Enhancing the Dispersibility of TiO2 Nanorods and ... - ACS Publications

Sep 26, 2016 - I, Friedrich-Alexander Universität Erlangen-Nürnberg (FAU), Henkestrasse 42, 91054 Erlangen, Germany. •S Supporting Information...
0 downloads 0 Views 1MB Size
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article 2

Enhancing the dispersability of TiO nanorods and gaining control over region-selective layer formation Sebastian Heinrich Etschel, Rik R. Tykwinski, and Marcus Halik Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.6b02480 • Publication Date (Web): 26 Sep 2016 Downloaded from http://pubs.acs.org on September 27, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Enhancing the dispersability of TiO2 nanorods and gaining control over region-selective layer formation Sebastian H. Etschel,[a,b] Rik R. Tykwinski,[b]* Marcus Halik[a]* a

Department Werkstoffwissenschaften, Lehrstuhl für Polymerwerkstoffe, Organic

Materials and Devices (OMD), Friedrich-Alexander Universität Erlangen-Nürnberg (FAU), Martensstrasse 7, 91058 Erlangen (Deutschland); http://omd.fau.de b

Department für Chemie und Pharmazie & Interdisciplinary Center for Molecular

Materials (ICMM), Lehrstuhl Organische Chemie I, Friedrich-Alexander Universität ErlangenNürnberg (FAU), Henkestrasse 42, 91054 Erlangen (Deutschland); http://www.chemie.unierlangen.de/tykwinski

KEYWORDS: Copper-catalyzed alkyne-azide cycloaddition, core-shell nanoparticles, regionselective layer formation, phosphonic acid self-assembled monolayers

ACS Paragon Plus Environment

1

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 23

ABSTRACT: We demonstrate that the dispersability and reactivity of core-shell TiO2 nanorods (NRs) can be controlled significantly through functionalization with a combination of ligands based on phosphonic acid derivatives (PAs). Specifically, a glycol based PA allows dispersion of the NRs in methanol (MeOH). On the other hand, incorporating an alkyne terminated PA in the ligand shell of the NRs allows for a copper-catalyzed alkyne-azide cycloaddition (CuAAC) reaction with an azide-patterned aluminum oxide (AlOx) substrate and forms a region-selectively deposited film of NRs. We clearly demonstrate that the quality of the NR films correlates strongly with the stability of the NR dispersions in the reaction medium. In particular, tuning the concentration of alkyne PA in the ligand shell inhibits aggregation of the NRs on the substrate, while reactivity for the CuAAC reaction is maintained. The surface coverage with NRs fits the Langmuir model. This study illustrates that surface functionalization of AlOx substrates can be effectively and conveniently controlled through enhancing the dispersability of the NRs using mixed ligand shells.

INTRODUCTION: Controlling the layer formation of nanomaterials on a substrate is a crucial aspect for their application in most electronic devices.1 The Huisgen-1,3-dipolar cycloaddition, also referred to as copper-catalyzed alkyne-azide cycloaddition (CuAAC) reaction in case of a Cu+ catalyst, has been introduced by Sharpless and Meldal in 2001 and quickly became known as the prototypic Click reaction due to its simplicity.2–4 The CuAAC reaction also shows rather remarkable substrate tolerance, and it has been quickly applied to the field of materials and surface chemistry.5–11 For example, it has been demonstrated that the CuAAC reaction can be used to direct layered deposition of core-shell functionalized nanoparticles (NPs) that bear either

ACS Paragon Plus Environment

2

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

an alkyne or an azide headgroup in their organic ligand shell, and accordingly functionalized self-assembled monolayer (SAM) substrates.12–14 Recently, we reported on a method in which three different layers of core-shell NPs could be hierarchically assembled using a layer-by-layer process, controlled by the CuAAC.15 We further demonstrated that the reported deposition technique can be implemented into the fabrication process of capacitor structures, which offers a new level of control for the production of electronically active NP layers.15 During deposition, however, aggregation of a NP species can hamper spatial control of the layer formation and thus strongly influence the thickness.15 This problem is even more pronounced for systems with strong tendency for agglomeration, such as larger NPs and nanorods (NRs), which limits their application to controlled surface assembly.16 Various strategies have been reported for stabilizing nanofluids and dispersions, including electrostatic stabilization and the use of surfactants.17 The main benefit of the core-shell approach, however, is that no further surfactants are required for stabilization of the dispersion, and functional groups can be introduced in order to systematically alter the chemical properties of the nanomaterial.10 Previous research in this realm has shown that the utilization of glycol phosphonic acids (PAs) as a ligand can increase the dispersability of metal oxide NPs in polar solvents significantly.18 The enhanced dispersability of the NPs is found to have a beneficial effect on the film formation properties during a spray coating process.19 In this work we introduce the concept of mixed ligand shells of glycol PAs and alkyne terminated PAs as a means to increase the dispersability of TiO2 NRs in methanol (MeOH), while effectively retaining their functionality for the CuAAC reaction. Thereby, the strong affinity of PAs towards metal oxide substrates is a key aspect for the controlled grafting of mixed ligand shells.20,21 We show that the dispersability of the TiO2 NRs can be enhanced, and at the same time their reactivity for the CuAAC can be preserved. The presented mixed ligand shell

ACS Paragon Plus Environment

3

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 23

approach is therefore an easy and universally applicable concept for customizing dispersions of nanomaterials. The improved dispersability of the functionalized NRs results in pronounced control over the film formation based on the CuAAC. The full control in film formation of even larger nanomaterials that still exhibit reactive sites on the surface may enable a generic technique towards controlled assembly of gradient materials. The introduction of a defined nm and µm sized porosity gradient into a material can have a dramatic influence on the mechanic, electronic, and optical properties of the material.22,23 EXPERIMENTAL SECTION: iso-Propanol (i-PrOH, ≥99.5%, Carl ROTH) and methanol (MeOH,

≥99%,

Carl

ROTH)

were

used

as

purchased.

Phosphonic

acids

12-

azidododecylphosphonic acid (≥95%; CAS 721457-32-5; Sikémia), (2-{2-[2-methoxyethoxy]ethoxy}ethyl)phosphonic acid (≥95%; CAS 96962-42-4; Sikémia), 10-undecynylphosphonic acid (≥95%; CAS 1220675-30-8; Sikémia), and (12,12,13,13,14,14,15,15,16,16,17,17,18,18,18pentadecafluorooctadecyl)phosphonic acid (custom synthesis by Dr. Schlörholz) were used as received. Titanium (IV) oxide nanorods (TiO2 NRs, 1 wt% aqueous solution; diameter 20–40 nm, length 100 nm; PlasmaChem GmbH) were diluted to 0.2 wt% aqueous solution. Dynamic light scattering (DLS) and zeta potential (Zetasizer Nano, Malvern) was measured for dispersions of TiO2 NRs in MeOH (0.015 wt%). Each presented data point is a mean value of three independent measurements; each measurement consists of 15 scans. In case of the pristine TiO2 NRs, an aqueous dispersion of 0.015 wt% was used for DLS and zeta potential measurements. Functionalization of TiO2 NRs: The according volumetric ratio of glycol PA and alkyne PA was prepared, by mixing the corresponding volumes of the solubilized PAs with the same concentration (overall volume 3 mL, 10 mM in i-PrOH). For instance, NRs with a ligand shell of

ACS Paragon Plus Environment

4

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

50% alkyne were functionalized from a mixture formed from glycol PA (1.5 mL, 10 mM in iPrOH) and alkyne PA (1.5 mL, 10 mM in i-PrOH). Pristine TiO2 NRs (10 mL, 0.2 wt% aq.) were then added. Note that the NRs were added after the PA solutions were thoroughly mixed, in order to guarantee for a homogeneous distribution of the ligands over the NR sample. The heterogeneous mixture was sonicated (30 min) followed by centrifugation (12000 rpm, 10 min). The supernatant was removed and the NRs were again dispersed in MeOH (10 mL) by sonication for washing. Three cycles of centrifugation (12000 rpm, 10 min), removing of the supernatant and dispersing (10 mL MeOH, sonication) completed the washing of the NRs. Finally, NRs 1 were dispersed in MeOH (10 mL), yielding a 0.2 wt% dispersion of TiO2 NRs. Region-selective deposition of TiO2 NRs by CuAAC: CuSO4•5H2O (256 µL, 5.0 mM in MeOH) and sodium-ascorbate (512 µL, 5.0 mM in MeOH) were added to degassed MeOH (12 mL). Core-shell functionalized TiO2 NRs (1 mL, 0.2 wt% in MeOH) were added to the solution, followed by the substrate wafer. The mixture was sonicated for 2 h, followed by stirring for 18 h under N2 inert atmosphere at 40 °C. The substrate was then removed from the reaction mixture and rinsed with MeOH (10 mL) and i-PrOH (3 x 10 mL) to yield layers of TiO2 NRs covalently bound to the substrate. In case of the time dependent study, TiO2 NRs with 1% alkyne and 99% glycol in the ligand shell were used and the reaction was carried out for the indicated time t. The coverage density was determined graphically from SEM images (5,000x magnification) using the software ImageJ. RESULTS AND DISCUSSION: The TiO2 NRs were chosen for the present study due to their unique properties, promising for potential application in electronic devices,24 and their rod shape (~40 x 100 nm), which is easy to monitor with atomic force microscopy (AFM) and scanning electron microscopy (SEM). Additionally, the TiO2 NRs were an attractive target for the present

ACS Paragon Plus Environment

5

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 23

study, due to their intrinsically poor dispersability in MeOH, serving as an ideal platform to monitor the impact of the used mixed ligand shells. The decreased dispersability of rod-like NOs in comparison to spherical NPs has been discussed in literature already.25 The binding of PAs to TiO2 has been previously described,19,26,27 and a functionalization protocol was adapted to the TiO2 NRs based on previous work.18 Different molar ratios of alkyne PA 2 and glycol PA 3 were explored for the formation of functionalized TiO2 NRs 1 with a mixed ligand shell (Figure 1). Infrared spectroscopy (IR) data for the functionalization of 1 can be found in the SI†. Figure 1a shows the dynamic light scattering (DLS, black trace) and zeta potential measurements (green trace) of TiO2 NRs 1 with 0% (1(0)), 1% (1(1)), 10% (1(10)), 50% (1(50)), and 100% (1(100)) of alkyne PA 2 in MeOH, compared to an aqueous dispersion of pristine TiO2 NRs. Note that the designation 1(X) refers to NRs 1 functionalized with X% molar fraction of alkyne PA 2 and 100–X% of glycol PA 3. Based on the mean diameter d as measured by DLS (including standard deviations), it is clear that the TiO2 NRs, functionalized with 100% alkyne PA 2 (1(100)) are not dispersible in MeOH (Figure 1a, black trace). Decreasing the molar fraction of 2 in the ligand shell of 1, while increasing the molar fraction of glycol PA 3, however, leads to a significant decrease in d, as determined by DLS. Finally, the diameter d of NRs 1(1) and 1(0) reaches the initial value of the pristine TiO2 NRs. This trend is further supported by the general appearance of the dispersions of NRs 1 in MeOH (Figure 1b). TiO2 NRs 1 functionalized with 2 in a molar fraction higher than 5% precipitate from solution within minutes if not stirred or sonicated, whereas dispersions of 1 bearing 0–5% of alkyne ligand 2 are reasonably stable for about two days. The zeta potential measurements show a similar trend as found for the DLS analyses (Figure 1a). Namely, pure TiO2 NRs show a high zeta potential, consistent with a stable dispersion (and small diameter).18,28 A decrease in zeta potential is found for aggregates of

ACS Paragon Plus Environment

6

Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

1(100). The zeta potential then increases again in absolute value with the decreasing molar fraction of 2 consistent with stabilization of the dispersed NRs 1 in MeOH.

Figure 1. a) Dynamic light scattering (DLS, black trace) and Zeta potential (green trace) measurements of TiO2 NRs 1 with mixed ligand shells of different molar ratios of alkyne PA 2 and glycol PA 3 in MeOH (pristine TiO2 NRs were measured as an aqueous dispersion); b) Schematic depiction of TiO2 NRs 1 functionalized with varying molar ratios of 2 and 3, photograph of dispersions of NRs 1 (0.2 wt%) in MeOH functionalized with different molar ratios of 2 and 3. The chemical behavior of TiO2 NRs 1 was then examined based on the ligand shell composition, and NRs 1(100), 1(50), 1(10), 1(1), and 1(0) were used for region-selective layer deposition by CuAAC according to Scheme 1 to give samples 8(100), 8(50), 8(10), 8(1) and 8(0), respectively. Note that the designation 8(X) again refers to samples 8, on which NRs 1(X)

ACS Paragon Plus Environment

7

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 23

with X% of alkyne PA 2 in their ligand shell were deposited according to Scheme 1. Heavily pdoped silicon substrates 4 (terminated with 10 nm of AlOx by atomic layer deposition, ALD) were patterned in a lithographic step with semi-perfluorinated PA 6 and CuAAC-active azide PA 7, yielding SAM patterned substrates 5. The details of the lithographic patterning of step a) can be found in the SI†.15 Segregation of the fluorinated, passivating PA 6 and the reactive azide PA 7 controls the region-selective deposition of NRs 1. Namely, the CuAAC reaction of alkyne terminated NRs 1 and the azide terminated domains of 5 provide for selective deposition of layers of 1, to give samples 8. Scheme 1 shows sample 8(100), functionalized with 1(100) bearing 100% molar fraction of 2 in the NR ligand shell.

ACS Paragon Plus Environment

8

Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Scheme 1. Region-selective deposition of mixed ligand shell TiO2 NRs 1 by CuAAC on SAM patterned substrates 5 to give functionalized samples 8; only the deposition of NPs 1(100) is shown, giving functionalized sample 8(100); conditions for lithographic patterning (a) can be found in the SI†; the procedure for deposition of NRs 1 by CuAAC of (b) are described in the Experimental Section. Samples 8(100), 8(50), 8(10), 8(1), and 8(0) were characterized by SEM (Figure 2). The SEM images (and expansions) show perfect region-selectivity of the deposition.15 Comparing the images, the impact of the dispersability on the layer formation is significant. Loose islands on samples 8(100) and 8(50) using NRs 1(100) and 1(50), become increasing dense, giving an almost perfectly closed layer when the molar fraction of 2 is reduced to 10% on NRs 1(10), giving 8(10). The layers of 8(1) using NRs 1(1), appear less dense, yet more homogeneous than the for 10% sample (8(10)). This trend can be rationalized by comparing the insets, showing heavy bundling of the NRs of 8(100) and 8(50) as consequence of their stronger agglomeration / aggregation. This bundling is less pronounced for 8(10) where NRs 1(10) are used, while the bundles disappear for NRs 1(1) on sample 8(1). The resulting layer occurs as region-selective assembly of almost isolated NRs. The plot in Figure 2 summarizes a statistical evaluation of the

ACS Paragon Plus Environment

9

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 23

layer coverage of samples 8 in dependence on the molar fraction of 2 used on NRs 1. Decreasing the amount of 2 on the NRs 1 from 100% to 50% causes a slight decrease in the coverage, i.e., 8(50) in comparison to 8(100). The coverage, determined in percent reaches a maximum for 8(10) and drops again for 8(1). Using 0% of 2 on NRs 1(0) (i.e., 100% of the glycol 3), a mean coverage of about 1% is observed. This value can be attributed exclusively to aggregation of NRs 1(0) to the substrate, as in this case no reactive groups are present on the rods that can undergo the CuAAC. The dense coverage of samples 8(10) might be a combined effect of reactivity and aggregation. The amount of 10% of 2 on NRs 1(10) is suitable for the CuAAC, but the dispersability of the rods is worse compared to NRs 1(1) (Figure 1). The stronger aggregation of NRs 1(10) might cause the denser pattern, leading to the observed higher coverage. The pronounced aggregation of samples 8(10), compared to the 8(1) was also monitored by AFM and is summarized in the SI†. Samples 8(1) feature monolayer binding (confirmed by AFM, SI†) of the NRs 1(1) with a mean coverage of about 26%, which is lower than for 8(10), as can be seen by the plot in Figure 2. Yet the monolayer character of samples 8(1) indicates that little to no aggregation is present for these samples and the CuAAC is the only origin of material deposition on 8(1). This statement is consistent to the negligible mean coverage of about 1% for the NRs 1(0). As both the NRs 1(0) and 1(1) have similar properties according to DLS and zeta potential (Figure 1), the only reason for the significantly higher coverage of 8(1) can be the CuAAC causing covalent immobilization of the NRs 1(1).

ACS Paragon Plus Environment

10

Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 2. a) SEM scans of 8(100), 8(50), 8(10), and 8(1), layered region-selectively with NRs 1(100), 1(50), 1(10), and 1(1); insets show expansions (2.8 x 2.1 µm) in each case. b) Plot of coverage density as mean values and standard deviations of 3–6 measurements (values were determined with the analysis package ImageJ and details can be found in the SI†).

ACS Paragon Plus Environment

11

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 23

For such monolayer films, the Langmuir model can be applied in order to investigate the kinetics behind the film formation.29,30. The major aspects that define Langmuir kinetics are monolayer binding to a finite number of equivalents and the presence of equivalent adsorption sites. Furthermore, there should be no lateral diffusion or interaction of the bound or adsorbed species.31 All of these points can be applied to our covalently immobilized TiO2 NR films on samples 8(1). The mixed ligand shell on 1(1) renders the TiO2 NRs sufficiently dispersible to prevent aggregation, and the used of 1% of alkyne PA 2 covalently links the NRs to the substrate creating a monolayer that inhibits lateral diffusion. The Langmuir isotherm can thus be expressed by the following equation as

 =

 ∙  ∙  1+∙

with q being the coverage (as a value between 0–1), Q being the maximum coverage, K being the equilibrium constant (defined as kdesorp/kadsorp) and t being the reaction time. Linearization of the above mentioned equation yields the following term, which allows the extraction of K and Q by a linear fit of the measured dependence of q on t.31

1 1  = + ∙  ∙  Figure 3 summarizes the measured coverage in dependence on the reaction time t of samples 8(1) on which NRs 1(1) were immobilized. Figure 3a thereby provides an overview over the evolution of the coverage on samples 8(1) in dependence on t, including the standard deviation (as a mean value of 3–4 data points). The value for the coverage of about 1% at 0 h was taken from NRs 1(0) (8(0), Fig. 2), which cannot undergo the CuAAC. This value is attributed solely

ACS Paragon Plus Environment

12

Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

to aggregation and serves as a baseline, since this 1% of coverage will occur on the not passivated sites (the azide PA 7 pattern, Scheme 1) as soon as the substrate 5 is exposed to NRs 1(1). Close examination of the plot in Figure 3a shows an increase in the layer coverage at longer reaction times, which is not expected according to the Langmuir model. This deviation can be rationalized by considering the possible overlap of the NRs after immobilization, which causes a variation in the layer thickness that is not strictly considered by the classical Langmuir model. Furthermore, steric hindrance from the NRs, already bound to the substrate, makes vacant binding sites less accessible for NRs not yet immobilized on the substrate yet. A plot of the height histogram extracted from an AFM measurement of sample 8(1) shows that indeed very little measured pixels on the AFM scans exceed the 40 nm diameter of the NRs, making the latter explanation more feasible (SI†). With increasing coverage, the influence of the steric hindrance becomes stronger, and, thus, giving the observed deviation from the Langmuir model. Nevertheless the obtained layer formation remains in reasonable agreement with the Langmuir kinetics, as can be derived by the plot in Figure 3b. The mean values for the coverage presented in Figure 3a were taken for the linearized plot in Figure 3b (except q for 0 h), and fit to the Langmuir model. The linear fit from Figure 3b yields a theoretical maximum coverage Q of about 49% (for NRs 1(1)), which is also in reasonable accordance to the samples displayed in Figure 2. For the case in which NRs 1(1) were deposited by CuAAC, no coverage above 42% was observed so far (SI†). The rather low maximum coverage that is observed might be explained by considering the steric hindrance caused by the TiO2 NRs covalently bound to the substrate (vide infra). The extracted value for the equilibrium constant K = 0.0967 h–1 should, however, be interpreted with care. In case of the proposed deposition technique using the CuAAC reaction, multiple processes run in parallel until the NRs

ACS Paragon Plus Environment

13

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 23

1 are deposited on the surface.32 The presence of several independent processes (such as catalyst reduction, the interaction between alkyne and catalyst, surface adsorption / desorption, and chemical bond formation) makes the interpretation of the obtained equilibrium constant K a complex issue. The value of 0.0967 h–1 shows that the equilibrium is strongly on the side of the bound species, considering that K is defined as kdesorp/kadsorp. This finding is in accordance to that expected in the case of covalent immobilization of the NRs by the CuAAC reaction. Observations showed indeed that the coverage on the substrates 8(1) is not influenced by additional rinsing with MeOH after the functionalization protocol and remains stable with no observed desorption of the TiO2 NRs 1(1). The time dependent study of the coverage q on samples 8(1) yielded a reproducible correlation between the reaction time t and the coverage q and could be fit to the known Langmuir model. This trend nicely shows that enhanced controllability of the layer formation was obtained by increasing the dispersability of the used NRs 1 with the mixed ligand shell approach.

ACS Paragon Plus Environment

14

Page 15 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 3. a) Plot of the coverage vs the reaction time t including standard deviations from 3–4 measurements for each data point. b) Linearized plot according to the Langmuir model and extracted values for the equilibrium constant K (as kdesorp/kadsorp) and the maximum coverage Q; value for 0 h left out for Langmuir fit. The potential of layer-by-layer growth on deposited NRs has been demonstrated for a simple case (sample 9 - SI†) of NR layers 8(100) by using functionalized TiO2-NPs with SAM providing the complementary binding motif (12-azidododecylphosphonic acid 7). Even though the initial NR layer is of poor quality, the experiment shows clearly that spherical NPs undergo a selective CuAAC reaction to growth on NR surface only and not on the substrate.

ACS Paragon Plus Environment

15

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 23

SUMMARY AND CONCLUSION: In summary, we have shown that tailor-made mixed ligand shells can be used to control the dispersability and the resulting film formation properties of NRs. By adding solubilizing glycol chains to the ligand shell, the dispersability of core-shell nanomaterials can be enhanced significantly, with preserving the reactivity of alkyne termination at the same time. We further demonstrated that the composition of the shell of the NRs, and resulting dispersability, has a strong impact on the layer formation using the CuACC reaction. By conducting a time dependent study of the CuACC deposition, we observed a trend in the increase of the layer coverage, which can be described by the Langmuir model. This finding demonstrates that a considerable amount of control over the layer formation can be achieved through controlling dispersability of NRs via a mixed ligand shell.

ASSOCIATED CONTENT † Supporting Information: Lithographic pattering process, IR data, Additional SEM and AFM data, and listed values used for the plots in Figure 2 and Figure 3 can be found in the Supporting Information (SI). This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Authors * E-Mail: [email protected] * E-Mail: [email protected] Funding Sources

ACS Paragon Plus Environment

16

Page 17 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

This project was funded by the Cluster of Excellence, Engineering of Advanced Materials (EAM) funded by the Deutsche Forschungsgemeinschaft (DFG). ACKNOWLEDGMENT Sebastian H. Etschel is supported by the Graduate School Molecular Science (GSMS) and the Graduate School Advanced Materials and Processing (GS-AMP).

ACS Paragon Plus Environment

17

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 23

REFERENCES (1)

Xia, Y.; Whitesides, G. M. Soft Lithography. Angew. Chem. Int. Ed. 1998, 37, 550–575;

Angew. Chem. 1998, 110, 568–594. (2)

Tornøe, C. W.; Christensen, C.; Meldal, M. Peptidotriazoles on Solid Phase. J. Org.

Chem. 2002, 67, 3057–3064. (3)

Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. A Stepwise Huisgen

Cycloaddition Process: Copper(I)-Catalyzed Regioselective "Ligation" of Azides and Terminal Alkynes. Angew. Chem. Int. Ed. 2002, 114, 2708–2711; Angew. Chem. 2002, 114, 2708–2711. (4)

Meldal, M.; Tornøe, C. W. Cu-catalyzed azide-alkyne cycloaddition. Chem. Rev. 2008,

108, 2952–3015. (5)

Lummerstorfer, T.; Hoffmann, H. Click Chemistry on Surfaces. J. Phys. Chem. B 2004,

108, 3963–3966. (6)

Collman, J. P.; Devaraj, N. K.; Chidsey, C. E. D. “Clicking” Functionality onto Electrode

Surfaces. Langmuir 2004, 20, 1051–1053. (7)

Binder, W. H.; Sachsenhofer, R. ‘Click’ Chemistry in Polymer and Materials Science.

Macromol. Rapid Commun. 2007, 28, 15–54. (8)

Evans, R. A. The Rise of Azide–Alkyne 1,3-Dipolar 'Click' Cycloaddition and its

Application to Polymer Science and Surface Modification. Aust. J. Chem. 2007, 60, 384–395. (9)

Haensch, C.; Hoeppener, S.; Schubert, U. S. Chemical modification of self-assembled

silane based monolayers by surface reactions. Chem. Soc. Rev. 2010, 39, 2323–2334.

ACS Paragon Plus Environment

18

Page 19 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(10) Nicosia, C.; Huskens, J. Reactive self-assembled monolayers. Mater. Horiz. 2014, 1, 32– 45. (11) Binder, W. H.; Kluger, C. Combining Ring-Opening Metathesis Polymerization (ROMP) with Sharpless-Type “Click” Reactions. Macromolecules 2004, 37, 9321–9330. (12) Upadhyay, A. P.; Behara, D. K.; Sharma, G. P.; Bajpai, A.; Sharac, N.; Ragan, R.; Pala, R. G. S.; Sivakumar, S. Generic process for highly stable metallic nanoparticle-semiconductor heterostructures via click chemistry for electro/photocatalytic applications. ACS Appl. Mater. Interfaces 2013, 5, 9554–9562. (13) Rianasari, I.; Jong, M. P. de; Huskens, J.; van der Wiel, Wilfred G Covalent Coupling of Nanoparticles with Low-Density Functional Ligands to Surfaces via Click Chemistry. Int. J. Mol. Sci. 2013, 14, 3705–3717. (14) Matyjewicz, J.; Lesniewski, A.; Niedziolka-Jonsson, J. Click chemistry modification of glassy carbon electrode with gold nanoparticles for electroactive ion discrimination. Electrochem. Commun. 2014, 48, 73–76. (15) Etschel, S. H.; Portilla, L.; Kirschner, J.; Drost, M.; Tu, F.; Marbach, H.; Tykwinski, R. R.; Halik, M. Region-Selective Deposition of Core-Shell Nanoparticles for 3D Hierarchical Assemblies by the Huisgen 1,3-Dipolar Cycloaddition. Angew. Chem. Int. Ed. 2015, 54, 9235– 9238; Angew. Chem. 2015, 127, 9367–9370. (16) Li, X.; Zhu, D.; Wang, X. Evaluation on dispersion behavior of the aqueous copper nanosuspensions. J. Colloid Interface Sci. 2007, 310, 456–463.

ACS Paragon Plus Environment

19

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 23

(17) Yu, W.; Xie, H. A Review on Nanofluids: Preparation, Stability Mechanisms, and Applications. J. Nanomater. 2012, 2012, 1–17. (18) Portilla, L.; Halik, M. Smoothly tunable surface properties of aluminum oxide core-shell nanoparticles by a mixed-ligand approach. ACS Appl. Mater. Interfaces 2014, 6, 5977–5982. (19) Portilla, L.; Etschel, S. H.; Tykwinski, R. R.; Halik, M. Green processing of metal oxide core-shell nanoparticles as low-temperature dielectrics in organic thin-film transistors. Adv. Mater. 2015, 27, 5950–5954. (20) Bauer, T.; Schmaltz, T.; Lenz, T.; Halik, M.; Meyer, B.; Clark, T. Phosphonate- and carboxylate-based self-assembled monolayers for organic devices: a theoretical study of surface binding on aluminum oxide with experimental support. Appl. Mater. Interfaces 2013, 5, 6073– 6080. (21) Bhairamadgi, N. S.; Pujari, S. P.; Trovela, F. G.; Debrassi, A.; Khamis, A. A.; Alonso, J. M.; Al Zahrani, A. A.; Wennekes, T.; Al-Turaif, H. A.; van Rijn, C.; Alhamed, Y. A.; Zuilhof, H. Hydrolytic and Thermal Stability of Organic Monolayers on Various Inorganic Substrates. Langmuir 2014, 30, 5829–5839. (22) Lu, K. Making strong nanomaterials ductile with gradients. Science 2014, 345, 1455– 1456. (23) Ryckman, J. D.; Jiao, Y.; Weiss, S. M. Three-dimensional patterning and morphological control of porous nanomaterials by gray-scale direct imprinting. Sci. Rep. 2013, 3, 1502.

ACS Paragon Plus Environment

20

Page 21 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

(24) Zhou, Y.; Hu, J.; Dang, B.; He, J. Titanium oxide nanoparticle increases shallow traps to suppress space charge accumulation in polypropylene dielectrics. RSC Adv 2016, 6, 48720– 48727. (25) Pardo, P.; Calatayud, J. M.; Alarcón, J. Improvement of boehmite nanoparticles' aqueous dispersability by controlling their size, shape and crystallinity. RSC Adv. 2014, 4, 48389–48398. (26) Zeininger, L.; Petzi, S.; Schonamsgruber, J.; Portilla, L.; Halik, M.; Hirsch, A. Very Facile Polarity Umpolung and Noncovalent Functionalization of Inorganic Nanoparticles: A Tool Kit for Supramolecular Materials Chemistry. Chem. Eur. J. 2015, 21, 14030–14035. (27) Zeininger, L.; Portilla, L.; Halik, M.; Hirsch, A. Quantitative Determination and Comparison of the Surface Binding of Phosphonic Acid-, Carboxylic Acid-, and CatecholLigands on TiO2 Nanoparticles. Chem. Eur. J. 2016, DOI:10.1002/chem.201601920. (28) Jiang, J.; Oberdörster, G.; Biswas, P. Characterization of size, surface charge, and agglomeration state of nanoparticle dispersions for toxicological studies. J. Nanopart. Res. 2009, 11, 77–89. (29) Moaz, R.; Sagiv, J. On the Formation and Structure of Self-Assembling Monolayers. J. Colloid Interface Sci. 1984, 100, 465–496. (30) Van der Boom, M. E.; Evmenenko, G.; Yu, C.; Dutta, P.; Marks, T. J. InterruptedGrowth Studies of the Self-Assembly of Intrinsically Acentric Siloxane-Derived Monolayers. Langmuir 2003, 19, 10531–10537. (31) Foo, K. Y.; Hameed, B. H. Insights into the modeling of adsorption isotherm systems. Chem. Eng. J. 2010, 156, 2–10.

ACS Paragon Plus Environment

21

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 23

(32) Iacobucci, C.; Reale, S.; Gal, J.-F.; Angelis, F. de Dinuclear copper intermediates in copper(I)-catalyzed azide-alkyne cycloaddition directly observed by electrospray ionization mass spectrometry. Angew. Chem. Int. Ed. 2015, 54, 3065–3068; Angew. Chem. 2015, 127, 3108– 3111.

ACS Paragon Plus Environment

22

Page 23 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

TOC:

ACS Paragon Plus Environment

23