Open access peer-reviewed chapter - ONLINE FIRST

Janus Nanosheets Derived from K4Nb6O17·3H2O via Regioselective Interlayer Surface Modification

By Ryoko Suzuki, Mitsuhito Sudo, Megumi Hirano, Naokazu Idota, Masashi Kunitake, Taisei Nishimi and Yoshiyuki Sugahara

Submitted: July 28th 2018Reviewed: January 8th 2019Published: February 18th 2019

DOI: 10.5772/intechopen.84228

Downloaded: 102

Abstract

Inorganic Janus nanosheets were successfully prepared using the difference in reactivity between interlayers I and II of layered hexaniobate K4Nb6O17·3H2O. Janus nanosheets exhibit the highest anisotropy among Janus compounds due to their morphology. It is therefore important to prepare Janus nanosheets with stable shapes in various solvents, robust chemical bonds between nanosheets and fuctional groups and high versatility due to surface functional groups. K4Nb6O17·3H2O, which possesses two types of interlayers and two types of organophosphonic acids that react with metal oxides to form robust covalent bonds, was employed to prepare Janus nanosheets for this study. Interlayer I was modified by octadecylphosphonic acid, followed by modification by carboxypropylphosphonic acid mainly at interlayer II. Preparation of Janus nanosheets with two organophosphonate moieties was confirmed by 31P MAS NMR. After these regioselective and sequential modifications, the products were exfoliated into single-layered nanosheets in THF. Two types of derivatives with different repeating distances were recovered from a dispersion containing nanosheets exfoliated by different processes, centrifugation, and solvent evaporation. AFM analysis of the exfoliated nanosheets revealed that the products were Janus compounds. There are high expectations for application of these types of Janus nanosheets in various fields and for design of various Janus nanosheets using this preparation method.

Keywords

  • Janus nanosheets
  • K4Nb6O17·3H2O
  • organophosphonic acid
  • grafting reaction
  • intercalation

1. Introduction

A Janus compound has two surface properties, and each of these properties appears on one of two sides of the compound [1]. Janus compounds are expected to be applied as functional materials, including electronic paper [2], solid surfactants [3], optics materials [4], and drug delivery system (DDS) vectors [5]. The morphology of Janus compounds is classified into three categories: 0-dimensional compounds such as particles; one-dimensional compounds including cylinders, tubes, and rods; and two-dimensional compounds, typically sheets or discs [6, 7, 8, 9, 10].

There are various methods of preparing Janus nanoparticles. Regioselective surface modification of nanoparticles can produce Janus nanoparticles [11]. By forming nanoparticles with two raw materials, Janus nanoparticles with different compositions can be prepared [2]. Self-assembly and subsequent cross-linking of polymer chains can also produce Janus nanoparticles [12]. Janus rods can be prepared by forming silica using TEOS by the sol-gel method on Fe3O4 regioselectively [13]. Janus cylinders can be prepared by masking one side of a cylinder and conducting subsequent modification of the other side [14].

Janus nanosheets exhibit the highest anisotropy among Janus compounds due to their unique morphology. They are also useful as emulsifiers, because Janus nanosheets cannot rotate at the interfaces of micelles [15]. Most Janus nanosheets reported so far have consisted of polymers. Stupp et al. reported the preparation of Janus nanosheets by polymerization of oligomers with polymerizable groups [16]. Polymerization of oligomers led to the formation of sheet morphology, because the polymerizable groups were located at the center of an oligomer. Walther et al. used triblock copolymers, polystyrene-block-polybutadiene-block-poly(tert-butyl methacrylate), for preparing Janus nanosheets [17]. Janus nanosheets were prepared by cross-linking polybutadiene domains of the triblock copolymers, and the resulting Janus nanosheets had two types of surfaces, polystyrene, and poly(tert-butyl methacrylate) moieties. These methods are based on selective polymerization or cross-linking. On the other hand, Janus nanosheets were prepared by dropping poly(ε-caprolactone) at the interface between water and pentyl acetate and evaporating solvents to crystallize polymers [10]. The properties of Janus compounds were realized by folding the polymer an odd number of times, which exposed carboxyl groups on one side of the nanosheets. Although polymer-based Janus nanosheets have been prepared, these Janus nanosheets became swollen and deformed in organic solvents because they were composed of polymers [18].

It is therefore obvious that the preparation of inorganic Janus nanosheets is an important issue. There are only a few reports of preparation of Janus nanosheets derived from inorganic compounds, however, comprising silica nanosheets prepared by the sol-gel method [19, 20, 21] and fluorohectorite, which is a layered clay mineral [22].

Silica-based Janus nanosheets were prepared by hydrolysis and condensation of two trifunctional organoalkoxysilanes. First, phenylalkoxysilane was assembled on the surface of oil in a W/O emulsion and reacted by the sol-gel method. The product, which formed a hollow shell, was further reacted with aminoalkyltrialkoxysilane. The resultant product had phenyl groups and amino groups on the inner and outer surfaces of hollow particles, respectively. Janus nanosheets were obtained by breaking the hollow silica particles using a colloid milling method. The resulting Janus nanosheets have a thickness of 65 nm and a curvature originating from the hollow particle morphology [19]. Another method of preparing silica-based Janus nanosheets using a CaCO3 template was also reported. First, 3-butyldianhydride mercaptopropyltrimethoxysilane was assembled on the surface of CaCO3 particles and reacted using a sol-gel process. The products were further reacted with octadecyltrichlorosilane. Janus nanosheets were obtained by removing the templates and crushing the resultant hollow particles. Janus nanosheets with single a nanometer thickness were prepared by this method [21]. These sol-gel preparation methods, which were developed by Yang et al., required adsorption of organosilanes on liquid-liquid interfaces or self-assembly on templates. Thus, these methods restricted the reaction system design and choices of molecules.

Another method of preparing Janus nanosheets using inorganic layered material was reported in which the Janus nanosheets consisted of a sheet of a clay mineral, fluorohectorite, and two cationic polymers. The interlayer of fluorohectorite was selectively exfoliated into double-layered nanosheets. Both surfaces of the products, which had a negative charge, were modified with a cationic polymer, protonated polyethyleneimine-ethylene oxide. After that, the products were exfoliated into single-layered nanosheets, and one surface which had not been modified was reacted with another cationic polymer, protonated dendritic poly(amidoamine). The products had one cationic polymer on one side and the other polymer on the other side of the nanosheets [22]. This preparation method has the advantage that any cationic compound could be used as the modifier, regardless of its functional group. On the other hand, the products use in water was restricted due to the presence of ionic bonds between the surfaces of the clay sheets and the polymers.

Preparation of graphene-based Janus nanosheets has also been developed [23]. For example, two-step functionalization was carried in one report [24]. First, the graphene surface of the substrate was functionalized by a photochlorination reaction. The functionalized graphene was then peeled off the substrate to expose the unmodified side using PMMA film as a mediator. The exposed fresh side of the graphene was further functionalized by phenylation reaction. Graphene-based Janus nanosheets were also prepared using a Pickering emulsion [25]. A graphene oxide (GO) dispersion was mixed with hydrochloric acid and wax, and this mixture was ultrasonicated to prepare a Pickering emulsion. The micelles were then washed with a sodium hydroxide aqueous solution, and GO was adsorbed on the surface of micelles to form a monolayer. The exposed GO surface of the micelles was further modified with alkylamine. Finally, Janus nanosheets were obtained by dissolving wax in chloroform. In the preparation of graphene-based Janus nanosheets, it is necessary that a single layer of graphene be adsorbed on a substrate or at the liquid-liquid interphase to achieve regioselective functionalization.

These earlier studies show that the following conditions are desirable: nanosheets should have covalent bonds with organic groups, the choice of functional groups should not be limited, and regioselective surface modification should be easily achieved. Another preparation method that satisfies the above conditions should therefore be developed.

Some inorganic layered materials have structures in which negatively charged nanosheets and metal cations are piled up alternately. Nanosheets are very thin, just a few atoms thick. On the other hand, their lateral sizes are large, possibly on a micrometer scale. Therefore, these materials have high aspect ratios. From the viewpoints of structure and composition, inorganic layered materials are classified into several categories, including clay minerals [26, 27, 28], layered silicate [29, 30, 31], and layered transition metal oxide [32, 33, 34]. These inorganic layered materials have been used as hosts for inorganic-organic hybrids. There are only two types of reaction for preparing inorganic-organic hybrids: intercalation reactions forming noncovalent bonds and grafting reactions forming covalent bonds. Grafting reactions that form covalent bonds between layer surfaces and organic groups utilize alcohols, carboxylic acids, silane coupling agents, and phosphorous coupling reagents such as organophosphonic acids as modifiers. Grafted organophosphonate moieties, in particular, are seldom eliminated from nanosheet surfaces because organophosphonate moieties form stable M–O–P bonds with nanosheet surfaces, except for Si–O–P bonds, which are known to become hydrolyzed under certain conditions [35]. Surface modification with monolayer can be easily achieved, moreover because homocondensation reactions do not occur between organophosphonic acids under mild conditions [36]. There have been many reports of surface modification by silane coupling agents for clay minerals [37], layered silicates [38, 39, 40], and layered transition metal oxides [41, 42]. In the case of surface modification by alcohol, there have been reports of polysilicates with ethylene glycol [43] and aliphatic alcohols [44], while layered perovskites have been modified with n-alcohol [45] and alcohol with fluoroalkyl groups [44, 46]. Also, layered perovskites were grafted with phenyl or n-alkylphosphonic acids using the aforementioned n-alkoxy derivatives as intermediates [47].

Layered hexaniobate (K4Nb6O17·3H2O) has a unique structure among layered transition metal oxides; K4Nb6O17·3H2O has two types of interlayers that are piled up alternately and exhibit different reactivities [48]. Interlayer I possesses hydrated water and shows high reactivity, which anhydrous interlayer II exhibits low reactivity. There have been a certain number of reports of reactions between K4Nb6O17·3H2O and organic molecules using the differences in reactivity between interlayer I and interlayer II [33].

Intercalation of small ammonium ions occurred sequentially, first in interlayer I and then in interlayer II [49]. On the other hand, bulky ammonium ions were intercalated only into interlayer I [50]. Compounds that were modified only in interlayer I were called A-type, and compounds that were modified in both interlayers I and II were called B-type. Kimura et al. modified the K4Nb6O17·3H2O surfaces with phenylphosphonic acid using A-type and B-type ion-exchanged intercalation compounds of K4Nb6O17·3H2O [51]. In their report, bulky dioctadecyldimethylammonium ions were intercalated into only interlayer I, and A-type phenylphosphonate derivatives were obtained using this A-type ammonium intercalation compound as an intermediate. On the other hand, dodecylammonium ions were intercalated into both interlayers I and II and a B-type phenylphosphonate derivative was obtained using this B-type ammonium intercalation compound as an intermediate. Thus, regioselective surface modification of K4Nb6O17·3H2O by organophosphonic acid was successfully achieved.

A variety of nanosheets have been obtained by exfoliation of layered materials [52], and various methods have been reported for their exfoliation. A simple method of exfoliation is dispersing layered materials in water. Water molecules can intercalate in the interlayer and promote exfoliation [53]. Bulky ammonium ions intercalated in the interlayer can expand the interlayer distance and decrease interactions between the negative charge and positive charge to cause exfoliation. [54]. Mechanical exfoliation using ultrasonication has also been employed [55]. On the other hand, in situ polymerization of organic monomers in the interlayer can also lead to exfoliation of layered materials. A modifier grafted onto the interlayer surface is reacted with monomers and generates polymer chains that expand the interlayer distance. This polymerization method is called the “grafting from” method and is often reported in the field of graphene [56]. Introducing small molecules, such as carboxyl acids, into the interlayer as an initiator group could cause polymerization from the surface of graphene [57, 58]. Another report of the “grafting from” method utilized a layered perovskite, HLaNb2O7·xH2O, which was modified with organophosphonic acid bearing an initiation group on the interlayer surface, and N-isopropylacrylamide (NIPAAm) was polymerized from the initiation group by atom transfer radical polymerization [59]. The interlayer distance was expanded by polymerization, and nanosheets dispersed in water were obtained. Because a thermos-responsive polymer, poly(N-isopropylacrylamide), PNIPAAm, was bound to the nanosheet surfaces, the nanosheets were hydrophilic and dispersed in water at below the lower critical solution temperature (LCST). On the contrary, the nanosheets become hydrophobic and aggregate at over the LCST in water.

In this research, the preparation of Janus nanosheets was achieved by taking advantage of the presence of two types of interlayers with different reactivities in K4Nb6O17·3H2O. Interlayer II of an A-type organophosphonic acid derivative of K4Nb6O17·3H2O was reacted with another type of organophosphonic acid. Both sides of niobate nanosheets were modified by two organophosphonic acids regioselectively, because organophosphonic acid could not undergo an exchange reaction and homocondensation. Janus nanosheets could be obtained by exfoliation of the product into single-layer nanosheets in an appropriate solvent, THF. Lipophilic octadecylphosphonic acid (ODPA) and hydrophilic carboxypropylphosphonic acid (CPPA) were chosen as the organophosphonic acids. The properties of both sides of Janus nanosheets were explored by the AFM phase imaging technique. This report is based on a study first reported in Chemical Communications (Scheme 1) [60].

Scheme 1.

Preparation of Janus nanosheet.

2. Experimental section

An A-type alkylammonium intercalation compound, (2C182MeN)1.0(K, H)3[Nb6O17] (2C182MeN = dioctadecyldimethylammonium ion), was prepared based on the previous report [49]. Octadecylphosphonic acid (ODPA) was synthesized as described elsewhere [61, 62]. Dodecyltrimethylammonium chloride, carboxypropylphosphonic acid (CPPA), 2-butanone, acetone, and tetrahydrofuran (THF) were used without further purification.

First, interlayer I of the A-type alkylammonium intercalation compound was modified by ODPA. The A-type alkylammonium intercalation compound (0.05 g) and ODPA (0.048 g) were used to adjust the Nb:ODPA molar ratio to 1:4 and reacted in 2-butanone (20 mL) at 150°C for 7 days. After the reaction, the crude product was centrifuged, washed with THF and HCl (pH = 3), and air-dried (ODPA_NbO). The cations (K+, H+) in interlayer II were then exchanged with the dodecylammonium ion (C12N+) to expand interlayer II. ODPA_NbO (0.1 g) and dodecyltrimethylammonium chloride (0.19 g) were used to adjust the ODPA:C12N+ molar ratio to 1:10 and reacted in water (10 mL) at 80°C for 3 days. After the reaction, the crude product was centrifuged, washed with water, and air-dried (ODPA_C12N_NbO). Then, interlayer II of ODPA_C12N_NbO was modified by CPPA. ODPA_C12N_NbO (0.05 g) and CPPA (0.05 g) were used to adjust the Nb:CPPA molar ratio to 1:10 before reaction in 2-butanone (10 mL) at 80°C for 3 days. After reaction, the crude product was centrifuged, washed with THF and HCl (pH = 3), and corrected as a precipitate (ODPA_CPPA_NbO). After centrifugation, the product dispersed in supernatant was corrected by slow evaporation of THF (ODPA_CPPA_NbO_evaporation). TEM and AFM samples were prepared by stirring ODPA_CPPA_NbO in THF to exfoliate them into single-layer nanosheets.

3. Analysis

XRD analysis (convergence method) was carried out with a Rigaku RINT-1000 diffractometer (Mn-filtered FeKα radiation). XRD analysis (parallel beam method) was performed with a Rigaku SmartLab diffractometer (oblique incidence, FeKα radiation). IR analysis was conducted with a JASCO FT/IR-460 Plus spectrometer by the KBr method. Solid-state 31P magic angle spinning (MAS) NMR spectra were recorded on a JEOL JNM-ECX400 spectrometer. The measurement conditions were as follows: resonance frequency: 160.26 MHz; pulse angle: 90°; pulse delay: 30 s; and MAS frequency: 12 kHz. Triphenylphosphine (−8.4 ppm) was used as a reference. Solid-state 13C cross-polarization (CP)/MAS NMR spectra were recorded on a JEOL JNM-ECX-400 spectrometer. The measurement conditions were as follows: resonance frequency: 99.55 MHz; pulse delay: 5 s; contact time: 1.5 ms; and MAS frequency: 12 kHz. Hexamethylbenzene (17.4 ppm) was used as a reference. ICP-AES measurement was performed using a Thermo Jarrell Ash ICAP-574II instrument. Samples (about 10 mg) were dissolved by heating at 150°C overnight in HF (1 mL), HCl (3 mL), and HNO3 (4 mL). H3BO3 (70 mL) was added as a masking reagent for HF. HF (1 mL), HCl (3 mL), HNO3 (4 mL), and H3BO3 (70 mL) were added to each standard solution for matrix matching. The amounts of C, H, and N in the samples were measured by elemental analysis using a PerkinElmer PE2400II instrument. Transmission electron microscope (TEM) images were observed with a JEOL JEM-1011 microscope operating at 100 kV. A TEM sample was prepared by dropping drops of a dispersion on a Cu 150P grid and drying under reduced pressure. Atomic force microscope (AFM) images were observed with an Agilent 5500 AFM/SPM microscope in the acoustic AC mode under ambient conditions. An ordinary commercial silicon cantilever was used as an AFM tip (e.g., a RTESP-300 from Bruker: resonance frequency ≈ 300 kHz, and spring constant ≈ 40 N/m). Samples for AFM were prepared by spin coating of the dispersion on a Si wafer.

4. Results and discussion

Figure 1 shows 13C CP/MAS NMR spectra of the products. In the spectrum of ODPA_NbO (Figure 1a), signals assignable to the octadecyl group were observed. It is likely that the ODPA moiety was introduced into interlayer I, since dioctadecyldimethylammonium ions, whose presence was required for interlayer modification with organophosphonic acids [51], were present only in interlayer I. In the spectrum of ODPA_C12N_NbO (Figure 1b), signals assignable to alkyl chains (octadecyl and dodecyl) were observed at 15–43 ppm [51]. In addition, a signal assignable to a carbon atom adjacent to a nitrogen atom was observed at 43 ppm [63], indicating the presence of C12N+. Since C12N+ is known to be intercalated into both interlayer I and interlayer II [48], the intercalation of C12N+ into interlayer II was likely to occur. In the spectrum of ODPA_CPPA_NbO (Figure 1c), signals due to alkyl chains were observed at 15–36 ppm. On the other hand, a signal originating from C12N+ at 43 ppm disappeared and a signal due to C=O groups of CPPA was observed at 178 ppm [64]. These results suggest the removal of C12N+ and introduction of the CPPA moiety to ODPA_CPPA_NbO.

Figure 1.

13C CP/MAS NMR spectra of (a) ODPA_NbO, (b) ODPA_C12N_NbO, and (c) ODPA_CPPA_NbO.

Figure 2 shows IR spectra of the products. In the spectrum of ODPA_NbO (Figure 2a), absorption bands due to ν (C–H), σs (CH2), and ν (P–O) modes were observed at 2956–2849, 1468, and 1011 cm−1, respectively [65], indicating that ODPA moiety was present in ODPA_NbO. In the spectrum of ODPA_C12N_NbO (Figure 2b), an adsorption band at 1540 cm−1 assignable to the σ (N–H) mode was observed in addition to the aforementioned adsorption band, indicating that octadecylammonium ions were present in ODPA_C12N_NbO. In the spectrum of ODPA_CPPA_NbO (Figure 2c), a new adsorption band that was assignable to the σ (C=O) mode of the CPPA moiety was observed at 1700 cm−1 [66], indicating the presence of the CPPA moiety in ODPA_CPPA_NbO. It was reported that an adsorption band due to the νas (CH2) of alkyl chain was shifted from 2924.7 cm−1 to a lower wavenumber by increasing the packing density of the alkyl chain [67]. In the case of the all-trans octadecyl alkyl chain, νas (CH2) was observed at 2917.8 cm−1 [67, 68] and a σs (CH2) band was observed at 1468 cm−1 [68]. In the spectrum of ODPA_NbO, adsorption bands assignable to νas (CH2) and νs (CH2) modes were observed at 2918 and 2848 cm−1, respectively, and a σs (CH2) adsorption band was observed at 1468 cm−1. Thus, the alkyl chain in ODPA_NbO was likely to be in an all-trans conformation. On the other hand, νas (CH2), νs (CH2), and σs (CH2) adsorption bands were observed at 2923, 2852, and 1456 cm−1 in the spectrum of ODPA_CPPA_NbO, respectively, indicating that the alkyl chain in ODPA_CPPA_NbO was likely to contain gauche-blocks.

Figure 2.

IR spectra of (a) ODPA_NbO, (b) ODPA_C12N_NbO, and (c) ODPA_CPPA_NbO.

Figure 3 shows 31P MAS NMR spectra of the products. A signal was observed at 28 ppm in the spectrum of ODPA_NbO (Figure 3a). This signal was shifted upfield from the chemical shift of the ODPA molecule (33 ppm at 31P MAS NMR) by 5 ppm, indicating that interlayer surface modification by ODPA had proceeded and an Nb–O–P bond had been formed [47]. In the spectrum of ODPA_C12N_NbO (Figure 3b), a signal was observed at 25 ppm. This signal was shifted upfield from 28 ppm, the chemical shift of ODPA_NbO, by 3 ppm. This shift suggests that C12N+ would change the electronic environment around the P atom by an ion exchange reaction with H+ of the P–OH group [47], although the details were not yet clarified. Thus, it is likely that C12N+ was intercalated not only in interlayer II, but probably also in interlayer I upon the reaction with ODPA_NbO. In the spectrum of ODPA_CPPA_NbO (Figure 3c), a new signal was observed at 31 ppm in addition to the signal at 28 ppm. The signal at 28 ppm was observed in the same position as that of the ODPA moiety of ODPA_NbO, confirming maintenance of the ODPA moiety at interlayer I. Because a signal of a CPPA molecule was observed at 34 ppm, a signal at 31 ppm was assignable to the CPPA moiety. This signal was shifted upfield by 3 ppm, indicating that the CPPA moiety was grafted onto the interlayer surface and a Nb–O–P bond was formed. The above results suggested that ODPA and CPPA formed covalent bonds with the [Nb6O17]4− sheet surface. Since bands assignable to P–OH groups and P=O groups were observed in the IR spectrum of ODPA_CPPA_NbO, ODPA and CPPA were likely to be in a monodentate environment on the surface of [Nb6O17]4− sheet.

Figure 3.

31P MAS NMR spectra of (a) OPDA_NbO, (b) ODPA_C12N_NbO, and (c) ODPA_CPPA_NbO.

Table 1 shows the molar ratio calculated from the ICP measurement and elemental analysis. The molar ratio of ODPA_NbO was P:Nb = 1.3:6.0. On the other hand, the molar ratios of ODPA_C12N_NbO and ODPA_CPPA_NbO were P:Nb = 1.3:6.0 and P:Nb = 3.5:6.0, respectively. Intercalation of C12N+ proceeded without release of the ODPA moiety in ODPA_NbO, because the molar ratio of P and Nb of ODPA_NbO did not change after reaction with a dodecylammonium chloride solution. Also, the molar ratio of P to 6 Nb in ODPA_CPPA_NbO increased by 2.2 (3.5 – 1.3), confirming grafting of the CPPA moiety. Assuming Nb = 6.0, the maximum modification amounts for interlayer I and II are 2.0 [51]. Since the Nb–O–P bond was stable with respect to hydrolysis and no homocondensation between two P–OH groups of phosphonic acid occurred under mild conditions [36], the amount of the ODPA moiety in interlayer I was estimated to be 1.3 (65% of the maximum modification amount), that of the CPPA moiety at interlayer I was in the range of 0.2–0.7 (10–35% of maximum modification amount), and that of the CPPA moiety in interlayer II was in the range of 1.5–2.0 (75–100% of the maximum modification amount). Thus, an organic derivative with interlayer I and interlayer II dominantly modified with hydrophobic ODPA and hydrophilic CPPA, respectively, were successfully prepared (Scheme 2).

Nb/–K/–P/–N/–
ODPA_NbO6.02.61.30.082
ODPA_C12N_NbO6.00.581.31.8
ODPA_CPPA_NbO6.00.493.5

Table 1.

Molar ratios of ODPA_NbO, ODPA_C12N_NbO, and ODPA_CPPA_NbO.

Scheme 2.

Proposed structure of ODPA_CPPA_NbO.

Based on the nitrogen ratio of ODPA_NbO, it seems that a small amount of unreacted A-type alkylammonium intercalation compound was present in ODPA_NbO or a small number of released 2C182MeN ions were present in interlayer I via ion exchange. Since the amount of K+ in interlayer II in ODPA_C12N_NbO decreased, an ion exchange reaction between K+ and H+ ions at interlayer II and C12N+ proceeded. Because no nitrogen was detected in ODPA_CPPA_NbO, C12N+ was completely removed from interlayer I and II after the reaction with CPPA.

A THF dispersion of nanosheets was easily attained by dispersing ODPA_CPPA_NbO in THF. The resulting dispersion was cast on a TEM grid, and TEM observation was carried out (Figure 4). A sheet-like morphology with low contrast was observed. Spots observed in the electron diffraction (ED) pattern can be assigned to 200, 202, and 002 of the orthorhombic cells, and the lattice parameters were calculated to be a = 0.80 nm and c = 0.64 nm. This ED pattern was thus a b-axis incidence pattern of K4Nb6O17·3H2O [69]. Based on these results, ODPA_CPPA_NbO was synthesized while maintaining the crystal structure of the [Nb6O17]4− nanosheets.

Figure 4.

TEM image of exfoliated ODPA_CPPA_NbO. The inset shows the corresponding ED pattern.

Figure 5 shows XRD patterns of the products. The d values of low-angle diffractions due to repeating distances were as follows: the d value of ODPA_NbO, A-type derivative (Figure 5a), was 5.67 nm and the d value of ODPA_C12N_NbO (Figure 5b) was 4.03 nm. If intercalation of C12N+ into interlayer II proceeded while maintaining an A-type stacking sequence, the d value of ODPA_C12N_NbO is likely to have increased from that of ODPA_NbO. It is possible that a B-type stacking sequence was generated due to exfoliation and restacking during the reaction, resulting in a smaller repeating distance. Also, the d values of ODPA_CPPA_NbO (Figure 5c) and ODPA_CPPA_NbO_evaporation (Figure 5d) were 2.41 and 4.74 nm, respectively. The stacking sequence would therefore be changed by reaction between ODPA_C12N_NbO and CPPA.

Figure 5.

XRD patterns of (a) ODPA_NbO, (b) ODPA_C12N_NbO, (c) ODPA_CPPA_NbO, and (d) ODPA_CPPA_NbO_evaporation.

Here, the difference between these two d values is discussed. If ODPA_CPPA_NbO is a B-type derivative, the thickness of an organic moiety layer (sum of an ODPA monolayer and a CPPA monolayer) can be calculated by subtracting 0.82 nm, the niobate layer thickness, from 2.41 nm to make 1.59 nm [51]. The repeating distance of an A-type derivative could thus be estimated as the sum of a double niobate layer thickness and a double organic layer thickness. The repeating distance of an A-type derivative can therefore be estimated as follows: (1.59 nm × 2) + (0.82 nm × 2) = 4.82 nm. This value is approximately equal to d = 4.74 nm of ODPA_CPPA_NbO_evaporation. From these estimations, it is proposed that ODPA_CPPA_NbO is a B-type derivative and ODPA_CPPA_NbO_evaporation is an A-type derivative. As shown in Scheme 3, a B-type derivative could be generated by forced restacking via centrifugation of exfoliated nanosheets (Scheme 3a and b). On the other hand, an A-type derivative, in which hydrophilic groups faced each other and hydrophobic groups faced each other, was obtained by slow evaporation under mild conditions (Scheme 3c).

Scheme 3.

The estimated structures of ODPA_CPPA_NbO: Possible routes from (a) ODPA_C12N_NbO to (b) ODPA_CPPA_NbO and (c) ODPA_CPPA_NbO_evaporation.

The crystallite sizes calculated from diffraction of the repeating distances using Scherrer’s formula were 3.67 and 7.71 nm for ODPA_CPPA_NbO and ODPA_CPPA_NbO_evaporation, respectively. The crystallite size of ODPA_CPPA_NbO_evaporation was larger than that of ODPA_CPPA_NbO. It should be noted that underestimation could occur with use of lowest-angle diffractions due to the presence of strain [32]. The crystallite sizes could therefore reflect the average thickness of the particles in the stacking direction, making the number of stacked ODPA_CPPA_NbO nanosheets lower than that of stacked ODPA_CPPA_NbO_evaporation nanosheets. On the other hand, the estimated crystallite size could be interpreted as average thickness of a portion of the stacked sheets with an A-type or B-type stacking sequence. Based on this interpretation, ODPA_CPAN_NbO formed via forced restacking by centrifugation has lower stacking regularity or more random stacking than ODPA_CPPA_NbO_evaporation nanosheets restacked under mild conditions.

Figure 6 shows an AFM image of a sample prepared by spin coating of a THF dispersion of ODPA_CPPA_NbO on a Si wafer. It contained many nanosheets that showed a relatively uniform thickness in the range of 2.5–3.0 nm (Figure 6A). This thickness range is approximately equal to the d value of B-type ODPA_CPPA_NbO, indicating that ODPA_CPPA_NbO was exfoliated into single-layer nanosheets that were casted on a Si wafer.

Figure 6.

Topographic (A) and phase (B) AFM images of ODPA_CPPA_NbO Janus nanosheets.

As marked by the a and b arrows in Figure 6, two different colored nanosheet surfaces were observed in the phase image (Figure 6B). This indicates the presence of two different faces (36–37° and 38–39°) in each Janus nanosheet. This phase difference in the AFM phase image corresponds to the tapping phase gap in the vibration amplitude, and it was reported that the phase difference could occur with a difference in the crystallinity, viscosity, and adhesion of the sample surface [70, 71].

The Janus nanosheets consisted of a hydrophobic surface, which was dominantly covered with the ODPA moiety, and a hydrophilic surface, which was modified with the CPPA moiety. As a result, two chemically different surfaces gave different phases due to differences in the interactions between the apex of the AFM probe and the surfaces of the nanosheets and distinguished visually in the phase image. The origin of the phase contrast would be due to differences in viscosity and hydrophilicity/hydrophobicity. Since the apex of the AFM probe used in this measurement was hydrophilic, it is likely that the high-phase surface and a low-phase surface were assignable to the hydrophilic CPPA moiety and hydrophobic ODPA moiety, respectively.

The c arrow in Figure 6 marks the overlapping area of two nanosheets (a and b). Obviously, these area possessed double-layer thickness. The color of the phase image of this area (Figure 6B) indicates that nanosheet b partially overlapped nanosheet a. Thus, these results indicate that hydrophilic and lipophilic surfaces are facing each other. These results also indicate that the nanosheets prepared in this study exhibited hydrophobicity on one side and hydrophilicity on the other.

5. Conclusions

Janus nanosheets were successfully prepared by regioselective and sequential surface modification and exfoliation of K4Nb6O17·3H2O, whose interlayer I and interlayer II were dominantly modified by ODPA and CPPA, respectively. Since organophosphonic acids bearing various functional groups can be easily synthesized, Janus nanosheet surfaces can exhibit various properties in addition to hydrophobicity and hydrophilicity. The Janus nanosheets prepared by the present method can be dispersed in many solvents, moreover, because organophosphonic moieties are bound to niobate nanosheets by covalent bonds. The Janus nanosheets prepared in this study can be expected to be applied in surface chemistry research because of the hydrophobicity and hydrophilicity on opposing sides of the nanosheets. Also, by changing the functional groups of organophosphonic acids, novel two-dimensional materials with various functions with potential applications in various fields can be realized.

Acknowledgments

This work was financially supported in part by a Grant-in-Aid for Scientific Research on Innovative Areas “New Polymeric Materials Based on Element-Blocks (No. 2401)” (JSPS KAKENHI Grant Numbers JP24102002), “Coordination Asymmetry (JP 23655205),” and Grant-in-Aid for Challenging Exploratory Research (JP17H05378). Reproduced from Ref. [60] with permission from the Royal Society of Chemistry.

Conflict of interest

There are no conflicts to declare.

Download

chapter PDF

© 2019 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution 3.0 License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite and reference

Link to this chapter Copy to clipboard

Cite this chapter Copy to clipboard

Ryoko Suzuki, Mitsuhito Sudo, Megumi Hirano, Naokazu Idota, Masashi Kunitake, Taisei Nishimi and Yoshiyuki Sugahara (February 18th 2019). Janus Nanosheets Derived from K4Nb6O17·3H2O via Regioselective Interlayer Surface Modification [Online First], IntechOpen, DOI: 10.5772/intechopen.84228. Available from:

chapter statistics

102total chapter downloads

More statistics for editors and authors

Login to your personal dashboard for more detailed statistics on your publications.

Access personal reporting

We are IntechOpen, the world's leading publisher of Open Access books. Built by scientists, for scientists. Our readership spans scientists, professors, researchers, librarians, and students, as well as business professionals. We share our knowledge and peer-reveiwed research papers with libraries, scientific and engineering societies, and also work with corporate R&D departments and government entities.

More about us