinimide (NHS) esters were described in this work in detail. Firstly, the porous silicon chip was prepared by nanosilver-assisted electroless chemical etching. Then the chip was etched successively with hydrofluoric acid/nitric acid solutions until it emitted dazzling red fluorescence which claimed the achieved SiQDs on silicon substrates (SiQDs/Si). Finally, surface NHS esters were fabricated on such an SiQDs/Si chipthrough stepwise modifications, which were tested by the amidation between the NHS esters and n-octylamine. The fluorescence emission of the SiQDs/Si chip almost remained unchanged during the successively chemical modifications, which indicated the SiQDs had capabilities of enduring the sustained high temperature and organic media. Meanwhile, the SiQDs did not leave from the silicon substrate during the surface tuning. The SiQDs obtained by ultrasonication of an SiQDs/Si chip in water were investigated by transmission electron and atomic force microscopies. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Silicon quantum dots1 arouse extensive interests in recent years. Compared with QDs of heavy metals (e.g., CdSe), SiQDs possess some unique advantages such as biocompatibility [1] and intrinsic low toxicity [2]. Therefore, SiQDs have potential applications in luminescent materials preparation, biology and biomedicine of in vivo fluorescence imaging. Since the optical properties of SiQDs depend on their sizes, many previous efforts have been paid to invent methods of preparing silicon nanocrystals or SiQDs. For example, nanocrystalline of SiASiO2 [3–5] could be obtained by heating hydrogen silsesquioxanes to a predetermined temperature in a high-temperature furnace with an inert atmosphere and were annealed for 1 h in a 4% H2 and 96% N2 atmosphere. SiQDs could also be generated in inverse micelles by reducing silicon tetrachloride with a strong reductant of lithium aluminum hydride (LiAlH4) [6,7]. Silicon
⇑ Corresponding author. 1
E-mail address: [email protected] (X. Liu). SiQDs.
http://dx.doi.org/10.1016/j.jcis.2014.04.007 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
tetrabromide was also chosen as a precursor to synthesize the SiQDs by mixing it with inert Ar and H2. Such a mixture was introduced into a plasma chamber at a defined pressure to accomplish the formation of SiQDs [8]. As we all know that such silicon sources are tough to be coped with since they hydrolyze readily in moist atmosphere and what is more the synthetic procedures are rather complex. Isolating the modified SiQDs from the solutions is still a complex operation since they are so tiny. Especially, the loss will increase with the increased process steps such as their surface modifications. Electrochemistry synthesis catalyzed by polyoxometalates was a novel approach to obtain tunable sizes of SiQDs which derived from a wafer [9–13]. This approach did not require the tedious operations and the harsh reaction conditions. Electrolytic current and ingredients of the electrolyte were the two main process parameters. Silicon powders were utilized as another cheap silicon source to synthesize SiQDs using wetting etching chemistry [14]. Different sizes of SiQDs were achieved by controlling the etching time and the proportion of hydrofluoric acid (HF) and nitric acid (HNO3). Nevertheless, further applications of such SiQDs need proper surface modifications. For example, the up-conversion property of silicon nanospheres were derived from the interaction between surface carboxyl groups and surface amine groups of SiQDs [12]. High specific
118
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123
Scheme 1. Schematic descriptions about the chemical modifications on an SiQDs/Si chip (both sides of the silicon chip were covered by the achieved SiQDs actually). The porous silicon (PSi) chip was etched firstly by HF/HNO3 to prepare SiQDs/Si chip, shown in (1). Then the SiQDs/Si chip with surface SiAH species (denoted with H-SiQDs/Si) was obtained by HF etching (2). Successively the chip involved the reaction (3) with 10-undecenoic acid (UA) to obtain surface-carboxylated SiQDs/Si (abbreviated with UA-SiQDs/Si). The surface ACOOH groups of UA-SiQDs/Si were activated by NHS (denoted with NHS-UA-SiQDs/Si) and thus the n-octylamine (OA) molecules could be anchored onto SiQDs/Si surfaces (denoted with OA-UA-SiQDs/Si), which shown in (4) and (5) respectively. Ultimately, such a SiQDs/Si chip covered by hydrocarbon chains was ultrasonically disposed to obtain aqueous SiQDs solutions with a photoluminescence (PL) emission of 596 nm.
surface areas of SiQDs may bring plentiful surface SiAH species when etched with HF or ammonium fluoride solution [15]. Some unsaturated organic compounds with terminal C@C could combine with such silicon hydrides [2,11,16–18]. Surface SiAH species may therefore widen the utilizations of SiQDs. Herein, we will report a convenient route of manufacturing the SiQDs on silicon substrate2 and of stepwise modifications on them to construct the interface of N-hydroxysuccinimide3 esters. Surface activation by NHS esters was a conventional strategy which applied in surface bonding of some biomolecules to prepare biochemical assays or biochips [15,19]. When carboxyl groups were introduced onto surface of silicon wafer [20], gold [21] or titanium [22], they are easily esterified by NHS under extremely mild conditions. The resultant NHS esters were active enough to react with primary amino groups of biomolecules by virtue of amidation at room temperature. In our work, n-octylamine is chosen as a model compound to demonstrate the activities of such surface NHS esters. The fluorescent stabilities of SiQDs/Si during the modifications were also estimated. The whole procedure is depicted schematically in Scheme 1.
2. Experimental 2.1. Materials Hydrogen peroxide (H2O2, 30%), HF (40%), HNO3 (67%), silver nitrate (AgNO3, 99.8%), tetrahydrofuran4 (P99%), UA (95%), NHS (99%), N,N0 -dicyclohexylcarbodiimide5 (P99%), and n-octylamine6 (99%) were purchased from Aladdin Reagent (Shanghai) Company. All the reagents were used as received without any further purification. Double distilled water was used throughout the work. The silicon chips were obtained by slicing a wafer (Si(1 1 1), thickness: 450 ± 50 lm, p-type, boron doped, electrical resistivity: 8–13 X cm) into sizes of 10 mm 5 mm. Before use, they were cleaned thoroughly in piranha solutions (H2O2 (30 wt% in H2O)/H2SO4 (98 wt%) 1:3 (v:v)) at 150 °C for more than 2 h to remove any fouling, as was described in our previous report [23], and were washed with copious water and dried with N2.
4 2 3
SiQDs/Si. NHS.
5 6
THF. DCC. OA.
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123
2.2. Preparation of an SiQDs/Si chip A cleaned silicon chip was immersed in a mixture containing 5.0 M HF and 5 mM AgNO3 to deposit silver nanoparticles for 20 s [24]. Successively, it was put into a mixture of 5.0 M HF and 0.6 M H2O2 at 50 °C for 1 h to prepare a PSi chip, as was described elsewhere [25]. The PSi chip was subsequently immersed in 7.2 M HNO3 solutions to remove any silver particles. Finally, the PSi chip was further etched by a mixture of 3.6 M HNO3 and 2.5 M HF at 50 °C for 60–90 s to obtain an SiQDs/Si chip, as described in (1) of Scheme 1. The etching reaction was extraordinarily violent that the solutions were entirely agitated by the resultant gases. During this period, etching time was vital for the formation of SiQDs. Any delay would bring about the dissolving of the produced SiQDs. Finally, the achieved SiQDs/Si chip was washed with copious water to wipe out any absorbed inorganic ions and was irradiated with a portable ultraviolet lamp (365 nm) to judge the PL.
119
spectroscopies (KRATOS AMICUS). The XP spectra were recorded by K-Alpha with a monochromatized Al Ka X-ray source. Survey scans over a binding energy of 0–1350 eV was executed for elemental information. High-resolution scans (step = 0.05 eV) of several elements such as C, O and Si were run for obtaining the changes of their binding energies13 to explain the stepwise chemical modifications. The spectra were analyzed with nonlinear least-square fitting using different weighting of Gaussian–Lorentzian functions. The C 1s, O 1s and Si 2p core-level spectra were deconvoluted to explain the surface ingredients. The PL of SiQDs/Si, UA-SiQDs/Si and NHS-SiQDs/Si chips were demonstrated by photographing in a darkroom under irradiation of an ultraviolet lamp (365 nm) and emission spectra executed by fluorescence spectrophotometer (LS 45, PerkinElmer) to confirm that such chemical modifications did not quench the PL properties of the SiQDs.
3. Results and discussion 2.3. Chemical modifications on an SiQDs/Si chip 3.1. Manufacture of an SiQDs/Si chip The dense surface SiAHx (x = 1, 2 or 3) [26] were readily obtained when the SiQDs/Si chip was placed in 0.25 M HF solutions within a short period of 1 min, as is shown in (2) of Scheme 1. The achieved SiQDs enhanced surface areas of the chip which accelerated the reaction between surface Si and HF. According to our observations, more concentrated HF solutions or prolonged etching time would weak the photoluminescence of the SiQDs/Si chip obviously. The obtained HASiQDs/Si chip was washed with water and dried with flowing nitrogen. Then it was put into neat 5 mL UA immediately. The reaction system was airtight with an oil-sealed device and was flown with nitrogen for at least 10 min to vent the air before heating. Such hydrosilylation reaction between surface SiAHx and UA proceeded at 140 °C for 4 h to obtain the UA-SiQDs/Si chip, described in (3) of Scheme 1. Successively, the obtained UA-SiQDs/Si chip was immersed in THF solutions of 0.2 g NHS and 0.1 g DCC at 40 °C for 4 h to achieve surface NHS esters (NHS-UA-SiQDs/Si), which described in (4) of Scheme 1. At last, the NHS-UA-SiQDs/Si chip was placed in 20 mL aqueous solutions containing 20 lL OA at room temperature for 2 h to estimate the activity of surface NHS esters depicted in (5) of Scheme 1. Since the stepwise reactions took place on the surface of silicon substrate, the excessive materials at the end of the reaction could be wiped out easily by several times of rinsing or soaking. Generally, it was rather convenient to carry out chemical modifications on such an SiQDs/Si chip [27]. Finally, ultrasonic processing in aqueous media gave the hydrocarbon chains-tagged SiQDs with a PL emission at 596 nm as is shown in Scheme 1. 2.4. Characterization techniques The morphology changes from the PSi to SiQDs/Si were observed by scanning electron microscopy7 (Nano SEM 430, FEI Company) and were demonstrated by X-ray powder diffraction8 (XPert Pro MPD, PANalytical B.V.). The SiQDs were illuminated by transmission electron microscope9 (Tecnai G2 F20 S-TWIN, FEI Company) and atomic force microscopy10 with tapping mode using silicon nitride tips in air at room temperature and analyzed using the Nanoscope software (Veeco Instruments Inc.). The stepwise modifications on an SiQDs/Si chip were judged by measurements of infrared11 (Nicolet 380, Thermo Nicolet Corporation) and X-ray photoelectron12 7 8 9 10 11 12
SEM. XRD. TEM. AFM. IR. XP.
The PSi chip was prepared according to the mechanism of galvanic displacement reaction [28,29]. And its morphology is exhibited in SEM image of Fig. 1(a). There are a lot of pores distributed on PSi chip. Etching reaction catalyzed by nanosilver leads to an increases in both the surface areas and surface silicon atoms. In light of the mechanism of galvanic displacement reaction, lengthening etching time only results in the more deep pores which benefits for the formation of silicon nanowires [25]. The nanosilver on silicon surface accelerates the decomposition of H2O2 to produce oxygen. Thus the silicon atoms near silver nanoparticles will be oxidized and successively be consumed by HF. Then the nanosilver settles gradually as the reaction goes on until the reaction ends [25]. That is to say, SiQDs cannot be generated in this case. Our experiments prove that it is the successive etching with HF/HNO3 etchants brings SiQDs into being. The etchants permeated into the pores of the PSi chip giving a wealth of tiny blocks on their surfaces shown in Fig. 1(b). In my opinion, the different crystalline planes may have different reaction rates which contribute forming tiny blocks on surface of PSi. Such etching reactions on PSi surfaces take place violently with a great deal of liberated gases. In the course of etching progress, the surface Si atoms are oxidized into SiO2 by HNO3 and the SiO2 are removed by HF instantly which described in the chemical equations of 3Si + 4HNO3 ? 3SiO2 + 4NO + 2H2O and SiO2 + 6HF ? H2SiF6 + 2H2O. The overall reaction can be described in the equation of 3Si + 18HF + 4HNO3 ? 3H2SiF6 + 4NO + 8H2O [14,30]. Therefore, the etching time is a key factor since too much etching time will make the resulted SiQDs leave from the silicon substrate while too short time will not in favor to the generation of SiQDs. XRD patterns indicate the magnificent difference shown in Fig. 1(c). SiQDs/Si chip gives a more intensive diffraction at (1 1 1) crystal plane exhibited in (1) than that in (2) of PSi chip even the intensity is multiplied by ten. The chip emitted a dazzling red fluorescence when irradiated by an ultraviolet lamp of 365 nm wavelength. This is a simple way to judge the PL property of the chip. Ultrasonic treatment of an SiQDs/Si chip will lead to the shift of the surface SiQDs from silicon substrate to aqueous medium. The SiQDs solution displays the red fluorescence when illuminated by an ultraviolet lamp, which shown in the inset of Fig. 1(d). The PL spectra of Fig. 1(d) confirm that SiQDs give PL emission at 600 nm while the PSi does not, which exhibited in (1) and (2) respectively. Moreover, despite we postpone the supersonic time the SiQDs/Si chip still emits a strong red fluorescence. This 13
BE.
120
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123
Fig. 1. SEM images of PSi and SiQDs/Si chips corresponding to (a) and (b), respectively. XRD patterns of the PSi and SiQDs/Si chips are shown in (c) corresponding to (1) and (2) respectively. PL spectra of the ultrasonic solutions of PSi and SiQDs/Si chips corresponding to (1) and (2) respectively are displayed in (d). The PL of the SiQDs solution is seen in the insert photograph of (d).
exclaims that the SiQDs on silicon substrate are rigid enough to sustain the following chemical modifications. The SiQDs can be removed from the silicon substrate by an ultrasonic treatment, which demonstrated by the TEM image of Fig. 2(a). It illuminates that the size of the SiQDs is about 3–4 nm. The top insert of Fig. 2(a), obtained by DigitalMicrograph software (http://www.gatan.com/) through drawing a square on a lattice fringe and its fast Fourier transform, demonstrates the d-spacing of 0.31 nm which matches the (1 1 1) plane of crystalline Si. AFM image of Fig. 2(b) also exhibits the homogeneous size distribution of the SiQDs. The corresponding depth histogram of Fig. 2(c) and section analysis of 2(d) show that the heights of the SiQDs are centered at about 3.5 nm which is consistent with the TEM observation.
3.2. Chemical modifications on an SiQDs/Si chip Surface SiAHx species were readily generated [31] and coupled with vinyl groups of some unsaturated compounds by covalent bond of SiAC such as UA. Heating [31], illumination under visible light [32] or microwave irradiation [33] would promote such hydrosilylations. The obtained UA-SiQDs/Si chip is investigated by both the IR and XP spectra, which shown in Figs. 3(a) and 4(a), respectively. Band 1716 cm1 is assigned to stretching vibration of C@O in carboxyl groups [15,34] shown in Fig. 3(a). Peaks of 2916 and 2850 cm1 are ascribed to asymmetrical and symmetrical stretching vibrations of CAH, respectively, in methylene. However, not all the Si-Hx species involve the hydrosilylation reaction, which is confirmed by the remaining absorptions between 2043
Fig. 2. TEM and AFM images of the SiQDs corresponding to (a) and (b), respectively. A fast Fourier transform of a lattice fringe is shown in the top insert of (a). The depth histogram and the section analysis of the SiQDs are displayed respectively in (c) and (d).
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123
Fig. 3. IR spectra of (a), (b) and (c) corresponding to the chips of UA-SiQDs/Si, NHSUA-SiQDs/Si and OA-UA-SiQDs/Si respectively. The spectra ranges marked with a shadow area are ascribed to surface SiAHx species.
and 2306 cm1. Some SiAHx sites may not be wetted by the added UA. This is due to the steric hindrance [31,35] of the SiQDs/Si chip. XP spectra with wide scan of UA-SiQDs/Si chip demonstrate the elemental components shown in Fig. 4(a). High-resolution scan of C 1s, O 1s and Si 2p on UA-SiQDs/Si witnesses the chemical attachment of UA which shown in Fig. 5(a–c) respectively. The C 1s core-level spectrum can be curve-fitted with two peak compo-
121
nents having BE at about 284.8 and 289.4 eV, attributable to the CAC/CAH and O@CAO [27,36] respectively shown in Fig. 5(a). The corresponding integrated areas (10:1) are in well accordance with the theoretically atomic components of UA. Fig. 5(b) displays the BE of O 1s of 532.3 eV. The high-resolution of Si 2p core-level spectrum in Fig. 5(c) can be deconvoluted into four components with BE of 102.7, 101.3, 100.3 and 99.4 eV. Some surface silicon atoms are oxidized giving SiAO peak of 102.7 eV while SiAC locates at 99.4 eV [37] which confirms the existence of chemical bond of surface SiAC. According to the aforementioned analyses of IR, not all the SiAHx species involve the hydrosilylation. The BE at 101.3 and 100.3 eV should arisen from @SiAH2 and „SiAH respectively [38]. The terminal ACOOH can be converted into NHS esters catalyzed by DCC at room temperature [39]. The resultant surface NHS esters are demonstrated by the complete disappearance of peak 1716 cm1 shown in Fig. 3(a) and appearance of newly tripartite peaks shown in Fig. 3(b). Band 1818, 1787 and 1738 cm1 in Fig. 3(b) are assigned respectively to stretching vibration of C@O (v(C@O)) in ester, asymmetrical as well as symmetrical v(C@O) in the NHS imide moiety [15]. The XP spectrum of insert of Fig. 4(b) also illuminates the existence of elemental N on NHS-UA-SiQDs/ Si which should be arisen from the esterification between surface ACOOH and NHS. NHS esters are regarded as fruitful groups in activating surface carboxyl groups of several solid substrates. Surface NHS esters had
Fig. 4. XP spectra with wide scans of UA-SiQDs/Si, NHS-UA-SiQDs/Si and OA-UA-SiQDs/Si chips corresponding to (a), (b) and (c) respectively. The high-resolution scan of N 1s core-level spectra of NHS-UA-SiQDs/Si is shown in the inset of (b).
Fig. 5. XP spectra with high-resolution scans of C 1s, O 1s and Si 2p of UA-SiQDs/Si and of OA-UA-SiQDs/Si corresponding to (a–f) respectively.
122
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123
Fig. 6. The PL spectra of (a), (b) and (c) corresponding to the ultrasonic solutions of UA-SiQDs/Si, NHS-UA-SiQDs/Si and OA-UA-SiQDs/Si chips respectively (excitation wavelength: 300 nm). The top insets of (a–c) show the photographs of the corresponding chips obtained by irradiation of an ultraviolet lamp (365 nm) in a dark room. The left inset of (c) is the photo of the ultrasonic solution of OA-UA-SiQDs/Si chip.
been utilized extensively as efficient linkages of conjugating organic or biological molecules bearing primary amines to a solid surface via amidation [19,40]. In this work, we choose OA as a model compound to demonstrate the activity of the NHS esters on surface of the SiQDS/Si chip. The primary amino groups of OA react with the NHS esters bringing about the almost complete disappearance of the tripartite peaks which is shown in Fig. 3(c). The newly occurred peaks of 1643 and 1549 cm1 are assigned to amide I (v(C@O)) and II [34] of OA-UA-SiQDs/Si respectively, which reveals the covalent bonding between OA and SiQDs/Si chip. XP spectrum with wide scan on OA-UA-SiQDs/Si shown in Fig. 4(c) demonstrates the surface chemical ingredients. High-resolution scans of C 1s, O 1s and Si 2p are illuminated in Fig. 5(d–f) respectively. Compared with Fig. 5(a) of UA-SiQDs/Si, Fig. 5(d) of OA-UA-SiQDs/Si changes obviously. Component of O@CAO shown in Fig. 5(a) vanishes completely now. BE of 284.8 eV still belongs to C 1s of CAC/CAH species. The two fitted peaks at 286.4 and 288.2 eV should be assigned to NACAC and O@CAN, respectively [34]. The C 1s integrated areas of CAC/CAH, NACAC and O@CAN are roughly in proportion to 17:1:1 which agrees well with the theoretically atomic ratio of OA-UA-SiQDs/Si. These prove the amidation between surface NHS esters and OA. Fig. 5(e) exhibits that O 1s core-level spectrum of OA-UA-SiQDs/Si are curve-fitted with two peak components. BE of C@O locates at 532.1 eV while SiAO shifts to 529.9 eV which testifies the surface oxidation of SiQDs. Si 2p of SiAO at 102.7 eV shown in Fig. 5(f) also demonstrates the inevitable oxidation of SiQDs. Strangely, Si 2p of SiAC gives a comparatively weak intensity at BE of 99.4 eV. It may be derived from that the resultantly long hydrocarbon chains cover the most part of the SiQDs surface. This case is similar to the silicon surface grafted polymer brushes which even gives no any signal of Si 2p [41]. Importantly, the PL intensity of SiQDs/Si almost does not decline with such successive modifications which can be observed clearly in the insert photograph of Fig. 6(a–c). This proves that the SiQDs on silicon substrate do not detach from the substrate during the chemical modifications. The PL of the modified SiQDs/ Si chips are extremely stable in atmosphere for at least two months which proved by the insert images of Fig. 6. Surface chemical tuning may prevent the surface oxidation which in favor to the stability of the PL [42]. The UA-SiQDs/Si, NHS-UA-SiQDs/Si and OAUA-SiQDs/Si chips are ultrasonically disposed and the aqueous solutions give the PL emission peaks at 601, 605 and 596 nm respectively shown in Fig. 6(a–c). The lower insert of Fig. 6(c) shows that the OA-UA-SiQDs solution emits a dazzling red fluorescence clearly which agrees well with the PL measurement. This
forecasts that biomolecules with primary amino groups may be anchored by such NHS-modified SiQDs similarly to OA molecules to realize fluorescent labeling of the biomolecules. Despite the hydrocarbon chains, OA-UA-SiQDs can be dispersed in aqueous media which should be attributable to the partly oxidized surface of SiQDs by surrounding oxygen which is demonstrated by the XP spectra shown in Fig. 5(e) and (f). As is known to all, the surface silicon dioxides are hydrophilic. We consider that this method can avoid the complicate operations of isolating the products from the mixtures since the excessive reactants can be remove by several times of rinsing or washing when the reaction finishes. Final ultrasonication of the labeled SiQDs/Si chip realizes the separation of the functionalized SiQDs from Si substrate very easily. Accordingly, the techniques described in this work may simplify the labeling of biomolecules with SiQDs. 4. Conclusion SiQDs can be manufactured from cheap silicon chips. Readily prepared SiAHx species make it possible to tailor surface SiQDs via the hydrosilylation. The vast varieties of organic compounds with vinyl groups facilitate such surface tuning of SiQDs/Si. The surface SiQDs on silicon substrate are sustainable for properly chemical modifications such as the surface carboxylation with UA and subsequent esterification with NHS; meanwhile, the PL does not vanish. Additionally, it is quite easy to isolate the excessive reactants from surface of SiQDs/Si chip. The modified SiQDs can be separated from the substrate via a supersonic treatment. There is no doubt that these approaches facilitate the manufacture of SiQDs. The biomolecules with primary amino groups are likely to be attached to such NHS-functionalized SiQDs to realize their fluorescent labeling by means of such facile routes. Additionally, since the SiQDs/Si chips can suffer from heating and organic media, they have the promise to be used as an efficient catalyst in organic syntheses [10]. Acknowledgments This work was supported by Educational Commission of Anhui Province of China (KJ2012A051) and Student Innovation Training Program of AHUT (AH201310360273). References [1] J.H. Ahire, Q. Wang, P.R. Coxon, G. Malhotra, R. Brydson, R. Chen, Y. Chao, ACS Appl. Mater. Interfaces 4 (2012) 3285–3292.
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123 [2] F. Erogbogbo, K. Yong, I. Roy, R. Hu, W. Law, W. Zhao, H. Ding, F. Wu, R. Kumar, M.T. Swihart, P.N. Prasad, ACS Nano 5 (2011) 413–423. [3] C.M. Hessel, E.J. Henderson, J.G.C. Veinot, Chem. Mater. 18 (2006) 6139–6146. [4] Z. Yang, A.R. Dobbie, K. Cui, J.G.C. Veinot, J. Am. Chem. Soc. 134 (2012) 13958– 13961. [5] S. Regli, J.A. Kelly, A.M. Shukaliak, J.G.C. Veinot, J. Phys. Chem. Lett. 3 (2012) 1793–1797. [6] M.J.L. Portolés, R.P. Diez, M.L.D. Arciprete, P. Caregnato, J.J. Romero, D.O. Mártire, O. Azzaroni, M. Ceolín, M.C. Gonzalez, J. Phys. Chem. C 116 (2012) 11315–11325. [7] R.D. Tilley, J.H. Warner, K. Yamamoto, I. Matsui, H. Fujimori, Chem. Commun. (2005) 1833–1835. [8] S. Ohta, P. Shen, S. Inasawa, Y. Yamaguchi, J. Mater. Chem. 22 (2012) 10631– 10638. [9] Z. Kang, C.H.A. Tsang, Z. Zhang, M. Zhang, N. Wong, J.A. Zapien, Y. Shan, S. Lee, J. Am. Chem. Soc. 129 (2007) 5326–5327. [10] Z. Kang, C.H.A. Tsang, N. Wong, Z. Zhang, S. Lee, J. Am. Chem. Soc. 129 (2007) 12090–12091. [11] Y. He, Y. Su, X. Yang, Z. Kang, T. Xu, R. Zhang, C. Fan, S. Lee, J. Am. Chem. Soc. 131 (2009) 4434–4438. [12] Y. Zhang, X. Han, J. Zhang, Y. Liu, H. Huang, H. Ming, S. Lee, Z. Kang, Nanoscale 4 (2012) 7760–7765. [13] Y. Zhang, X. Han, R. Liu, Y. Liu, H. Huang, J. Zhang, H. Yu, Z. Kang, J. Phys. Chem. C 116 (2012) 20363–20367. [14] K. Sato, H. Tsuji, K. Hirakuri, N. Fukata, Y. Yamauchi, Chem. Commun. (2009) 3759–3761. [15] X. Liu, H. Han, H. Liu, S. Xiao, Surf. Sci. 604 (2010) 1315–1319. [16] P.K. Sudeep, T. Emrick, ACS Nano 3 (2009) 4105–4109. [17] J.A. Kelly, A.M. Shukaliak, M.D. Fleischauer, J.G.C. Veinot, J. Am. Chem. Soc. 133 (2011) 9564–9571. [18] J. Wang, Y. Liu, F. Peng, C. Chen, Y. He, H. Ma, L. Cao, S. Sun, Small 8 (2012) 2430–2435. [19] H. Asanuma, H. Noguchi, K. Uosaki, H. Yu, J. Am. Chem. Soc. 130 (2008) 8016– 8022. [20] H. Kitano, H. Suzuki, K. Matsuura, K. Ohno, Langmuir 26 (2010) 6767–6774.
123
[21] J.D. Dunn, E.A. Igrisan, A.M. Palumbo, G.E. Reid, M.L. Bruening, Anal. Chem. 80 (2008) 5727–5735. [22] S. Xiao, M. Textor, N.D. Spencer, Langmuir 14 (1998) 5507–5516. [23] X. Liu, H. Liu, P. Guo, S. Xiao, Phys. Status Solidi A 208 (2011) 1462–1470. [24] X. Liu, H. Cheng, P. Cui, Appl. Surf. Sci. 292 (2014) 695–701. [25] X. Zhong, Y. Qu, Y. Lin, L. Liao, X. Duan, ACS Appl. Mater. Interfaces 3 (2011) 261–270. [26] D. Gräf, M. Grundner, R. Schulz, J. Vac. Sci. Technol. A 7 (1989) 808–813. [27] X. Liu, H. Zheng, Y. Ma, Q. Yan, S. Xiao, J. Colloid Interface Sci. 358 (2011) 116– 122. [28] X. Li, K. Liang, B.K. Tay, E.H.T. Teo, Appl. Surf. Sci. 258 (2012) 6169–6176. [29] F. Miao, B. Tao, P.K. Chu, Dalton Trans. 41 (2012) 5055–5059. [30] M. Cao, X. Song, J. Zhai, J. Wang, Y. Wang, J. Phys. Chem. B 110 (2006) 13072– 13075. [31] S.A. Alekseev, V. Lysenko, V.N. Zaitsev, D. Barbier, J. Phys. Chem. C 111 (2007) 15217–15222. [32] A. Imanishi, S. Yamane, Y. Nakato, Langmuir 24 (2008) 10755–10761. [33] X. Liu, H. Zheng, Q. Yan, C. Wang, Y. Ma, Y. Tang, S. Xiao, Surf. Sci. 605 (2011) 1106–1113. [34] R. Voicu, R. Boukherroub, Vasiliki Bartzoka, T. Ward, J.T.C. Wojtyk, D.D.M. Wayner, Langmuir 20 (2004) 11713–11720. [35] A. Petit, M. Delmotte, A. Loupy, J. Chazalviel, F. Ozanam, R. Boukherroub, J. Phys. Chem. C 112 (2008) 16622–16628. [36] S.R. Puniredd, O. Assad, H. Haick, J. Am. Chem. Soc. 130 (2008) 13727–13734. [37] M. Rosso, A. Arafat, K. Schroën, M. Giesbers, C.S. Roper, R. Maboudian, H. Zuilhof, Langmuir 24 (2008) 4007–4012. [38] S. Mirabella, R. Agosta, G. Franzò, I. Crupi, M. Miritello, R.L. Savio, M.A. Di Stefano, S. Di Marco, F. Simone, A. Terrasi, J. Appl. Phys. 106 (2009) 103505. [39] D. Guo, S. Xiao, B. Xia, Shuai-Wei, J. Pei, Y. Pan, X. You, Z. Gu, Z. Lu, J. Phys. Chem. B 109 (2005) 20620–20628. [40] F. Rusmini, Z. Zhong, J. Feijen, Biomacromolecules 8 (2007) 1775–1789. [41] D. Xu, W.H. Yu, E.T. Kang, K.G. Neoh, J. Colloid Interface Sci. 279 (2004) 78–87. [42] Z. Kang, Y. Liu, C.H.A. Tsang, D.D.D. Ma, X. Fan, N. Wong, S. Lee, Adv. Mater. 21 (2009) 661–664.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Facile routes of manufacturing silicon quantum dots on a silicon wafer and their surface activation by esters of N-hydroxysuccinimide Xiang Liu ⇑, Heming Cheng, Tiantian Zhao, Changchang Zhang School of Chemistry and Chemical Engineering, Anhui University of Technology, Maanshan 243002, China
a r t i c l e
i n f o
Article history: Received 4 January 2014 Accepted 2 April 2014 Available online 12 April 2014 Keywords: Silicon quantum dots Porous silicon Chemical modifications N-hydroxysuccinimide Fluorescence
a b s t r a c t Fluorescent silicon quantum dots (SiQDs) could be prepared by reduction of hydrogen silsesquioxane, etching of silicon powers with wetting chemistry techniques or electrolysis of a wafer catalyzed by polyoxometalates. Chemical modifications are indispensable for the stability of the SiQDs photoluminescence and wider applications of SiQDs. Facile routes of manufacturing SiQDs derived from a silicon wafer and its surface functionalization by N-hydroxysuccinimide (NHS) esters were described in this work in detail. Firstly, the porous silicon chip was prepared by nanosilver-assisted electroless chemical etching. Then the chip was etched successively with hydrofluoric acid/nitric acid solutions until it emitted dazzling red fluorescence which claimed the achieved SiQDs on silicon substrates (SiQDs/Si). Finally, surface NHS esters were fabricated on such an SiQDs/Si chipthrough stepwise modifications, which were tested by the amidation between the NHS esters and n-octylamine. The fluorescence emission of the SiQDs/Si chip almost remained unchanged during the successively chemical modifications, which indicated the SiQDs had capabilities of enduring the sustained high temperature and organic media. Meanwhile, the SiQDs did not leave from the silicon substrate during the surface tuning. The SiQDs obtained by ultrasonication of an SiQDs/Si chip in water were investigated by transmission electron and atomic force microscopies. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Silicon quantum dots1 arouse extensive interests in recent years. Compared with QDs of heavy metals (e.g., CdSe), SiQDs possess some unique advantages such as biocompatibility [1] and intrinsic low toxicity [2]. Therefore, SiQDs have potential applications in luminescent materials preparation, biology and biomedicine of in vivo fluorescence imaging. Since the optical properties of SiQDs depend on their sizes, many previous efforts have been paid to invent methods of preparing silicon nanocrystals or SiQDs. For example, nanocrystalline of SiASiO2 [3–5] could be obtained by heating hydrogen silsesquioxanes to a predetermined temperature in a high-temperature furnace with an inert atmosphere and were annealed for 1 h in a 4% H2 and 96% N2 atmosphere. SiQDs could also be generated in inverse micelles by reducing silicon tetrachloride with a strong reductant of lithium aluminum hydride (LiAlH4) [6,7]. Silicon
⇑ Corresponding author. 1
E-mail address: [email protected] (X. Liu). SiQDs.
http://dx.doi.org/10.1016/j.jcis.2014.04.007 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
tetrabromide was also chosen as a precursor to synthesize the SiQDs by mixing it with inert Ar and H2. Such a mixture was introduced into a plasma chamber at a defined pressure to accomplish the formation of SiQDs [8]. As we all know that such silicon sources are tough to be coped with since they hydrolyze readily in moist atmosphere and what is more the synthetic procedures are rather complex. Isolating the modified SiQDs from the solutions is still a complex operation since they are so tiny. Especially, the loss will increase with the increased process steps such as their surface modifications. Electrochemistry synthesis catalyzed by polyoxometalates was a novel approach to obtain tunable sizes of SiQDs which derived from a wafer [9–13]. This approach did not require the tedious operations and the harsh reaction conditions. Electrolytic current and ingredients of the electrolyte were the two main process parameters. Silicon powders were utilized as another cheap silicon source to synthesize SiQDs using wetting etching chemistry [14]. Different sizes of SiQDs were achieved by controlling the etching time and the proportion of hydrofluoric acid (HF) and nitric acid (HNO3). Nevertheless, further applications of such SiQDs need proper surface modifications. For example, the up-conversion property of silicon nanospheres were derived from the interaction between surface carboxyl groups and surface amine groups of SiQDs [12]. High specific
118
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123
Scheme 1. Schematic descriptions about the chemical modifications on an SiQDs/Si chip (both sides of the silicon chip were covered by the achieved SiQDs actually). The porous silicon (PSi) chip was etched firstly by HF/HNO3 to prepare SiQDs/Si chip, shown in (1). Then the SiQDs/Si chip with surface SiAH species (denoted with H-SiQDs/Si) was obtained by HF etching (2). Successively the chip involved the reaction (3) with 10-undecenoic acid (UA) to obtain surface-carboxylated SiQDs/Si (abbreviated with UA-SiQDs/Si). The surface ACOOH groups of UA-SiQDs/Si were activated by NHS (denoted with NHS-UA-SiQDs/Si) and thus the n-octylamine (OA) molecules could be anchored onto SiQDs/Si surfaces (denoted with OA-UA-SiQDs/Si), which shown in (4) and (5) respectively. Ultimately, such a SiQDs/Si chip covered by hydrocarbon chains was ultrasonically disposed to obtain aqueous SiQDs solutions with a photoluminescence (PL) emission of 596 nm.
surface areas of SiQDs may bring plentiful surface SiAH species when etched with HF or ammonium fluoride solution [15]. Some unsaturated organic compounds with terminal C@C could combine with such silicon hydrides [2,11,16–18]. Surface SiAH species may therefore widen the utilizations of SiQDs. Herein, we will report a convenient route of manufacturing the SiQDs on silicon substrate2 and of stepwise modifications on them to construct the interface of N-hydroxysuccinimide3 esters. Surface activation by NHS esters was a conventional strategy which applied in surface bonding of some biomolecules to prepare biochemical assays or biochips [15,19]. When carboxyl groups were introduced onto surface of silicon wafer [20], gold [21] or titanium [22], they are easily esterified by NHS under extremely mild conditions. The resultant NHS esters were active enough to react with primary amino groups of biomolecules by virtue of amidation at room temperature. In our work, n-octylamine is chosen as a model compound to demonstrate the activities of such surface NHS esters. The fluorescent stabilities of SiQDs/Si during the modifications were also estimated. The whole procedure is depicted schematically in Scheme 1.
2. Experimental 2.1. Materials Hydrogen peroxide (H2O2, 30%), HF (40%), HNO3 (67%), silver nitrate (AgNO3, 99.8%), tetrahydrofuran4 (P99%), UA (95%), NHS (99%), N,N0 -dicyclohexylcarbodiimide5 (P99%), and n-octylamine6 (99%) were purchased from Aladdin Reagent (Shanghai) Company. All the reagents were used as received without any further purification. Double distilled water was used throughout the work. The silicon chips were obtained by slicing a wafer (Si(1 1 1), thickness: 450 ± 50 lm, p-type, boron doped, electrical resistivity: 8–13 X cm) into sizes of 10 mm 5 mm. Before use, they were cleaned thoroughly in piranha solutions (H2O2 (30 wt% in H2O)/H2SO4 (98 wt%) 1:3 (v:v)) at 150 °C for more than 2 h to remove any fouling, as was described in our previous report [23], and were washed with copious water and dried with N2.
4 2 3
SiQDs/Si. NHS.
5 6
THF. DCC. OA.
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123
2.2. Preparation of an SiQDs/Si chip A cleaned silicon chip was immersed in a mixture containing 5.0 M HF and 5 mM AgNO3 to deposit silver nanoparticles for 20 s [24]. Successively, it was put into a mixture of 5.0 M HF and 0.6 M H2O2 at 50 °C for 1 h to prepare a PSi chip, as was described elsewhere [25]. The PSi chip was subsequently immersed in 7.2 M HNO3 solutions to remove any silver particles. Finally, the PSi chip was further etched by a mixture of 3.6 M HNO3 and 2.5 M HF at 50 °C for 60–90 s to obtain an SiQDs/Si chip, as described in (1) of Scheme 1. The etching reaction was extraordinarily violent that the solutions were entirely agitated by the resultant gases. During this period, etching time was vital for the formation of SiQDs. Any delay would bring about the dissolving of the produced SiQDs. Finally, the achieved SiQDs/Si chip was washed with copious water to wipe out any absorbed inorganic ions and was irradiated with a portable ultraviolet lamp (365 nm) to judge the PL.
119
spectroscopies (KRATOS AMICUS). The XP spectra were recorded by K-Alpha with a monochromatized Al Ka X-ray source. Survey scans over a binding energy of 0–1350 eV was executed for elemental information. High-resolution scans (step = 0.05 eV) of several elements such as C, O and Si were run for obtaining the changes of their binding energies13 to explain the stepwise chemical modifications. The spectra were analyzed with nonlinear least-square fitting using different weighting of Gaussian–Lorentzian functions. The C 1s, O 1s and Si 2p core-level spectra were deconvoluted to explain the surface ingredients. The PL of SiQDs/Si, UA-SiQDs/Si and NHS-SiQDs/Si chips were demonstrated by photographing in a darkroom under irradiation of an ultraviolet lamp (365 nm) and emission spectra executed by fluorescence spectrophotometer (LS 45, PerkinElmer) to confirm that such chemical modifications did not quench the PL properties of the SiQDs.
3. Results and discussion 2.3. Chemical modifications on an SiQDs/Si chip 3.1. Manufacture of an SiQDs/Si chip The dense surface SiAHx (x = 1, 2 or 3) [26] were readily obtained when the SiQDs/Si chip was placed in 0.25 M HF solutions within a short period of 1 min, as is shown in (2) of Scheme 1. The achieved SiQDs enhanced surface areas of the chip which accelerated the reaction between surface Si and HF. According to our observations, more concentrated HF solutions or prolonged etching time would weak the photoluminescence of the SiQDs/Si chip obviously. The obtained HASiQDs/Si chip was washed with water and dried with flowing nitrogen. Then it was put into neat 5 mL UA immediately. The reaction system was airtight with an oil-sealed device and was flown with nitrogen for at least 10 min to vent the air before heating. Such hydrosilylation reaction between surface SiAHx and UA proceeded at 140 °C for 4 h to obtain the UA-SiQDs/Si chip, described in (3) of Scheme 1. Successively, the obtained UA-SiQDs/Si chip was immersed in THF solutions of 0.2 g NHS and 0.1 g DCC at 40 °C for 4 h to achieve surface NHS esters (NHS-UA-SiQDs/Si), which described in (4) of Scheme 1. At last, the NHS-UA-SiQDs/Si chip was placed in 20 mL aqueous solutions containing 20 lL OA at room temperature for 2 h to estimate the activity of surface NHS esters depicted in (5) of Scheme 1. Since the stepwise reactions took place on the surface of silicon substrate, the excessive materials at the end of the reaction could be wiped out easily by several times of rinsing or soaking. Generally, it was rather convenient to carry out chemical modifications on such an SiQDs/Si chip [27]. Finally, ultrasonic processing in aqueous media gave the hydrocarbon chains-tagged SiQDs with a PL emission at 596 nm as is shown in Scheme 1. 2.4. Characterization techniques The morphology changes from the PSi to SiQDs/Si were observed by scanning electron microscopy7 (Nano SEM 430, FEI Company) and were demonstrated by X-ray powder diffraction8 (XPert Pro MPD, PANalytical B.V.). The SiQDs were illuminated by transmission electron microscope9 (Tecnai G2 F20 S-TWIN, FEI Company) and atomic force microscopy10 with tapping mode using silicon nitride tips in air at room temperature and analyzed using the Nanoscope software (Veeco Instruments Inc.). The stepwise modifications on an SiQDs/Si chip were judged by measurements of infrared11 (Nicolet 380, Thermo Nicolet Corporation) and X-ray photoelectron12 7 8 9 10 11 12
SEM. XRD. TEM. AFM. IR. XP.
The PSi chip was prepared according to the mechanism of galvanic displacement reaction [28,29]. And its morphology is exhibited in SEM image of Fig. 1(a). There are a lot of pores distributed on PSi chip. Etching reaction catalyzed by nanosilver leads to an increases in both the surface areas and surface silicon atoms. In light of the mechanism of galvanic displacement reaction, lengthening etching time only results in the more deep pores which benefits for the formation of silicon nanowires [25]. The nanosilver on silicon surface accelerates the decomposition of H2O2 to produce oxygen. Thus the silicon atoms near silver nanoparticles will be oxidized and successively be consumed by HF. Then the nanosilver settles gradually as the reaction goes on until the reaction ends [25]. That is to say, SiQDs cannot be generated in this case. Our experiments prove that it is the successive etching with HF/HNO3 etchants brings SiQDs into being. The etchants permeated into the pores of the PSi chip giving a wealth of tiny blocks on their surfaces shown in Fig. 1(b). In my opinion, the different crystalline planes may have different reaction rates which contribute forming tiny blocks on surface of PSi. Such etching reactions on PSi surfaces take place violently with a great deal of liberated gases. In the course of etching progress, the surface Si atoms are oxidized into SiO2 by HNO3 and the SiO2 are removed by HF instantly which described in the chemical equations of 3Si + 4HNO3 ? 3SiO2 + 4NO + 2H2O and SiO2 + 6HF ? H2SiF6 + 2H2O. The overall reaction can be described in the equation of 3Si + 18HF + 4HNO3 ? 3H2SiF6 + 4NO + 8H2O [14,30]. Therefore, the etching time is a key factor since too much etching time will make the resulted SiQDs leave from the silicon substrate while too short time will not in favor to the generation of SiQDs. XRD patterns indicate the magnificent difference shown in Fig. 1(c). SiQDs/Si chip gives a more intensive diffraction at (1 1 1) crystal plane exhibited in (1) than that in (2) of PSi chip even the intensity is multiplied by ten. The chip emitted a dazzling red fluorescence when irradiated by an ultraviolet lamp of 365 nm wavelength. This is a simple way to judge the PL property of the chip. Ultrasonic treatment of an SiQDs/Si chip will lead to the shift of the surface SiQDs from silicon substrate to aqueous medium. The SiQDs solution displays the red fluorescence when illuminated by an ultraviolet lamp, which shown in the inset of Fig. 1(d). The PL spectra of Fig. 1(d) confirm that SiQDs give PL emission at 600 nm while the PSi does not, which exhibited in (1) and (2) respectively. Moreover, despite we postpone the supersonic time the SiQDs/Si chip still emits a strong red fluorescence. This 13
BE.
120
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123
Fig. 1. SEM images of PSi and SiQDs/Si chips corresponding to (a) and (b), respectively. XRD patterns of the PSi and SiQDs/Si chips are shown in (c) corresponding to (1) and (2) respectively. PL spectra of the ultrasonic solutions of PSi and SiQDs/Si chips corresponding to (1) and (2) respectively are displayed in (d). The PL of the SiQDs solution is seen in the insert photograph of (d).
exclaims that the SiQDs on silicon substrate are rigid enough to sustain the following chemical modifications. The SiQDs can be removed from the silicon substrate by an ultrasonic treatment, which demonstrated by the TEM image of Fig. 2(a). It illuminates that the size of the SiQDs is about 3–4 nm. The top insert of Fig. 2(a), obtained by DigitalMicrograph software (http://www.gatan.com/) through drawing a square on a lattice fringe and its fast Fourier transform, demonstrates the d-spacing of 0.31 nm which matches the (1 1 1) plane of crystalline Si. AFM image of Fig. 2(b) also exhibits the homogeneous size distribution of the SiQDs. The corresponding depth histogram of Fig. 2(c) and section analysis of 2(d) show that the heights of the SiQDs are centered at about 3.5 nm which is consistent with the TEM observation.
3.2. Chemical modifications on an SiQDs/Si chip Surface SiAHx species were readily generated [31] and coupled with vinyl groups of some unsaturated compounds by covalent bond of SiAC such as UA. Heating [31], illumination under visible light [32] or microwave irradiation [33] would promote such hydrosilylations. The obtained UA-SiQDs/Si chip is investigated by both the IR and XP spectra, which shown in Figs. 3(a) and 4(a), respectively. Band 1716 cm1 is assigned to stretching vibration of C@O in carboxyl groups [15,34] shown in Fig. 3(a). Peaks of 2916 and 2850 cm1 are ascribed to asymmetrical and symmetrical stretching vibrations of CAH, respectively, in methylene. However, not all the Si-Hx species involve the hydrosilylation reaction, which is confirmed by the remaining absorptions between 2043
Fig. 2. TEM and AFM images of the SiQDs corresponding to (a) and (b), respectively. A fast Fourier transform of a lattice fringe is shown in the top insert of (a). The depth histogram and the section analysis of the SiQDs are displayed respectively in (c) and (d).
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123
Fig. 3. IR spectra of (a), (b) and (c) corresponding to the chips of UA-SiQDs/Si, NHSUA-SiQDs/Si and OA-UA-SiQDs/Si respectively. The spectra ranges marked with a shadow area are ascribed to surface SiAHx species.
and 2306 cm1. Some SiAHx sites may not be wetted by the added UA. This is due to the steric hindrance [31,35] of the SiQDs/Si chip. XP spectra with wide scan of UA-SiQDs/Si chip demonstrate the elemental components shown in Fig. 4(a). High-resolution scan of C 1s, O 1s and Si 2p on UA-SiQDs/Si witnesses the chemical attachment of UA which shown in Fig. 5(a–c) respectively. The C 1s core-level spectrum can be curve-fitted with two peak compo-
121
nents having BE at about 284.8 and 289.4 eV, attributable to the CAC/CAH and O@CAO [27,36] respectively shown in Fig. 5(a). The corresponding integrated areas (10:1) are in well accordance with the theoretically atomic components of UA. Fig. 5(b) displays the BE of O 1s of 532.3 eV. The high-resolution of Si 2p core-level spectrum in Fig. 5(c) can be deconvoluted into four components with BE of 102.7, 101.3, 100.3 and 99.4 eV. Some surface silicon atoms are oxidized giving SiAO peak of 102.7 eV while SiAC locates at 99.4 eV [37] which confirms the existence of chemical bond of surface SiAC. According to the aforementioned analyses of IR, not all the SiAHx species involve the hydrosilylation. The BE at 101.3 and 100.3 eV should arisen from @SiAH2 and „SiAH respectively [38]. The terminal ACOOH can be converted into NHS esters catalyzed by DCC at room temperature [39]. The resultant surface NHS esters are demonstrated by the complete disappearance of peak 1716 cm1 shown in Fig. 3(a) and appearance of newly tripartite peaks shown in Fig. 3(b). Band 1818, 1787 and 1738 cm1 in Fig. 3(b) are assigned respectively to stretching vibration of C@O (v(C@O)) in ester, asymmetrical as well as symmetrical v(C@O) in the NHS imide moiety [15]. The XP spectrum of insert of Fig. 4(b) also illuminates the existence of elemental N on NHS-UA-SiQDs/ Si which should be arisen from the esterification between surface ACOOH and NHS. NHS esters are regarded as fruitful groups in activating surface carboxyl groups of several solid substrates. Surface NHS esters had
Fig. 4. XP spectra with wide scans of UA-SiQDs/Si, NHS-UA-SiQDs/Si and OA-UA-SiQDs/Si chips corresponding to (a), (b) and (c) respectively. The high-resolution scan of N 1s core-level spectra of NHS-UA-SiQDs/Si is shown in the inset of (b).
Fig. 5. XP spectra with high-resolution scans of C 1s, O 1s and Si 2p of UA-SiQDs/Si and of OA-UA-SiQDs/Si corresponding to (a–f) respectively.
122
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123
Fig. 6. The PL spectra of (a), (b) and (c) corresponding to the ultrasonic solutions of UA-SiQDs/Si, NHS-UA-SiQDs/Si and OA-UA-SiQDs/Si chips respectively (excitation wavelength: 300 nm). The top insets of (a–c) show the photographs of the corresponding chips obtained by irradiation of an ultraviolet lamp (365 nm) in a dark room. The left inset of (c) is the photo of the ultrasonic solution of OA-UA-SiQDs/Si chip.
been utilized extensively as efficient linkages of conjugating organic or biological molecules bearing primary amines to a solid surface via amidation [19,40]. In this work, we choose OA as a model compound to demonstrate the activity of the NHS esters on surface of the SiQDS/Si chip. The primary amino groups of OA react with the NHS esters bringing about the almost complete disappearance of the tripartite peaks which is shown in Fig. 3(c). The newly occurred peaks of 1643 and 1549 cm1 are assigned to amide I (v(C@O)) and II [34] of OA-UA-SiQDs/Si respectively, which reveals the covalent bonding between OA and SiQDs/Si chip. XP spectrum with wide scan on OA-UA-SiQDs/Si shown in Fig. 4(c) demonstrates the surface chemical ingredients. High-resolution scans of C 1s, O 1s and Si 2p are illuminated in Fig. 5(d–f) respectively. Compared with Fig. 5(a) of UA-SiQDs/Si, Fig. 5(d) of OA-UA-SiQDs/Si changes obviously. Component of O@CAO shown in Fig. 5(a) vanishes completely now. BE of 284.8 eV still belongs to C 1s of CAC/CAH species. The two fitted peaks at 286.4 and 288.2 eV should be assigned to NACAC and O@CAN, respectively [34]. The C 1s integrated areas of CAC/CAH, NACAC and O@CAN are roughly in proportion to 17:1:1 which agrees well with the theoretically atomic ratio of OA-UA-SiQDs/Si. These prove the amidation between surface NHS esters and OA. Fig. 5(e) exhibits that O 1s core-level spectrum of OA-UA-SiQDs/Si are curve-fitted with two peak components. BE of C@O locates at 532.1 eV while SiAO shifts to 529.9 eV which testifies the surface oxidation of SiQDs. Si 2p of SiAO at 102.7 eV shown in Fig. 5(f) also demonstrates the inevitable oxidation of SiQDs. Strangely, Si 2p of SiAC gives a comparatively weak intensity at BE of 99.4 eV. It may be derived from that the resultantly long hydrocarbon chains cover the most part of the SiQDs surface. This case is similar to the silicon surface grafted polymer brushes which even gives no any signal of Si 2p [41]. Importantly, the PL intensity of SiQDs/Si almost does not decline with such successive modifications which can be observed clearly in the insert photograph of Fig. 6(a–c). This proves that the SiQDs on silicon substrate do not detach from the substrate during the chemical modifications. The PL of the modified SiQDs/ Si chips are extremely stable in atmosphere for at least two months which proved by the insert images of Fig. 6. Surface chemical tuning may prevent the surface oxidation which in favor to the stability of the PL [42]. The UA-SiQDs/Si, NHS-UA-SiQDs/Si and OAUA-SiQDs/Si chips are ultrasonically disposed and the aqueous solutions give the PL emission peaks at 601, 605 and 596 nm respectively shown in Fig. 6(a–c). The lower insert of Fig. 6(c) shows that the OA-UA-SiQDs solution emits a dazzling red fluorescence clearly which agrees well with the PL measurement. This
forecasts that biomolecules with primary amino groups may be anchored by such NHS-modified SiQDs similarly to OA molecules to realize fluorescent labeling of the biomolecules. Despite the hydrocarbon chains, OA-UA-SiQDs can be dispersed in aqueous media which should be attributable to the partly oxidized surface of SiQDs by surrounding oxygen which is demonstrated by the XP spectra shown in Fig. 5(e) and (f). As is known to all, the surface silicon dioxides are hydrophilic. We consider that this method can avoid the complicate operations of isolating the products from the mixtures since the excessive reactants can be remove by several times of rinsing or washing when the reaction finishes. Final ultrasonication of the labeled SiQDs/Si chip realizes the separation of the functionalized SiQDs from Si substrate very easily. Accordingly, the techniques described in this work may simplify the labeling of biomolecules with SiQDs. 4. Conclusion SiQDs can be manufactured from cheap silicon chips. Readily prepared SiAHx species make it possible to tailor surface SiQDs via the hydrosilylation. The vast varieties of organic compounds with vinyl groups facilitate such surface tuning of SiQDs/Si. The surface SiQDs on silicon substrate are sustainable for properly chemical modifications such as the surface carboxylation with UA and subsequent esterification with NHS; meanwhile, the PL does not vanish. Additionally, it is quite easy to isolate the excessive reactants from surface of SiQDs/Si chip. The modified SiQDs can be separated from the substrate via a supersonic treatment. There is no doubt that these approaches facilitate the manufacture of SiQDs. The biomolecules with primary amino groups are likely to be attached to such NHS-functionalized SiQDs to realize their fluorescent labeling by means of such facile routes. Additionally, since the SiQDs/Si chips can suffer from heating and organic media, they have the promise to be used as an efficient catalyst in organic syntheses [10]. Acknowledgments This work was supported by Educational Commission of Anhui Province of China (KJ2012A051) and Student Innovation Training Program of AHUT (AH201310360273). References [1] J.H. Ahire, Q. Wang, P.R. Coxon, G. Malhotra, R. Brydson, R. Chen, Y. Chao, ACS Appl. Mater. Interfaces 4 (2012) 3285–3292.
X. Liu et al. / Journal of Colloid and Interface Science 426 (2014) 117–123 [2] F. Erogbogbo, K. Yong, I. Roy, R. Hu, W. Law, W. Zhao, H. Ding, F. Wu, R. Kumar, M.T. Swihart, P.N. Prasad, ACS Nano 5 (2011) 413–423. [3] C.M. Hessel, E.J. Henderson, J.G.C. Veinot, Chem. Mater. 18 (2006) 6139–6146. [4] Z. Yang, A.R. Dobbie, K. Cui, J.G.C. Veinot, J. Am. Chem. Soc. 134 (2012) 13958– 13961. [5] S. Regli, J.A. Kelly, A.M. Shukaliak, J.G.C. Veinot, J. Phys. Chem. Lett. 3 (2012) 1793–1797. [6] M.J.L. Portolés, R.P. Diez, M.L.D. Arciprete, P. Caregnato, J.J. Romero, D.O. Mártire, O. Azzaroni, M. Ceolín, M.C. Gonzalez, J. Phys. Chem. C 116 (2012) 11315–11325. [7] R.D. Tilley, J.H. Warner, K. Yamamoto, I. Matsui, H. Fujimori, Chem. Commun. (2005) 1833–1835. [8] S. Ohta, P. Shen, S. Inasawa, Y. Yamaguchi, J. Mater. Chem. 22 (2012) 10631– 10638. [9] Z. Kang, C.H.A. Tsang, Z. Zhang, M. Zhang, N. Wong, J.A. Zapien, Y. Shan, S. Lee, J. Am. Chem. Soc. 129 (2007) 5326–5327. [10] Z. Kang, C.H.A. Tsang, N. Wong, Z. Zhang, S. Lee, J. Am. Chem. Soc. 129 (2007) 12090–12091. [11] Y. He, Y. Su, X. Yang, Z. Kang, T. Xu, R. Zhang, C. Fan, S. Lee, J. Am. Chem. Soc. 131 (2009) 4434–4438. [12] Y. Zhang, X. Han, J. Zhang, Y. Liu, H. Huang, H. Ming, S. Lee, Z. Kang, Nanoscale 4 (2012) 7760–7765. [13] Y. Zhang, X. Han, R. Liu, Y. Liu, H. Huang, J. Zhang, H. Yu, Z. Kang, J. Phys. Chem. C 116 (2012) 20363–20367. [14] K. Sato, H. Tsuji, K. Hirakuri, N. Fukata, Y. Yamauchi, Chem. Commun. (2009) 3759–3761. [15] X. Liu, H. Han, H. Liu, S. Xiao, Surf. Sci. 604 (2010) 1315–1319. [16] P.K. Sudeep, T. Emrick, ACS Nano 3 (2009) 4105–4109. [17] J.A. Kelly, A.M. Shukaliak, M.D. Fleischauer, J.G.C. Veinot, J. Am. Chem. Soc. 133 (2011) 9564–9571. [18] J. Wang, Y. Liu, F. Peng, C. Chen, Y. He, H. Ma, L. Cao, S. Sun, Small 8 (2012) 2430–2435. [19] H. Asanuma, H. Noguchi, K. Uosaki, H. Yu, J. Am. Chem. Soc. 130 (2008) 8016– 8022. [20] H. Kitano, H. Suzuki, K. Matsuura, K. Ohno, Langmuir 26 (2010) 6767–6774.
123
[21] J.D. Dunn, E.A. Igrisan, A.M. Palumbo, G.E. Reid, M.L. Bruening, Anal. Chem. 80 (2008) 5727–5735. [22] S. Xiao, M. Textor, N.D. Spencer, Langmuir 14 (1998) 5507–5516. [23] X. Liu, H. Liu, P. Guo, S. Xiao, Phys. Status Solidi A 208 (2011) 1462–1470. [24] X. Liu, H. Cheng, P. Cui, Appl. Surf. Sci. 292 (2014) 695–701. [25] X. Zhong, Y. Qu, Y. Lin, L. Liao, X. Duan, ACS Appl. Mater. Interfaces 3 (2011) 261–270. [26] D. Gräf, M. Grundner, R. Schulz, J. Vac. Sci. Technol. A 7 (1989) 808–813. [27] X. Liu, H. Zheng, Y. Ma, Q. Yan, S. Xiao, J. Colloid Interface Sci. 358 (2011) 116– 122. [28] X. Li, K. Liang, B.K. Tay, E.H.T. Teo, Appl. Surf. Sci. 258 (2012) 6169–6176. [29] F. Miao, B. Tao, P.K. Chu, Dalton Trans. 41 (2012) 5055–5059. [30] M. Cao, X. Song, J. Zhai, J. Wang, Y. Wang, J. Phys. Chem. B 110 (2006) 13072– 13075. [31] S.A. Alekseev, V. Lysenko, V.N. Zaitsev, D. Barbier, J. Phys. Chem. C 111 (2007) 15217–15222. [32] A. Imanishi, S. Yamane, Y. Nakato, Langmuir 24 (2008) 10755–10761. [33] X. Liu, H. Zheng, Q. Yan, C. Wang, Y. Ma, Y. Tang, S. Xiao, Surf. Sci. 605 (2011) 1106–1113. [34] R. Voicu, R. Boukherroub, Vasiliki Bartzoka, T. Ward, J.T.C. Wojtyk, D.D.M. Wayner, Langmuir 20 (2004) 11713–11720. [35] A. Petit, M. Delmotte, A. Loupy, J. Chazalviel, F. Ozanam, R. Boukherroub, J. Phys. Chem. C 112 (2008) 16622–16628. [36] S.R. Puniredd, O. Assad, H. Haick, J. Am. Chem. Soc. 130 (2008) 13727–13734. [37] M. Rosso, A. Arafat, K. Schroën, M. Giesbers, C.S. Roper, R. Maboudian, H. Zuilhof, Langmuir 24 (2008) 4007–4012. [38] S. Mirabella, R. Agosta, G. Franzò, I. Crupi, M. Miritello, R.L. Savio, M.A. Di Stefano, S. Di Marco, F. Simone, A. Terrasi, J. Appl. Phys. 106 (2009) 103505. [39] D. Guo, S. Xiao, B. Xia, Shuai-Wei, J. Pei, Y. Pan, X. You, Z. Gu, Z. Lu, J. Phys. Chem. B 109 (2005) 20620–20628. [40] F. Rusmini, Z. Zhong, J. Feijen, Biomacromolecules 8 (2007) 1775–1789. [41] D. Xu, W.H. Yu, E.T. Kang, K.G. Neoh, J. Colloid Interface Sci. 279 (2004) 78–87. [42] Z. Kang, Y. Liu, C.H.A. Tsang, D.D.D. Ma, X. Fan, N. Wong, S. Lee, Adv. Mater. 21 (2009) 661–664.
