Chemical Cross-Linking of Anatase Nanoparticle Thin Films for

May 3, 2018 - ... higher homogeneity and their structural integrity is preserved during .... hardness and roughness data of the coatings; FTIR data of...
0 downloads 3 Views 2MB Size
Subscriber access provided by Kaohsiung Medical University

Interface Components: Nanoparticles, Colloids, Emulsions, Surfactants, Proteins, Polymers

Chemical cross-linking of anatase nanoparticle thin films for enhanced mechanical properties Apostolos Salmatonidis, Jutta Hesselbach, Gerhard Lilienkamp, Tobias Graumann, Winfried Daum, Arno Kwade, and Georg Garnweitner Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b00479 • Publication Date (Web): 03 May 2018 Downloaded from http://pubs.acs.org on May 4, 2018

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 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 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.

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

Chemical cross-linking of anatase nanoparticle thin films for enhanced mechanical properties A. Salmatonidis,1 J. Hesselbach,1 G. Lilienkamp,2 T. Graumann,3 W. Daum,2 A. Kwade,1 G. Garnweitner1,4,*

1

Institute for Particle Technology, Technische Universität Braunschweig, 38106 Braunschweig, Germany

2

Institute for Energy Research and Physical Technologies, Technische Universität Clausthal, 38678 Clausthal-Zellerfeld, Germany

3

Fraunhofer Institute for Surface Engineering and Thin Films, 38108 Braunschweig, Germany

4

Laboratory for Emerging Nanometrology, Technische Universität Braunschweig, 38106 Braunschweig, Germany

ACS Paragon Plus Environment

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 Titania nanoparticle-based thin films are highly attractive for a vast range of commercial applications. Whilst their application on polymer-based substrates is particularly appealing, the requirement of low process temperatures results in low mechanical stability. Highly crystalline anatase nanoparticles were used as the building blocks for coatings through a two-stage process. The main benefits of this method, over the more common sol-gel ones, are the relatively low temperature required for the production of metal oxide coatings, allowing the use of polymer-based substrates, and the defined crystallinity of the resulting thin films. Although in several cases moderate temperatures can be utilized for drying the films, the mechanical stability of the respective coatings remains a critical issue. In this contribution, we present a strategy to achieve a network formation between TiO2 nanoparticles in a pre-formed thin film, based on the cross-linking of the functionalized nanoparticles. In a first stage, the nanoparticles were functionalized by dicarboxylic acids, concurrently leading to a stable colloidal dispersion that could be utilized for dip-coating to obtain TiO2 thin films with high homogeneity and optical transparence. During the second stage, the films were immersed in a solution of a diamine as linker molecule, in order to achieve a cross-linking between the nanoparticles within the film. It is demonstrated that indeed covalent bonding was realized, and functional coatings with significantly enhanced mechanical properties were obtained by our strategy.

1 ACS Paragon Plus Environment

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

1.

Introduction

Titanium dioxide (TiO2) is considered to be one of the most promising and highly investigated nanomaterials of the recent years due to its unique combination of properties, such as high refractive index, favorable band gap positions and oxidation potential, high hardness, chemical stability, and non-toxicity. The applications of TiO2 nanoparticles (NPs) are manifold,1–3 e.g. in pigments and paints,4 in cosmetic products and sunscreens,5–7 as well as in a wide range of products due to their photocatalytic activity (e.g. for water and air remediation, self-cleaning, anti-fogging and antibacterial surfaces).8–13 The photocatalytic activity of TiO2 was firstly introduced by the pioneering work of Fujishima & Honda on the electrochemical photolysis of water achieved in a photoelectrochemical cell with a Pt electrode described by a simple reaction equation:14



  + 2 ℎ →  + 

(Eq. 1),

which attracted extensive research attention and boosted the field of photocatalysis, resulting in numerous scientific articles.8,15–17 Whilst in many scientific reports the photocatalytic properties of TiO2 NPs are studied using particle dispersions, for practical applications the immobilization of the NPs in the form of coatings and thin films on a substrate is highly desired to facilitate NP handling and reduce risks of loss and contamination. Sputtering and chemical vapor deposition (CVD) can be used to obtain TiO2 thin films,18 however these techniques require expensive equipment and high energy input, and often result in smooth thin films with relatively low surface area and thus, low activity. Sol-gel manufactured TiO2 thin-films are commonly produced but, in order to achieve a high degree of crystallization of TiO2 – which is necessary to ensure high photocatalytic activity – as well as good mechanical stability, a calcination treatment is required.19,20 This however constitutes a stern limitation on the materials that can be used as substrate, and implies a number of adverse effects such as uncontrolled crystallite growth and loss of porosity due to sintering.

2 ACS Paragon Plus Environment

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

In the present work we propose an alternative strategy for the manufacturing of TiO2 NPbased coatings with tailored characteristics and high mechanical stability, based on wet chemistry. Abargues et al. have shown that by interconnecting Ag NPs, chemically modified with dicarboxylic acids, their optical properties could be tuned.21 Whilst previously, we have shown that metal oxide nanoparticles can be stabilized by both organic amines and carboxylic acids22,23 and can be sequentially functionalized in a step-by-step process,24 here we utilize the reaction of multifunctional ligands bound to the nanoparticle surface to induce a covalent network formation, in order to improve the mechanical properties of the produced coatings. Initially we synthesize the building blocks of the coatings, the TiO2 NPs of anatase modification, via a non-aqueous sol-gel method ensuring their homogeneous characteristics in terms of crystallinity, primary particle size and morphology.25,26 Subsequently, we chemically modify the anatase NPs in a way that both stable dispersions are obtained and reactive terminal groups are available on the surface of the NPs for further functionalization. As the TiO2 NPs are strongly agglomerated after the synthesis, the chemical surface modification treatment is performed in a planetary ball mill, with a combination of a mechanical disagglomeration process and the adsorption of the stabilizer resulting in colloidally stable particle dispersions. We utilize dicarboxylic acids as surface modifiers, in a way that one carboxylic acid group binds to the surface of the nanoparticles and the second carboxylic acid moiety remains as a terminal group for further reaction. The produced dispersions are then applied to generate homogenous transparent coatings. We thereby introduce a two-stage coating process: firstly, we apply the stable nanoparticle dispersion to produce a “primary” coating (coating stage). The second stage comprises a treatment in a molecular linker solution to achieve covalent linkage between the particles, utilizing the amine-carboxylic acid condensation reaction.

A diamine is used as the

molecular linker between the previously modified nanoparticles. The term linker hereby refers to a cross-linking agent that is used to interconnect two already functionalized nanoparticles, and it should not be confused with the term ligand which we use for the

3 ACS Paragon Plus Environment

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

molecular species that bind to the nanoparticle surface. Lastly, after a post-curing treatment by drying at mild temperatures (120

o

C) the final coatings are obtained. The low

temperatures required in the presented method represent an important advantage over the conventional sol-gel techniques, allowing the potential application on flexible polymer substrates. Moreover, dip-coating was used as the coating process in all stages; the simplicity and low cost of this coating technique renders our method facile and inexpensive. For comparison, a one-stage route was applied using the same building blocks (anatase nanoparticles), ligands and linker molecules (dicarboxylic acids and diamines), solvent, as well as process parameters. During the one-stage route the linker solution is added directly to the stable dispersion of TiO2 NPs and the resulting suspension is used for the fabrication of coatings. A comparison of the results of the two methods revealed that cross-linking occurs for both strategies, but illustrate the benefits and the effectiveness of the two-stage process.

Figure 1. Schematic illustration of the two-stage coating process

4 ACS Paragon Plus Environment

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

Figure 2. Schematic illustration of the one-stage coating process

2.

Experimental

2.1

NP synthesis and functionalization

Anatase nanoparticles were prepared via a non-aqueous sol-gel method in a closed reactor system (Polyclave type 3/1, Büchi Glas Uster) as was comprehensively described by Zimmermann et al. [13,14]. Specifically, the TiO2 nanoparticles were synthesized at a temperature of 200 oC utilizing titanium isopropoxide (Ti(OiPr)4, 97%, Sigma Aldrich) as precursor at a concentration of 560 mmol/L in the solvent benzyl alcohol (BnOH 99%, Sigma Aldrich), with an agitator speed of 250 RPM. The resulting nanoparticles were retrieved from the reaction medium by centrifugation (10 minutes at 7000 RPM) and then washed twice with chloroform. Ethanol (p.a.) was added to the washed precipitated nanoparticles and subsequently, an appropriate ligand (sebacic acid or adipic acid, both 99% from Sigma Aldrich) in powder form was added under stirring in different TiO2-to-ligand molar ratios. The solution was kept under stirring for 1 hour and once the ligand was dissolved completely, it was transferred to an alumina plated stainless steel milling container. The latter was filled 5 ACS Paragon Plus Environment

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

with grinding media (YSZ, 185 µm) to an extent of 50 vol.%. A laboratory scale planetary ball mill (Retsch PM400) was used with settings at 300 RPM for 6 hours. This process has several benefits; it promotes de-agglomeration of the anatase nanoparticles, it disperses the nanoparticles in the solvent, while at the same time a chemical NP surface functionalization takes place and a stable dispersion is obtained at the end of the process. The resulting dispersion is collected by careful removal of the supernatant in order to avoid mixing with the sedimented grinding media.

2.2

Fabrication of thin films

The coating technique that we used in all cases was dip-coating (Coater 4, IDLab, withdrawal at 500 mm min-1). Barth and Zimmermann have employed a similar method using different types of TiO2 NPs to observe the effect of the synthesis route on the characteristics of the applied coatings.27 In the present work an identical coating protocol has been used in all cases in order to acquire comparable results to that study. Standard objective glass slides were coated with the anatase dispersions and after drying (120 °C, 2 h) the acquired coatings were dipped into ethanol linker solutions with different concentrations. Two different diamines (linkers) were used in our experiments, hexamethylenediamine (HE, 98%, Sigma Aldrich), and p-phenylenediamine (PDA, 98%, Sigma Aldrich).

2.3

Characterization 2.3.1

Anatase dispersions

The anatase nanoparticle dispersions were characterized in terms of secondary particle size and particle size distribution with dynamic light scattering (DLS, Nanophox, Sympatec GmbH). The dicarboxylic acid-functionalized TiO2 nanoparticles were precipitated after intensive centrifugation and the addition of 1 ml distilled H2O (as non-solvent), washed two times with ethanol (technical grade), and left to dry overnight at room temperature in a 6 ACS Paragon Plus Environment

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

vacuum desiccator. In the case of the one-stage route, samples were also obtained after the linker (diamine) addition, following the same drying process as above. The obtained powders were quantitatively analyzed in terms of surface chemistry by thermogravimetric analysis (TGA, Mettler Toledo TGA/SDTA851) and FTIR spectroscopy (Bruker Vertex 70). 2.3.2

Anatase coatings

The mechanical properties of the coatings were determined by nanoindentation using a Berkovich tip (TI900, Hysitron, Inc.). The hardness was calculated based on measured mean maximal force of forty measurements per sample. The detailed measurement procedure was described by Barth et al.28 Coating roughness was measured in order to identify the influence of different formulation parameters (e.g. the ligand-to-linker ratio) on the structural properties of the respective coatings. Five roughness scans were performed on each sample using AFM (NanoWizard III BioAFM, JPK) with a gold coated cantilever (75 kHz, 3Nm–1, Multi75GD-G). X-ray photoelectron spectroscopy (XPS) was applied for an investigation of the elemental and chemical composition of the coatings using an Al Kα x-ray photon source and the hemispherical electron energy analyzer of a scanning Auger microscope (Omicron NanoSAM). Photocatalytic activity measurements were performed according to ISO 10678:2010 by the degradation of methylene blue in an aqueous solution (see Supporting Information section for further details).

3.

Results and discussion

3.1

Anatase nanoparticle dispersions

The self-synthesized TiO2 nanoparticles are of anatase modification with high crystallinity and a crystallite size of 13.2 nm (calculated from the PXRD measurements using Scherrer's equation for the (011) reflection)27 and were synthesized according to previous reports from our group.26,29 The two main ligands used were adipic acid (AdA) and sebacic acid (SebA).

7 ACS Paragon Plus Environment

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

In Figure 3 the ATR-FTIR spectra of the functionalized anatase nanoparticles are presented and as can be observed, for both ligands similar spectra are obtained, pointing to a successful surface modification. Previous studies have demonstrated that carboxylic acids can strongly bind to the surface of metal oxide nanoparticles through carboxylate ligand-tosurface interactions.23,30,31 The presence of peaks at 1568 cm -1 assigned to the asymmetric (vas) and at 1447 cm-1 assigned to the symmetric (vs) stretching vibrations (Figure 3), confirms that the acids are indeed strongly bound to the NPs. Furthermore, the peak at 1739 cm -1 assigned to the C=O stretching vibration and the broad peak at 3200 cm -1 attributed to the OH-stretching vibration indicate that the second carboxylic moieties of the ligands are not bound.32 Thus, we can conclude that the dicarboxylic acids have successfully been grafted to the anatase NP surface from the one side, while terminal carboxylic groups are also available for further functionalization. The molar ratio of TiO2 to ligand is a key parameter for the preparation of stable dispersions of the functionalized anatase nanoparticles. Figure 4 shows the effect of different molar ratios on the particle size distribution. The functionalized NPs with the same ratio (TiO2: ligand) have almost identical average secondary particle sizes (Q3-50%), while the difference in the chain length of the ligands has a rather minor effect. We have observed that there is an optimum range of molar ratio at about 4-8 (TiO2: ligand), in terms of secondary particle size with mono-modal distribution, solid content as well as stabilization efficiency (see Supporting Information section, Figures S1 and S2). The relative efficiency of the stabilization process (Est) is calculated as the ratio of TiO2 NPs mass quantity before and after the process:

=

 

× 100

(2)

where Min is the initial mass of TiO2 NPs and Mf the mass of the stable NPs in the resulting dispersion. A reasonable statement would be that the dispersed nanoparticle mass cannot exceed the initial one (Mf ≤ Min). The free surface area of NPs can be reduced due to 8 ACS Paragon Plus Environment

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

aggregation, agglomeration, or due to the strong binding to other organic molecules (e.g. benzyl alcohol) and thus, the binding of ligands to the surfaces of NPs can be limited. Consequently, some NPs will not achieve the needed level of surface modification in order to reach colloidal stability, and they will form a sediment. Additionally, there are always some NPs deposited on the walls of the container or on the dispersing media during the stabilization process. Hence, it applies that Mf < Min. In our previous studies where colloidal stability was the main objective, a simple addition of ligands was reported to successfully lead to the disintegration of agglomerates and the formation of a stable dispersion of primary ZrO2 and ITO nanoparticles.22,23 However, in our case a mechanical treatment was required to achieve a stable dispersion of nanoparticles with tailored characteristics and available functional moieties. After their functionalization, the anatase nanoparticles could not be precipitated or removed even by centrifugation, indicating the high stability of the obtained dispersions. Moreover, the dispersions showed very good long term stability, without any significant sedimentation visible after 2 months’ storage. In the course of the one-stage process, where the linker (diamine) solution was added directly to the stable dispersion of the functionalized TiO2 nanoparticles, an instant crosslinking effect was observed. Figure 5 presents the ATR-FTIR spectrum of pristine HE (black dashed curve), the spectrum of the SebA-modified TiO2 NPs (red curve), and the spectrum of the SebA-modified NPs after addition of HE and stirring for 30 minutes (blue curve). In the case of the SebA-functionalized and crosslinked NPs, the broadening of the signals in the range 3100-3300 cm-1 can be attributed to a combination of the OH-stretching vibration from the carboxylic acid groups, and the H-bonded OH-stretch signal due to alcohol residues. Nevertheless, the HE absorption band at 3330 cm-1, which is assigned to the N-H stretching vibration of the free amino groups, is completely disappeared after the addition of HE to the SebA functionalized NPs, indicating that the HE has completely reacted, resulting in a cross-linking of the nanoparticles. On the other hand, as the dicarboxylic acid is utilized in excess, it has only been partially consumed by the reaction. The latter can be confirmed

9 ACS Paragon Plus Environment

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

by the decreased intensity of the carboxyl peak after the addition of HE, which is also an indication of the underlying reaction between the amine and the carboxyl groups. Finally, the formation of the amide bond, which results in the cross-linkage, is evidenced by the strong peak at 1570 cm-1 that is attributed to the N-H bending vibration.32 Thus, a molecular network formation and linkage of the anatase NPs via covalent binding could be realized. In Figure 6, the effect of cross-linkage on the secondary particle size of the anatase-AdAfunctionalized NPs can be observed. After the addition of the diamines HE and PDA, a clear increase in the secondary particle size can be observed. Thus, we can conclude that the cross-linking among the functionalized NPs is realized through an ester formation as condensation reaction. On the other hand, the extensive growth of the secondary particle size in the case of PDA can be attributed to the formation of larger agglomerates due to higher hydrophobicity of PDA, with the NPs consequently becoming less compatible with the solvent.

Figure 3. ATR-FTIR spectra of the functionalized anatase nanoparticles (dried powders)

10 ACS Paragon Plus Environment

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

Figure 4. Effect of the molar ratio TiO2: ligand on the secondary particle size distribution as determined by DLS

Figure 5. ATR-FTIR spectra of the dried NPs after different process stages. SebA-functionalized NPs (red curve), cross-linked NPs (blue curve) and diamine reference (black dashed curve)

11 ACS Paragon Plus Environment

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

Figure 6. Secondary particle size distribution of the functionalized anatase NPs before (black curve) and after the additon of the linker (red, HE; blue, PDA) as determined by DLS

3.2

Anatase thin films

Figure 7. Photographs of thin films with SebA-modified TiO2 NPs. Primary coating (1), one-stage coating with HE (2), two stage coating with PDA (3), two stage coating with HE (4)

The final thin films obtained from the two-stage process were transparent and homogenous, in contrast to the films obtained from the one-stage coating process (Fig. 7, number 2) which

12 ACS Paragon Plus Environment

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

were non-uniform and showed lower transparency. A comparison of the microstructure of the films resulting from different process stages is displayed in Figure 8, where from the scanning electron microscopy images one can clearly observe the effect of the process and the structural alterations of the coatings. The surface of the one-stage coating is inhomogeneous with a very rough surface and somewhat “flower-like” microstructure (Fig. 8a and Fig. S7), while the two-stage coating shows a much smoother and uniform surface topography (Fig. 8b, c). Moreover, looking at the cross-section of the two-stage coating (Fig. 8c) we can observe a slight difference in the porosity of the top layer, which seems denser in comparison to the lower part of the coating. We attribute this difference to the cross-linking of the anatase NPs, localized only at the top surface, since – due to the linker treatment process – possibly, the linker does not fully penetrate the film. Hence, the two-stage process results in thin films with improved homogeneity utilizing the same building blocks and organic modifiers as in the one-stage process. This can be explained by the formation of large agglomerates upon linker addition prior to the coating stage, as described above. The coatings obtained from the functionalized NPs without linker have higher homogeneity and their structural integrity is preserved during the second stage, while large agglomerates forming after linker addition to the NP dispersion lead to low-quality coatings in the onestage process. In Fig. 8d the comparative XPS analysis of samples at different stages of the two-stage process is displayed, confirming the presence of nitrogen on the final coatings after addition of the diamines, in contrast to the nitrogen-free spectrum of the primary TiO2SebA coating. The N 1s binding energy of the nitrogen species detected in the final coatings amounts to 400.2 eV which is about 1.3 eV higher than that observed for amine groups in aliphatic hydrocarbons.33–35 Previous XPS results for amine groups in aromatic compounds show larger variations of the N 1s binding energy.35,36 Thus, at least for the case of HE, our observation of a positive N 1s core level shift of the final layer is consistent with a charge transfer due to a reaction of the linker molecule’s amine group with the end group of the SebA coating.

13 ACS Paragon Plus Environment

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

a

b

c

d

Figure 8. SEM image of the one-stage coating (a); surface (b) and cross-sectional SEM images (c) of the final two-stage coating; XPS of the primary and the final two-stage coatings (d)

In Figure 9 the effect of the different molar ratios of TiO2: ligand: linker on the hardness of the two-stage coatings is shown, utilizing SebA as ligand and PDA as linker. For the primary coatings (black squares) a linear trend with a negative slope can be observed, indicating that a higher amount of ligand on the NP surface results in lower hardness, which can be an effect of higher amount of organic material as well as larger aggregates (cf. Figure 4, where larger TiO2: ligand ratios resulted in smaller secondary particle sizes). Thus, NPs functionalized with the lowest ratio of dicarboxylic acid showed the highest hardness. The coatings obtained in the two-stage process with a diamine concentration of 2.5 mg/mL showed significantly higher hardness. For this specific linker concentration, a comparison of hardness between the coatings with lower (TiO2: ligand = 16) and higher (TiO2: ligand = 4) 14 ACS Paragon Plus Environment

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

degree of functionalization shows that the hardness of the coating with higher amount of ligand increased to a greater extent. This is attributed to a higher cross-linking efficiency between the NPs, which becomes even more pronounced when higher linker concentrations are used. For the NPs functionalized with a lower amount of ligand, a higher amount of added linker might induce a kind of oversaturation, where the linker is no longer fully effective and may only partially react, as was observed in a similar fashion for Al2O3-based coatings with an epoxy-based crosslinker recently.37 In a similar fashion, the use of a diamine concentration of 5 mg/mL might result in an oversaturation and lower degree of crosslinking, explaining the low hardness observed in particular for the NPs with low degrees of functionalization. It becomes clear that the optimal ligand: linker ratio is a decisive parameter in order to achieve the highest increase in hardness.

Figure 9. Effect of the linker concentration (PDA) in the post-curing solution, for different TiO2: ligand ratios, on the hardness of the two-stage coatings

15 ACS Paragon Plus Environment

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

Figure 10. Effect of the linker concentration (PDA) in the post-curing solution on the mechanical properties of the coatings obtained from the two-stage process

Figure 10 shows a correlation between roughness and hardness of the coatings, as an indication for the degree of the cross-linking.37 A post-curing of the coatings in a solution with 2.5 mg/mL linker concentration leads to smoother surfaces, as a result of a more homogeneous coating. A linker concentration of 5 mg/mL resulted in slightly higher roughness compared to the primary coatings, probably due to more organic material on the surface. The difference in surface roughness has an influence on the optical properties of the coatings. Higher roughness results in a higher extinction coefficient and thus, lower transparency (Figure S5). A preliminary assessment of the photocatalytic properties of the 2-stage and primary coatings showed that they indeed were photocatalytically active according to ISO 10678:2010, with significantly higher activity than a commercial reference. Both types of the screened coatings exhibited photon efficiencies higher than 0.03%, with the 2-stage coating 16 ACS Paragon Plus Environment

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

having only slightly lower efficiency than the primary one (Fig. S8). Whilst a decrease of activity might be attributed to a slight decrease of surface accessibility of the anatase NP on the surface of the 2-stage coating after the diamine post-curing treatment, the determined values are rather similar and substantially above the photocatalytic activity of the reference system, indicating only a small influence of the cross-linking treatment. During the performed preconditioning and photocatalytic activity tests, we did not observe any degradation of the organic cross-linking that would result in detachment or disintegration of the films in solution. However, further investigations on the long term performance and structural integrity of the coatings under extended UV exposure are required to assess the suitability of our crosslinking strategy also for more demanding applications such as water splitting that would involve gas evolution and pH changes.

4.

Conclusions

In the present work we have introduced a method to enhance the mechanical properties of titania coatings via a covalent cross-linking mechanism based on amide bond formation from the reaction between carboxylic acid and amino groups. The proper functionalization of the anatase nanoparticles by a dicarboxylic acid ligand resulted in free terminal carboxylic acid moieties grafted on their surface, and concurrently led to colloidally stable dispersions. The functionalization was achieved by applying a mechanically assisted stabilization treatment, where the molar ratio TiO2-to-ligand was the key parameter for an effective functionalization. Subsequently, a two-stage chemical solution deposition approach (dip-coating followed by immersion) was applied, which provides optimal control over the coating process as well as on the properties of the resulting thin films. Moreover, only low temperatures are required for the drying of the coatings, providing high potential for the application on polymer and thin foils as flexible substrates. The final coatings are highly transparent, homogenous, hard and smooth. The formation of a covalent network by the cross-linking of anatase nanoparticles was proven via a combination

17 ACS Paragon Plus Environment

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

of analytical techniques and showed to be beneficial for all the above properties of the resulting coatings. Utilization of the optimum TiO2: ligand: linker ratio turned out to be paramount for obtaining coatings with optimum hardness. The photocatalytic properties of the TiO2 coatings can find various applications (self-cleaning, air purifying) mainly outdoors as UV radiation is required for them to be activated; consequently, the improvement of their mechanical properties is highly beneficial in order to avoid attrition, maintain their structural integrity and provide advanced (e.g. anti-scratch) properties.

Acknowledgements This work was funded by the Deutsche Forschungsgemeinschaft, grants GA 1492/7-2 and KW 9/16-2. The authors gratefully acknowledge Mr. Peter Pfeiffer (Institute of Material Science, TU Braunschweig) and Dr. Frank Ludwig (Institute for Electrical Measurement Technology, TU Braunschweig) for the SEM images, Ms. Stephanie Michel for the nanoindentation and AFM measurements, and Mr. Wanja Dziony for the XPS measurements. Furthermore, we thank Prof. Carsten Schilde for numerous helpful discussions.

Supporting Information Stabilization efficiency, solids content and secondary particle size of the obtained NP dispersions; weight loss of functionalized and cross-linked NPs as determined by TGA; hardness and roughness data of the coatings; FTIR data of the pure solvent and ligands; SEM image of coatings at larger magnification; experimental details and results of photocatalytic activity measurements.

18 ACS Paragon Plus Environment

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

References

(1)

Okada, M.; Yamada, Y.; Jin, P.; Tazawa, M.; Yoshimura, K. Fabrication of Multifunctional Coating Which Combines Low-E Property and Visible-Light-Responsive Photocatalytic Activity. Thin Solid Films 2003, 442 (1–2), 217–221.

(2)

Faustini, M.; Nicole, L.; Boissière, C.; Innocenzi, P.; Sanchez, C.; Grosso, D. Hydrophobic, Antireflective, Self-Cleaning, and Antifogging Sol-Gel Coatings: An Example of Multifunctional Nanostructured Materials for Photovoltaic Cells. Chem. Mater. 2010, 22 (15), 4406–4413.

(3)

Prado, R.; Beobide, G.; Marcaide, A.; Goikoetxea, J.; Aranzabe, A. Development of Multifunctional Sol-Gel Coatings: Anti-Reflection Coatings with Enhanced Self-Cleaning Capacity. Sol. Energy Mater. Sol. Cells 2010, 94 (6), 1081–1088.

(4)

Marolt, T.; Škapin, A. S.; Bernard, J.; Živec, P.; Gaberšček, M. Photocatalytic Activity of Anatase-Containing Facade Coatings. Surf. Coatings Technol. 2011, 206 (6), 1355–1361.

(5)

Weir, A.; Westerhoff, P.; Fabricius, L.; Hristovski, K.; Von Goetz, N. Titanium Dioxide Nanoparticles in Food and Personal Care Products. Environ. Sci. Technol. 2012, 46 (4), 2242– 2250.

(6)

Committee, S.; Sccs, C. S. Scientific Committee on Consumer Safety Additional Coatings for Titanium Dioxide ( Nano Form ) as UV-Filter in Dermally Applied Cosmetic Products; 2016.

(7)

Schilling, K.; Bradford, B.; Castelli, D.; Dufour, E.; Nash, J. F.; Pape, W.; Schulte, S.; Tooley, I.; van den Bosch, J.; Schellauf, F. Human Safety Review of “nano” Titanium Dioxide and Zinc Oxide. Photochem. Photobiol. Sci. 2010, 9 (4), 495.

(8)

Fujishima, A.; Zhang, X.; Tryk, D. A. TiO2 Photocatalysis and Related Surface Phenomena. Surf. Sci. Rep. 2008, 63 (12), 515–582.

(9)

Di Paola, A.; García-López, E.; Marcì, G.; Palmisano, L. A Survey of Photocatalytic Materials for Environmental Remediation. J. Hazard. Mater. 2012, 211–212, 3–29.

(10)

Diasanayake, M. A. K. L.; Senadeera, G. K. R.; Sarangika, H. N. M.; Ekanayake, P. M. P. C.; Thotawattage, C. A.; Divarathne, H. K. D. W. M. N. R.; Kumari, J. M. K. W. TiO2 as a Low Cost, Multi Functional Material. Mater. Today Proc. 2016, 3, S40–S47.

(11)

Graziani, L.; Quagliarini, E.; Osimani, A.; Aquilanti, L.; Clementi, F.; Yéprémian, C.; Lariccia, V.; Amoroso, S.; D’Orazio, M. Evaluation of Inhibitory Effect of TiO2 Nanocoatings against Microalgal Growth on Clay Brick Façades under Weak UV Exposure Conditions. Build. Environ. 2013, 64, 38–45.

(12)

Lu, Y.; Sathasivam, S.; Song, J.; Crick, C. R.; Carmalt, C. J.; Parkin, I. P. Robust Self-Cleaning Surfaces That Function When Exposed to Either Air or Oil. Science 2015, 347 (6226), 1132– 1135.

(13)

Watanabe, T.; Nakajima, A.; Wang, R.; Minabe, M.; Koizumi, S.; Fujishima, A.; Hashimoto, K. Photocatalytic Activity and Photoinduced Hydrophilicity of Titanium Dioxide Coated Glass. Thin Solid Films 1999, 351 (1–2), 260–263.

(14)

Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238 (5358), 37–38.

(15)

Nakata, K.; Fujishima, A. TiO2 Photocatalysis: Design and Applications. J. Photochem. Photobiol. C Photochem. Rev. 2012, 13 (3), 169–189.

(16)

Fujishima, A.; Rao, T. N.; Tryk, D. A. Titanium Dioxide Photocatalysis. J. Photochem. Photobiol. C Photochem. Rev. 2000, 1 (1), 1–21.

19 ACS Paragon Plus Environment

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

(17)

Hussain, M.; Ceccarelli, R.; Marchisio, D. L.; Fino, D.; Russo, N.; Geobaldo, F. Synthesis, Characterization, and Photocatalytic Application of Novel TiO2 Nanoparticles. Chem. Eng. J. 2010, 157 (1), 45–51.

(18)

Masakazu, A. Preparation, Characterization, and Reactivities of Highly Functional Titanium Oxide-Based Photocatalysts Able to Operate under UV–Visible Light Irradiation: Approaches in Realizing High Efficiency in the Use of Visible Light. Bull. Chem. Soc. Jpn. 2004, 77 (8), 1427–1442.

(19)

Choi, H.; Sofranko, A. C.; Dionysiou, D. D. Nanocrystalline TiO2 Photocatalytic Membranes with a Hierarchical Mesoporous Multilayer Structure: Synthesis, Characterization, and Multifunction. Adv. Funct. Mater. 2006, 16 (8), 1067–1074.

(20)

Van Gestel, T.; Vandecasteele, C.; Buekenhoudt, A.; Dotremont, C.; Luyten, J.; Leysen, R.; Van der Bruggen, B.; Maes, G. Alumina and Titania Multilayer Membranes for Nanofiltration: Preparation, Characterization and Chemical Stability. J. Memb. Sci. 2002, 207 (1), 73–89.

(21)

Abargues, R.; Albert, S.; Valdés, J. L.; Abderrafi, K.; Martínez-Pastor, J. P. MolecularMediated Assembly of Silver Nanoparticles with Controlled Interparticle Spacing and Chain Length. J. Mater. Chem. 2012, 22 (41), 22204.

(22)

Grote, C.; Chiad, K. J.; Vollmer, D.; Garnweitner, G. Unspecific Ligand Binding Yielding Stable Colloidal ITO-Nanoparticle Dispersions. Chem. Commun. 2012, 48 (10), 1464–1466.

(23)

Grote, C.; Cheema, T. A.; Garnweitner, G. Comparative Study of Ligand Binding during the Postsynthetic Stabilization of Metal Oxide Nanoparticles. Langmuir 2012, 28 (40), 14395– 14404.

(24)

Kockmann, A.; Hesselbach, J.; Zellmer, S.; Kwade, A.; Garnweitner, G. Facile Surface Tailoring of Metal Oxide Nanoparticles via a Two-Step Modification Approach. RSC Adv. 2015, 5 (75), 60993–60999.

(25)

Zimmermann, M.; Temel, B.; Garnweitner, G. Parameter Studies of the Synthesis of Titanium Dioxide Nanoparticles: Effect on Particle Formation and Size. Chem. Eng. Process. 2013, 74.

(26)

Zimmermann, M.; Garnweitner, G. Spontaneous Water Release Inducing Nucleation during the Nonaqueous Synthesis of TiO2 Nanoparticles. CrystEngComm 2012, 14 (24), 8562.

(27)

Barth, N.; Zimmermann, M.; Becker, A. E.; Graumann, T.; Garnweitner, G.; Kwade, A. Influence of TiO2 Nanoparticle Synthesis on the Properties of Thin Coatings. Thin Solid Films 2015, 574, 20–27.

(28)

Barth, N.; Schilde, C.; Kwade, A. Influence of Particle Size Distribution on Micromechanical Properties of Thin Nanoparticulate Coatings. Phys. Procedia 2013, 40, 9–18.

(29)

Zimmermann, M.; Ibrom, K.; Jones, P. G.; Garnweitner, G. Formation of a Dimeric Precursor Intermediate during the Nonaqueous Synthesis of Titanium Dioxide Nanocrystals. ChemNanoMat 2016, 2 (12), 1073–1076.

(30)

De Roo, J.; Justo, Y.; De Keukeleere, K.; Van den Broeck, F.; Martins, J. C.; Van Driessche, I.; Hens, Z. Carboxylic-Acid-Passivated Metal Oxide Nanocrystals: Ligand Exchange Characteristics of a New Binding Motive. Angew. Chemie-International Ed. 2015, 54, 1–5.

(31)

De Roo, J.; Van Den Broeck, F.; De Keukeleere, K.; Martins, J. C.; Van Driessche, I.; Hens, Z. Unravelling the Surface Chemistry of Metal Oxide Nanocrystals, the Role of Acids and Bases. J. Am. Chem. Soc. 2014, 136 (27), 9650–9657.

(32)

Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. Spectrometric Identification of Organic Compounds, 4th ed.; Wiley: New York, 1981.

(33)

Barber, M.; Connor, J. A.; Guest, M. F.; Hillier, I. H.; Schwarz, M.; Stacey, M. Bonding in Some Donor-Acceptor Complexes Involving Boron Trifluoride. Study by Means of ESCA and

20 ACS Paragon Plus Environment

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

Molecular Orbital Calculations. J. Chem. Soc., Faraday Trans. 2 1973, 69 (0), 551–558. (34)

Lindberg, B.; Maripuu, R.; Siegbahn, K.; Larsson, R.; Gölander, C. G.; Eriksson, J. C. ESCA Studies of Heparinized and Related Surfaces. 1. Model Surfaces on Steel Substrates. J. Colloid Interface Sci. 1983, 95 (2), 308–321.

(35)

Lindberg, B. J.; Hedman, J. Molecular Spectroscopy by Means of Esca. 6. Group Shifts for N, P and as Compounds. Chem. Scr. 1975, 7 (4), 155–166.

(36)

Shenglong, W.; Fosong, W.; Xiaohui, G. Polymerization of Substituted Aniline and Characterization of the Polymers Obtained. Synth. Met. 1986, 16 (1), 99–104.

(37)

Hesselbach, J.; Kockmann, A.; Lüke, S.; Overbeck, A.; Garnweitner, G.; Schilde, C.; Kwade, A. Enhancement of the Mechanical Properties of Nanoparticulate Thin Coatings via Surface Modification and Cross-Linking Additive. Chem. Eng. Technol. 2017, 40 (9), 1561–1568.

21 ACS Paragon Plus Environment

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

Table of Contents Graphic

22 ACS Paragon Plus Environment