Exceptional Sensitizer Dye Loading via a New Porous Titanium

Subscriber access provided by University of Sunderland

Article

Exceptional Sensitizer Dye Loading via a New Porous Titanium-Niobium Metal Oxide with Tris(2,2'-bipyridyl)ruthenium(II) in the Structure Tahmina Akter, and Geoffrey B. Saupe ACS Appl. Nano Mater., Just Accepted Manuscript • DOI: 10.1021/acsanm.8b01242 • Publication Date (Web): 04 Sep 2018 Downloaded from http://pubs.acs.org on October 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 42 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

ACS Applied Nano Materials

1

Exceptional Sensitizer Dye Loading via a New Porous Titanium-Niobium

2

Metal Oxide with Tris(2,2'-bipyridyl)ruthenium(II) in the Structure

3

Tahmina Akter and Geoffrey B. Saupe*

4

Department of Chemistry, University of Texas at El Paso

5

500 W. University Avenue, El Paso, TX 79968-0513, USA

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

* Corresponding Author :

21

email: [email protected] (Dr. Geoffrey B. Saupe)

22

1 ACS Paragon Plus Environment

ACS Applied Nano Materials 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

1

Page 2 of 42

Abstract:

2

Exceptionally high loadings of tris(2,2'-bipyridyl)ruthenium(II), abreviated as Ru(bpy)32+,

3

were successfully incorporated into a titanium niobium mixed oxide framework. Ru(bpy)32+

4

loadings of 295.6 mg/g were achieved, equivalent to 518.9 µmol/g. Through an aqueous-phase

5

mixing process, a solution of Ru(bpy)32+ was used to aggregate colloids of unilamellar anionic

6

nanosheets, derived from the exfoliation of the layered semiconductor KTiNbO5. The aggregates

7

formed were porous metal oxide composite materials, with Ru(bpy)32+ electrostatically bound

8

within the structure of the pore walls. The new solids were highly orange-colored, due to the

9

high internal content of Ru(bpy)32+. X-ray diffraction, chemical and theoretical analyses indicate

10

that the ruthenium dye is incorporated as a single layer between the metal oxide sheets, and that

11

the dye loading is maximized. In well rinsed samples, the Ru(bpy)32+ remained in the solid with

12

only trace amounts of leaching, even after 3 months in water at room temperature. The strategy

13

demonstrated here is general enough to be applicable to other layered metal oxides. This report

14

describes the first successful use of Ru(bpy)32+ and restacked KTiNbO5 to synthesize a stable,

15

dye-sensitized, porous metal oxide semiconductor material.

16

correlate with light capture efficiencies, giving this new technique promise in solar energy

17

conversion technologies.

Improved sensitizer loadings

18 19

Keywords: Sensitizer; Ru(bpy)32+; Porous metal oxide; Restacked; Exfoliated nanosheets;

20

KTiNbO5; Dye loading; Layered semiconductor.

21


Page 3 of 42 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

1


1. Introduction

2

Visible light dye sensitization of wide band gap metal oxide semiconductors has been

3

widely studied as an approach to capture terrestrial solar energy. The strategy is used for direct

4

water photolysis for hydrogen generation, dye-sensitized solar cells, and advanced photo

5

oxidation processes for water decontamination.1-6 Due to its long-lived photoexcited state and its

6

high molar absorptivity in the visible spectrum, tris(2,2'-bipyridyl)ruthenium(II), abreviated as

7

Ru(bpy)32+, has been studied extensively as a visible light sensistizer for heterogeneous metal

8

oxide wide band gap semiconductor-based photocatalysis.1,2,7 In general, long-lived photoexcited

9

states of sensitizer dyes are desirable, because they enhance the probability of electron transfer

10

from an excited state of a sensitizer molecule into the conduction band of a metal oxide

11

semiconductor, thus improving charge separation lifetimes and promoting subsequent electron

12

transfer reactions. Because light harvesting efficiencies correlate with the absorbance of a

13

photocatalytic material, creating materials with higher sensitizer loadings has the potential to

14

increase light absorption and solar energy conversion efficiencies. Ru(bpy)32+ and its analogs

15

possess electronic excited state potentials that are more negative than the conduction band

16

potentials of many metal oxide semiconductors, which is an energetic requirement for efficient

17

sensitization.8

18

Given suitable electronic potentials, between a photo excited sensitizer (donor) and the

19

conduction band of the semiconductor (acceptor), electron transfer will not occur unless the

20

sensitizer is adjacent to or in contact with the surface of the semiconductor. When this is not the

21

case, the sensitizer dye in solution must first diffuse to the semiconductor surfaces, where it is

22

then possible for electron transfer to occur. Despite long-lived photo excited states (hundreds of

23

nanoseconds) for Ru(bpy)32+, by comparison the rate of molecular diffusion in solution is very 3 ACS Paragon Plus Environment


Page 4 of 42

1

slow.1,9,10 Consequently, for non-tethered sensistizers, only the sensitizer molecules already

2

present at the semiconductor surface are useful donors when irradiating with light, and the rest of

3

the sensitizer in solution only acts as an attenuator of the incoming light source, thus reducing

4

the system efficiency. Solution phase sensitizers have therefore resulted in relatively poor

5

sensitization performance.1,10

6

To improve the electron transfer efficiencies, from sensitizers to the conduction band of

7

metal oxides, various attachment or anchoring schemes have been devised for tethering sensitizer

8

dyes to surfaces. For example, the bipyridine ligands of standard Ru(bpy)32+ have been modified

9

in ways that enable tethering of the modified Ru(bpy)32+. Dye compounds with carboxylate,

10

phosphonate, phosphonite, phosphoramidate and other linkers on the ligands of metal complexes

11

have been explored.5,6,10-14 Modifications of ruthenium based dyes have also been investigated in

12

pursuit of extending their visible light absorption spectra, therefore capturing more of the solar

13

spectrum and improving the useful optical density.15 Compared to regular Ru(bpy)32+, specialty

14

ligands with tethers may increase the complexity and cost of the sensitizer dyes, and may also

15

introduce unwanted photostability issues.10 Studies have reported the deactivation of sensitizer

16

dyes due to tether failure and desorption from the semiconductor surface, or from de-

17

intercalation and chemical or photo decomposition.1,15-17 Sensitizer failure appears to be the

18

primary cause of catalyst deactivation. The stability of sensitizer dyes is a key issue to be

19

solved, if direct water photolysis is to become a viable solar energy conversion technology.10,18

20

Although there are advantages to the stability and simplicity of Ru(bpy)32+, a major

21

disadvatage is that it does not bind well, if at all, to the external surfaces of most metal oxide

22

semiconductors. Other than by electrostatic attraction, Ru(bpy)32+ has no capability to tether

23

itself onto a surface. One option is to insert or intercalate Ru(bpy)32+ into the inner spaces of


Page 5 of 42 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


1

layered metal oxides via ion exchange, thus taking advantange of the high internal surface area

2

of such materials to improve loadings, and to possibly increase the chemical stability of

3

Ru(bpy)32+.

4

However, previous studies have shown that ion exchange of Ru(bpy)32+ into layered

5

semiconducors like KTiNbO5 or K4Nb6O17 does not occur readily, and may take several weeks19

6

without pre-expanding layers with precursors like bulky amines. Even so, the resulting dye

7

loadings into such oxides have been sub-optimal.19,20 Porphyrines and other dyes have similar

8

problems.21-24

9

photoinstability issues lead to catalyst deactivation and prevent optimal catalyst performance and

10

Low sensitizer loadings, sensitizer desorption, sensitizer decomposition and

longevity.

11

From our own group and from others, there have been a handful of reports about novel

12

materials that were synthesized via the restacking of colloidal nanosheets, which were derived

13

from exfoliated layered materials.25-34 Of these restacking reports, even fewer have focused on

14

the goal of creating visible light photocatalysts. However, after first creating restacked materials,

15

three of these studies subsequently tried to intercalate or adsorb molecular visible light

16

sensitizers,35-37 none of which were Ru(bpy)32+.

17

intercalation studies was aimed at heterogeneous photocatalysis. However, in 2003 Unal et al.

18

reported a restacked cesium titanate, and in 2006 a restacked K4Nb6O17, where Ru(bpy)32+ was

19

used to induce the restacking of a colloid.25,29 In the 2003 study by Unal, the authors used the

20

restacked material to create a thin film photoelectrode. In the 2006 study, a photoelectrode was

21

prepared, but the authors also tested the photocatalytic properties of their material, however the

22

stability of the composite was very poor and they claim that the Ru(bpy)32+ deintercalated

23

readily, causing rapid photoelectrode deactivation in less than an hour.29 There still remains a

Though using dyes, only one of these



Page 6 of 42

1

great need to discover robust, visible light sensitized, photocatalytic materials with exceptional

2

photon capture capabilities for use in direct water photolysis and for dye-sensitized solar cell

3

applications.

4

In this report we show that Ru(bpy)32+ can be immobilized quite effectively into a stable

5

structure of restacked metal oxide nanosheets, resulting in a new material with an exceptional

6

internal loading of the Ru(bpy)32+ sensitizer. The colloidal nanosheets were derived from the

7

exfoliation of the layered metal oxide semiconductor, KTiNbO5, which has a well known crystal

8

structure.38

9

Ru(bpy)32+ molecules, creating a disordered agglomerate, held together only by electrostatic

10

forces. The great extent of the Ru(bpy)32+ sensitizer loading gave the composite a deep orange

11

color. The new synthetic method is general enough to be used with other layered materials and

12

other cationic sensitizers.

13

Ru(bpy)32+ firmly held and evenly spread throughout the composite material. This is the first

14

successful use of Ru(bpy)32+ to create stable restacked layered materials. The new materials also

15

represent the first use of restacked KTiNbO5 to synthesize a visible light sensitized, porous metal

16

oxide semiconductor material.

When mixed, the anionic colloidal nanosheets restacked around the cationic

This work has resulted a robust composite, with high levels of

17 18

2. Experimental Methods

19

All water used was ultra-pure and deionized using a standard ion exchange resin system

20

until a conductivity of 18 MΩ was reached. All chemicals were used as received from the

21

manufacturer, as indicated.

22

2.1. Synthesis of KTiNbO5


Page 7 of 42 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


1

Potassium titanium niobium pentoxide (KTiNbO5) was synthesized by a conventional

2

solid-state synthetic method.39 Reagent grade potassium carbonate (99.9%, Fisher Chemicals),

3

titanium dioxide (99.5%, Sigma-Aldrich) and niobium (V) oxide (99.9%, Alfa Aesar) were

4

purchased and used as received. KTiNbO5 was prepared by grinding stoichiometric amounts of

5

K2CO3, TiO2 and Nb2O5 into fine powders with a mortar and pestle and were then mixed

6

thoroughly. A 10% excess of K2CO3 was used to counteract the loss of potassium as an oxide

7

vapor during the heating cycle. The mixture was heated in an alumina crucible in a high

8

temperature furnace (BF51800 series, Lindberg/Blue, USA) for 20 hours using the following

9

program. The temperature was raised from 20 to 120 °C at 10 °C per minute and held at 120 ºC

10

for 60 minutes. The temperature was then raised to 900° C at a rate of 10 °C per minute and held

11

at 900 °C for 60 minutes. Finally, the temperature was raised to 1100 °C at the same rate and

12

held for 20 hours and was then allowed to cool to 20 °C. The product material, KTiNbO5, was

13

stored at ambient conditions until needed.

14

2.2. Acid exchange and colloid preparation

15

HTiNbO5 was produced from the parent KTiNbO5 by a water solution-based acid

16

exchange process using a solid to liquid ratio of 5 g/300 mL. The KTiNbO5 powder was stirred

17

in a 4 M aqueous HCl solution for 24 hours for 24 hours. The solid was separated from the

18

liquid using centrifugation. The solid was re-suspended and again stirred in 4 M HCl. The

19

process of stirring with acid was done a total of four times. Finally, the solid was washed with

20

fresh water before allowing the material to air dry in a glass dish. The resulting material was

21

HTiNbO5.

22

Aqueous tetrabutylammonium hydroxide (TBAOH 40%, GFS Chemicals, Columbus,

23

Ohio, USA) was added dropwise to a stirred suspension of powdered HTiNbO5 to induce 7 ACS Paragon Plus Environment


Page 8 of 42

1

exfoliation, until the pH stabilized at 9.26 At this pH, the TBA+ cations fully exfoliate the

2

individual metal oxide sheets into a colloid. The resulting mixture was stirred for a day with

3

TBAOH added as needed to maintain the pH at 9. The pH was monitored using phenolphthalein

4

as an indicator. The indicator dye helps avoid problems with leaking potassium cations from pH

5

meter probes, as frequent pH determinations are necessary. The colloid container was capped to

6

avoid atmospheric CO2 contamination.

7

2.3. Precipitation of colloids to create Ru(bpy)32+ sensitized porous oxides

8

Based on the electrostatics ratio in KTiNbO5, the desired catalyst material

9

[Ru(bpy)3](TiNbO5)2 was prepared by restacking or agglomerating the colloidal sheets. Double

10

the stoichiometric amount of tris(2,2'-bipyridyl)ruthenium(II) dichloride (GFS Chemicals,

11

Columbus, Ohio, USA) was dissolved in water. The concentration of the resulting Ru(bpy)3Cl2

12

dye solution was about 0.07 M. Eight milliliters of the colloid of (TBA)TiNbO5 (conc. 0.015

13

g/mL) was added drop wise into the slowly stirred dye solution. An orange solid precipitated.

14

After the reaction, the vial was stored in the dark overnight. The precipitated mixture was then

15

centrifuged. The supernatant liquid was removed and pure water was added. The centrifuge and

16

rinsing steps were repeated 5 or more times until the supernatant water remained clear of any

17

dye. The new material is referred to as RuPOX, meaning ruthenium dye-loaded porous oxide.

18

Scheme 1 shows an overview of the entire synthetic process.

19

2.4. HTiNbO5 and HPOX dye loading and general dye content analyses

20

The Ru(bpy)32+ loaded HTiNbO5 was made by stirring 15 mg of HTiNbO5 powder in 5

21

mL of an aqueous 9 mM Ru(bpy)32+ solution for 24 hours. The solid was then rinsed well with

22

pure water. Acid restacked colloids of (TiNbO5¯ )n to make the acid restacked porous oxide , or


Page 9 of 42 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


1

HPOX, were made by adding drops of a 2 M HCl into the colloids to produce a solid porous

2

agglomerate. The solids were then rinsed well with pure water. HPOX was then soaked in 0.07

3

M solution of Ru(bpy)32+ for 24 hours.

4

The extent of loading of Ru(bpy)32+ in the RuPOX materials was measured by analyzing

5

for elemental ruthenium in the RuPOX solid, using inductively coupled plasma optical emission

6

spectroscopy (ICP-OES) (Perkin Elmer, Optima 4300 DV, Norwalk, CT, USA). In preparation

7

for ICP analyses, 1.000 mL aliquots of wet RuPOX were air dried. After drying, the samples

8

were then dissolved in 2-3 drops of 50% aqueous HF (Alfa Aesar, USA) and then the solution

9

was diluted with 1% nitric acid solution.

Quantifications of the Ru(bpy)32+ loadings on

10

HTiNbO5 and on HPOX were done by measuring the optical absorbance of Ru(bpy)32+ in

11

solution, after dissolving the solids in 50% aqueous HF, followed by dilution with water. This

12

procedure yielded equivalent results to using ICP-OES, as described above.

13

2.5. Other characterizations

14

X-ray powder diffraction (XRD) patterns of KTiNbO5, HTiNbO5, and the dye-sensitized

15

catalyst RuPOX were obtained to verify structure (Cu Kα radiation, Siemens D5000, Germany).

16

UV–Vis diffuse reflectance spectra (UV-3101 PC, Shimadzu, Japan) of the dried RuPOX

17

material were measured. Optical absorption spectra of dye solutions were obtained using Cary

18

300, Cary 50, or Cary 60 UV-VIS spectrophotometers (Varian, USA).

19

microscopy (SEM, S-4800, Hitachi, Japan) was used to inspect the morphology of the materials.

20

For the SEM analysis, samples were air dried and placed on carbon tape for the analysis.

21

Transmission electron microscopy (TEM) was performed with a Zeiss EM-10 microscope

22

(Oberkochen, Germany).

23

epifluorescence microscope (Zeiss, Germany). Fourier transform infrared spectra were acquired

Scanning electron

Epifluorescence spectra were obtained using Zeiss Axioskop



Page 10 of 42

1

using a Spectrum-100 (Perkin Elmer Inc., USA).

Thermogravimetric analyses (TGA/DSC

2

Metler Toledo, Switzerland) were performed in air using a heating rate of 10 °C/min over a

3

temperature range of 25-950 °C.

4

Scheme 1. Overview of the synthesis of the new dye-sensitized porous oxide, RuPOX. 5

6 7 8 9 10 11


Page 11 of 42 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


1 2 3

3. Results and Discussion With the synthetic strategies presented here, high levels of Ru(bpy)32+ have been

4

incorporated

inside

of

restacked

KTiNbO5

5

[Ru(bpy)3]xH2-2x(TiNbO5)2, where ‘bpy’ is bipyridine. The strategy uses only electrostatic forces

6

to keep the dye and the semiconductor stuck together in a composite, however, this has resulted

7

in a robust binding of the dye to the semiconductor. Although the new method is not a standard

8

intercalation reaction, the process creates an intercalated layered semiconductor material.

9

Ru(bpy)32+ is fully intercalated and electrostatically bound inside the new material’s pore walls,

10

which are composed of restacked layered nanosheets of metal oxide, derived from the exfoliation

11

of parent compound, KTiNbO5.

to

produce

the

novel

nanomaterial

12

To appreciate the structure of the new material, it is important to understand the building

13

blocks used in the synthesis. The parent material, KTiNbO5, is an ion-exchangeable layered

14

metal oxide semiconductor material.39 The crystal structure indicates that there are layers of

15

potassium ions in between layers of oxygen-coordinated titanium and niobium octahedra.38

16

Aqueous colloids of anionic nanosheets were produced by chemically exfoliating KTiNbO5,

17

which involves cation exchange with acid and a subsequent reaction with tetrabutylammonium

18

hydroxide (see Scheme 1).40

19

dimensional metal oxide nanosheets, which can be represented as (TiNbO5¯ )n. The nanosheets

20

are anionic, about one nanometer thick, one to several microns in diameter, and they are quite

21

stable in water.30,41

22

(tetrabutylammonium) as the counter ion.30 The structure of each nanosheet stems directly from

23

the crystalline layers of the parent metal oxide material, KTiNbO5. Thus, the sheets are single

The colloidal nanostructures produced in this way are two-

At pH 9, the anionic colloidal nanosheets mostly have TBA+



Page 12 of 42

1

crystal nanosheets with the photophysical properties of a wide band gap semiconductor

2

material.30,42

3

To create highly porous solids, our laboratory has been agglomerating colloidal metal

4

oxide sheets with smaller cations, like H+ from acids, or metal cations.33,43 Other reports have

5

investigated similar processes, however few have focused on the porosity of the agglomerated

6

materials or their use as photocatalysts.25-34,44 In one report, researchers have intentionally

7

restacked exfoliated materials slowly, thus recreating the dense structure resembling the

8

original.44

9

disordered solid with pores.33,43 In general, cations with significantly higher charge densities

10

However, the fast agglomeration of colloidal nanosheets can lead to a more

than TBA+ will cause colloids of metal oxide sheets to agglomerate, or precipitate.

11

In this report, tris(2,2'-bipyridyl)ruthenium(II), abreviated as Ru(bpy)32+, was used to

12

precipitate colliodal metal oxide sheets quickly into porous agglomerates. It is perhaps more

13

accurate to say that the anionic colloidal particle sheets agglomerate or flocculate with the

14

application of the cation, as it is not a true precipitation reaction. The technique of restacking

15

exfoliated materials is sometimes refered to in the literature as electrostatic self-assembly

16

deposition.29 Though not equivalent, that term can be confused with electrostatic deposition and

17

electrostatic self-assembly of layers via layer-by-layer deposition methodologies. The term

18

restacking will be used in this report. The addition of an aqueous solution of Ru(bpy)32+ to the

19

anionic colloidal sheets destabilizes the colloid, causing the negatively charged sheets to flock

20

together with the cation, resulting in the Ru(bpy)32+ molecules being sandwiched or intercalated

21

between the anionic sheets (see Scheme 2) As can be seen in Fig. 1, highly colored porous solids

22

were produced.

23

concentration of dye and a high absorbance. The new material, [Ru(bpy)3]xH2-2x(TiNbO5)2, is

The intensity of the orange color is strikingly rich, suggesting a high


Page 13 of 42 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


Scheme 2. Anionic colloidal nanosheets will restack using Ru(bpy)32+ as the cation. Though stacked, the nanosheets are agglomerated randomly, producing voids.

1

referred to as RuPOX, meaning ruthenium dye-loaded porous oxide.

The new dye-filled

2

material is a disordered solid with a high degree of porosity. In addition to the characterizations

3

provided here, we draw this conclusion in part from evidence found in the literature. As

4

mentioned earlier in the introduction section, in addition to work from our lab, there have been

5

literature reports of restacked colloids, which have produced porous aggregrates with high

6

surface areas and a turbostratic structure.26,32,43 For example, the Mallouk group exfoliated and

7

restacked K4Nb6O17 to produce porous materials with surface areas of 250-300 m2/g.45

8

The porosity in the RuPOX results from the random positions that the individual

9

exfoliated sheets adopt, during the sudden precipitation of the colloid by Ru(bpy)32+. The sheets

10

are folded, wrinkled, rolled, and randomly arranged, which creates voids, or pores.30,43 As

11

demonstrated below, this new synthetic procedure appears to maximize the loading of dye. The

12

wet porous oxide is saturated with the ruthenium dye and has the characteristics of a fluffy

13

orange-colored solid. Thermogravimetric analysis (TGA) of dried RuPOX and HTiNBO5 were

14

performed. The results of the TGA are presented in the Supporting Information (see Fig. S4). In

15

addition, comparable infrared spectra of the Ru(bpy)3Cl2 and the RuPOX powders are available

16

in the Supporting Information (see Fig. S3). 13 ACS Paragon Plus Environment


Page 14 of 42

1 2

3.1. Dye content analysis

3

To quantify the amount of dye that was incorporated into the porous oxide, elemental

4

analyses were done using ICP-OES. To ensure only the bound Ru(bpy)32+ dye in the solid was

5

measured, the measurements were done after extensive rinsing of the RuPOX materials in pure

6

water. The average ruthenium content was found to be 295.6 mg/g ±5%, which is equivalent to

7

518.9 µmol/g. The extent of the dye loading was rationalized in the following way. (For

8

calculation details see Supporting Information, S1.) Using an electrostatics-based stoichiometry

9

from an ideal formula of [Ru(bpy)3]x(TiNbO5)2, where x =1, the theoretical maximum amount of

10

dye that can be combined into the ionic solid was calculated to be 563.36 mg/g, which is

Figure 1.

An aqueous sample of the new RuPOX material, showing the intensity of the

orange color from the high loading of Ru(bpy)32+. The supernatant contains suspended particles of the orange solid and is not colored by free dye. 14 ACS Paragon Plus Environment

Page 15 of 42 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


1

equivalent to 988.93 µmol/g. Experimental results showed that very high dye loadings were

2

achieved, where x = 0.525 ±5 %. On an electrostatic basis, this means that 52.5 % of the

3

theoretical maximum ‘intercalated’ dye loading was attained in the synthesis.

4

reasonable approach, considering that the driving force for the precipitation is electrostatic.

5

Using conventional ion exchange, K.Yao et al. measured maximum intercalations and obtained

6

dye loadings of 11.5 % and 7 % of maximum using an electrostatics basis, while taking into

7

consideration both enantiomers of ruthenium trisphenanthroline.20 Meyer et al. reported the

8

extent of dye loading at 42 µmol/g for a porous oxide film on dye-sensitized solar cells.46 The

9

experimental results reported here (518.9 µmol/g) far exceed those numbers.

This is a

10

It is important to note that the charge densities of the cations at work are a major factor in

11

the limits to ion exchange capacity. The loading of the Ru(bpy)32+ dye inside the RuPOX

12

material may have reached a fundamental limit due to charge density and size limitations, where

13

no more Ru(bpy)32+ can possibly fit between sheets in a single layer, due to steric factors. This

14

possibility was analyzed by comparing the charge density of the original K+ in KTiNbO5 to that

15

of Ru(bpy)32+, which is being used to replace the K+ in the KTiNbO5 to produce RuPOX. A

16

similar analysis approach was used by Bujdak for a dye in clays.47 The charge density of each

17

(TiNbO5¯ )n anionic sheet does not change, whether in the parent solid, KTiNbO5, or in the final

18

product, RuPOX. Therefore, by comparing the charge densities of the two cations we can gain

19

insight into the limits of cation exchange using Ru(bpy)32+. The ratio of the unhydrated bare ion

20

charge densities (charge to volume ratios) for K+ versus Ru(bpy)32+ is 27.4 to 1, using a radius of

21

133 pm and 505 pm for K+ and Ru(bpy)32+, respectively.48,49 (For calculation details please see

22

Supporting Information S1.)



Page 16 of 42

1

However, the cations in these layered materials are sandwiched between anionic metal

2

oxide sheets in a two-dimensional layer. Therfore, it is more meaningful to consider the two-

3

dimensional surface area per charge ratio (square area per charge) of the cations versus the

4

anionic sheet.

5

nm2/charge.35 For K+ and Ru(bpy)32+ the 2-D charge densities are calculated to be 0.0556

6

nm2/charge and 0.401 nm2/charge, respectively (For details see Supporting Information S1). In

7

comparing the value of the sheet’s charge density to that of the Ru(bpy)32+, we find that the

8

Ru(bpy)32+ requires 3.26 times more area per charge than the sheet provides. By using the

9

inverse ratio (0.307), the sheet offers only 30.7 % of the charge density required by the

10

Ru(bpy)32+. These ratios oversimplify the analysis, but nonetheless, they demonstrate that the

11

sheer size of the Ru(bpy)32+ limits its ability to completely compensate the anionic charge of the

12

sheets. The ideal formula for the RuPOX material is [Ru(bpy)3]x(TiNbO5)2, where x = 1. The

13

area per charge calculation above implies that based on stoichiometry or electrostatics alone a

14

loading of Ru(bpy)32+ where x = 1 is not feasible. In fact these calculations imply that the

15

loading of Ru(bpy)32+ in the RuPOX solid should be much less than the ideal formula indicates,

16

where x = 0.307.

The surface area per charge ratio of a single (TiNbO5¯ )n sheet is 0.123

17

Our experimental data resulted in x = 0.525, which is greater than the theoretical x =

18

0.307, based on the area per charge analysis above. This is an interesting result and it is not clear

19

how the extra dye is contained in the RuPOX solid. In the syntheses, the RuPOX materials were

20

rinsed extensively before chemical analysis. As Ru(bpy)32+ is quite soluble in water and contains

21

no functional groups that might serve to immobilize it or dimerize it, it is unlikely that any

22

excess Ru(bpy)32+ exists on the surface of the RuPOX solid. In water the extent of surface

23

adsorption was found to be negligible in our tests with Ru(bpy)32+ on powdered KTiNbO5, 16 ACS Paragon Plus Environment

Page 17 of 42 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


1

because surface adsorption, if any, occurs only in equilibrium with the Ru(bpy)32+ in solution.

2

Therefore, thorough rinsing ensures no surface adsorption. It is possible that the neighboring

3

Ru(bpy)32+ molecules in between the oxide sheets could be forced closer to each other than their

4

theoretical diameter suggests. The charge density on the sheets may be high enough to induce a

5

denser packing or squeezing of the Ru(bpy)32+ molecules. Because Ru(bpy)32+ is not a solid

6

sphere, it may be possible for the Ru(bpy)32+ molecules to move closer to each other, if the

7

bipyridine ligands on neighboring Ru(bpy)32+ molecules could interleave or become distorted

8

from their ideal positions. There is also the possibility that because the RuPOX material has a

9

disordered porous structure, it is likely that randomly produced closed cell pores may exist in the

10

porous RuPOX solid, possibly trapping excess Ru(bpy)32+ during the synthesis, with chloride or

11

hydroxide as the counter ion. Both of these factors may be involved in producing the higher than

12

expected loadings.

13

From the chemical analysis of RuPOX, x = 0.525 in [Ru(bpy)3]x(TiNbO5)2, so the

14

remaining charge balance (1-x) is equal to 0.475, which represents a divalent cation capacity that

15

is being satisfied by other cations within the solid. If the other cations are monovalent, then the

16

0.475 number is doubled to 0.95.

17

combination of H+ and TBA+, as they are the only other cations present during the synthesis.

18

These possibilities lead to the formula H0.95-yTBAy[Ru(bpy)3]0.525(TiNbO5)2, where y is likely

19

very close to zero, due to TBA+ having a much lower charge density as compared to H+ and

20

Ru(bpy)32+. The presence of H+ is also most likely, because H+ can easily fit in between the

21

larger Ru(bpy)32+ to complete the charge balance. Sources of H+ abound, because the synthesis

22

is carried out in water. Furthermore, the pH of a well-rinsed, pure water suspension of RuPOX

23

material is 5.5, suggesting that a source of protons exists within the solid.

The remaining charge balance is likely supplied by a



Page 18 of 42

1

When soaking the highly rinsed RuPOX material in pure water for long periods, the

2

composite materials retained their strong orange color and the dye did not leach out into solution.

3

Even after storing the rinsed RuPOX materials in water solutions for 3 months, the Ru(bpy)32+

4

did not leach out into the clean water, and the supernatant solution remained uncolored. Only a

5

trace amount of Ru(bpy)32+ is detectable via UV-Vis spectrophotometry (see Supporting

6

Information, Fig. S2). This is additional evidence that the Ru(bpy)32+ molecules are located

7

between the metal oxide sheets and are not bound to external surfaces where they can release

8

into solution. The parent material, KTiNbO5, is ion exchangeable. Therefore, it is expected that

9

the restacked material would also be ion exchangeable. In the presence of higher charge density

10

cations, such as Na+ or H+, the Ru(bpy)32+ can be ion exchanged out of the solid. It was found

11

that the addition of the RuPOX material to water at pH 1 resulted in the release of free dye into

12

solution and can be seen with the naked eye. The presence of free dye was also characterized by

13

UV-Vis spectrophotometry (see Supporting Information Fig. S2). Na+ was also found to ion

14

exchange into RuPOX, forcing Ru(bpy)32+ out by using concentrated NaCl solutions (not

15

shown).

16

In addition to calculating dye loadings on an electrostatic charge stoichiometric basis, dye

17

loadings were also calculated using a surface area basis. Even though it is believed that all the

18

Ru(bpy)32+ dye molecules in the synthesized RuPOX material were intercalated and

19

electrostatically bound between the anionic sheets, the following calculation assumes that all the

20

dye is only bound to the external surface. This theoretical analysis provides additional insight on

21

the new material’s high capacity for a sensitizer dye and facilitates the comparison of the dye

22

loading numbers with other semiconductor surface-only adsorbed dye results typically found in

23

the literature.14,46,52,58 The surface area footprint of the dye and the total surface area of the


Page 19 of 42 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


1

porous solid were used. As the literature reports, the diameter of Ru(bpy)32+ is 1.01 nm,49 the

2

surface area footprint of each Ru(bpy)32+ dye molecule was estimated to be 1.02 nm2. A square

3

footprint was used, to include the unoccupied and usable spaces between the round molecular

4

footprints as they close pack on the surface. Assuming that the porous material’s available

5

surface area can be completely covered by a monolayer of dye molecules, the maximum number

6

of molecules that can be surface-adsorbed is equal to the total available surface area of the

7

material divided by the square footprint area of one dye molecule. For this calculation, we have

8

used a surface area of 150 m2/g. The surface area is based on several nitrogen adsorption BET

9

surface area analyses, which were obtained on similar porous samples, however they were

10

precipitated with different cations and dried using supercritical point CO2 techniques prior to the

11

BET measurements.43 Because the RuPOX sample in this study was never dried, the surface

12

area of 150 m2/g is a best estimate. With these assumptions, the theoretical maximum coverage

13

of adsorbed Ru(bpy)32+ per gram of porous oxide is calculated to be 244.2 µmol/g.

14

Using the same ICP-OES elemental analysis data for the Ru(bpy)32+ content of RuPOX

15

(518.9 µmol/g), the efficiency of the surface only dye coverage for the RuPOX samples was

16

found (518.9 µmol/g /244.2 µmol/g) to be 212.5 % of the theoretical maximum.

17

calculations assume that the adsorbed dye molecules would only be attached onto the available

18

external surface area and are not intercalated between the sheets. It is reasonable that this result

19

(518.9 µmol/g) is more than twice the amount of dye of the theoretical maximum, because the

20

dye molecules in the RuPOX material are actually located between the sheets, not adsorbed on

21

the surface. The area between the layered overlapping sheets is greater than the exposed surface

22

area. Hence, this new synthetic strategy greatly enhances visible light sensitization of metal

23

oxide semiconductors and more than doubles the theoretical maximum sensitizer loading of

These



Page 20 of 42

1

conventional surface sensitization schemes even in this new high surface area porous material.

2

Furthermore, because each sensitizer molecule directly touches the metal oxide, these enhanced

3

loadings on the RuPOX materials are possibly achieved without the drawbacks of the inner filter

4

effects caused by multilayer/overlayer sensitizer surface depositions, which occur with

5

carboxylated linker and other type sensitizers. Inner filter effects greatly diminish effective

6

electron transfer and quantum efficiency, and hence, catalyst performance.

7

The surface area only calculation above shows that the actual dye loading is quite

8

excellent. By mass, the dye loading on the new RuPOX material is 29.6 %. For comparison,

9

Rawling et al. reported a 1.11 % loading by mass of the N719 sensitizer and 37.3 µmol/g on

10

TiO2 nanoparticle electrode substrate.51 Our result of 518.9 µmol/g on the new porous RuPOX

11

material is 13.9 times higher. It must be noted that to our knowledge and from the lack of

12

evidence in the literature, Ru(bpy)32+ does not dimerize or form clusters. Therefore, within the

13

metal oxide framework of RuPOX, clusters or solids of Ru(bpy)32+ are not likely to exist. This

14

also supports the conclusion that single layers of Ru(bpy)32+ are present between the metal oxide

15

sheets.

16

To show the relative effectiveness of this new sensitizer restacked material method versus

17

other synthetic methods of sensitization using Ru(bpy)32+, a direct ion exchange of HTiNbO5

18

with Ru(bpy)32+ was performed. This was done by soaking the HTiNbO5 in a solution of

19

Ru(bpy)32+ for 24 hours. The loading of Ru(bpy)32+ into HTiNbO5 was found to be 4.41 µmol/g.

20

The restacking method described herein using colloids restacked by Ru(bpy)32+ delivered a value

21

of 518.9 µmol/g, which compared to the direct ion exchange result of 4.41 µmol/g represents a

22

value that is 117 times greater. Others have also reported similar difficulties with the direct

23

intercalation of these types of materials with bulky dyes like Ru(bpy)32+.19,25 20 ACS Paragon Plus Environment

Page 21 of 42 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


1

Another porous material was made by first restacking the (TiNbO5¯ )n nanosheets with

2

acid (H+), thus forming the acid restacked porous metal oxide, or HPOX. As reported in

3

literature, this process also creates a highly disordered random porous structure with a high

4

surface area, which is comparable with the new RuPOX material’s structure (also discussed in

5

section 3.0).43 Like RuPOX, HPOX is also ion exchangeable. Due its porous, higher surface

6

area structure, the HPOX could provide enhanced mass transport throughout. By soaking the

7

HPOX in solutions of Ru(bpy)32+ for 24 hours, the HPOX was tested for its ability for the direct

8

ion exchange of Ru(bpy)32+. The dye loading found for the HPOX synthetic route was 84.5

9

µmol/g. The loading of the Ru(bpy)32+ dye in the RuPOX material is about 6 times greater than

10

the direct loading using the HPOX solid ion exchange route. Table 1 summarizes the Ru(bpy)32+

11

loading results for the various materials. The paramenters used during the synthesis of the

12

RuPOX material in this work yielded a maximized loading. This technique also permits the

13

flexibility of making fractionally dye loaded RuPOX material.

Table 1. Comparison of dye loadings using the restacking method and other methods. The restacking of colloids with Ru(bpy)32+ provided a superior result. Ru(bpy)32+ loading

Material

(µmol/g) 1

Ru(bpy)32+ -HTiNbO5

4.4

2

Ru(bpy)32+ -HPOX

84.5

3

RuPOX

518.9

4

Ru(bpy)32+ -P25 titaniaref. 10

2.5

14 15 21 ACS Paragon Plus Environment


1

Page 22 of 42

3.2. UV-Vis Diffuse Reflectance Spectrophotometry and X-ray Diffraction

2

Figure 2 shows the absorption spectra of both an aqueous solution of Ru(bpy)3Cl2 and the

3

air-dried RuPOX solid, acquired using transmission UV-Vis and UV-Vis diffuse reflectance

4

(UV-Vis-DRS) absorption spectrophotometry, respectively.

5

Ru(bpy)32+ in aqueous solution is at 452 nm.52 The spectrum of the Ru(bpy)32+ trapped in the

6

solid differs from the solution spectrum. The absorption spectrum of the RuPOX indicates that

7

the MLCT absorption band of Ru(bpy)32+ has broadened and the peak is red shifted by about 12

8

nm. Others have reported similar phenomena for intercalated Ru(bpy)32+ in solids20,53 and also

9

with other dyes.29 Peak broadening can originate from dye-dye interactions and from dye-host

10

interactions, as well as from the distortion of the intercalated molecules.53 For comparison, a

The MLCT absorption band of

Figure 2. (a) UV-Vis spectrum of Ru(bpy)3Cl2 in solution (offset for easy comparison), (b) UV-Vis diffuse reflectance spectra (DRS) of dried RuPOX, and (c) powdered HTiNbO5


Page 23 of 42 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


1

UV-Vis-DRS spectrum of the source material, HTiNbO5 is shown in Fig. 2-c. The spectrum of

2

the HTiNbO5 shows that there is no absorption above 400 nm, where the RuPOX material shows

3

absorption from the incorporated Ru(bpy)32+. This is to be expected as HTiNbO5 is a wide band

4

gap semiconductor that only absorbs in the UV, whereas the Ru(bpy)32+ absorption spectrum

5

includes part of the visible region.

6

The RuPOX material is saturated with Ru(bpy)32+, which serves as a sensitizer to visible

7

light. As an initial test of the photoactivity, the RuPOX material was used as a photocatalyst to

8

photodegrade aqueous solutions of bromocresol green. The RuPOX material was found to be

9

capable of oxidation and reduction catalysis under visible light. Details and results of this can be

10

found in the Supporting Information file (see S3).

11

The X-ray diffraction pattern of the RuPOX material shows only one intense peak at

12

9.37º 2θ, corresponding to a d-spacing of 9.43 Å (0.943 nm, see Fig. 3). The RuPOX material is

13

composed of restacked sheets, and when restacking, the sheets tend to adopt a coplanar

14

orientation.36 The single peak in the pattern corresponds to the interlayer d-spacing of the

15

restacked sheets in this material, in the stacking direction. This layer line reflection is commonly

16

found in restacked materials.32 The disappearance of the all the higher angle reflections also

17

indicates that the material is highly disordered and is turbostratic,32 with the expected loss of the

18

in-plane registry of the layers, normally found in the parent material, KTiNbO5, seen in Fig. 3-b.

19

The XRD pattern of the synthesized parent KTiNbO5 powder was matched with the JCPDS

20

pattern library file no: 054-1155. KTiNbO5 has an orthorhombic crystal structure.32 For clarity,

21

not all peaks are labeled here, but the library pattern match is conclusive. The peak width of the

22

single peak in the RuPOX X-ray pattern indicates that the restacking of the nanosheets has

23

occurred in a consistent manner, with the interlayer spacing remaining uniform throughout the 23 ACS Paragon Plus Environment


Page 24 of 42

1

material, thus creating a single phase composite. If the Ru(bpy)32+ cations were not filling each

2

interlayer completely and uniformly, and there were areas in the material with little or no

3

Ru(bpy)32+, then the X-ray pattern would likely show a second reflection at a greater angle. The

4

second reflection would correspond to the interlayer areas without Ru(bpy)32+ and would

5

possibly contain a different cation such as H+. Those areas would have a different d-spacing

6

than the areas that were completely filled with Ru(bpy)32+. In the case of partially filled layers,

7

any corrugation or varying d-spacing caused by partially filled interlayer areas would likely have

8

resulted in the broadening of the peak at 9.37° 2θ. Therefore, the X-ray data of the synthesized

9

RuPOX material (see Fig. 3) is consistent with the colloidal nanosheets being restacked into a

10

layered material, with single height layers of Ru(bpy)32+ uniformly filling the sheet interlayer

11

areas. The d-spacing of 0.943 nm is close to, but lower than, the diameter of Ru(bpy)32+ (1.01

12

nm), as reported in the literature.49 This may indicate that some distortion of the Ru(bpy)32+

13

molecules in the RuPOX solid may be present, but this is not clear. As discussed above, there

14

were changes in the UV-Vis absorption spectrum of Ru(bpy)32+ in the RuPOX solid versus the

15

aqueous phase Ru(bpy)32+, which could be, in part, due to molecular distortion of Ru(bpy)32+ and

16

its interaction with the host material.20,29,53

17


Page 25 of 42 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


Figure 3. X-ray diffraction pattern of (a) the RuPOX material (dried film), and (b) the powdered parent compound, KTiNbO5

1

3.3. Electron microscopy

2

Figure 4 shows a scanning electron microscopy (SEM) image of (a) HTiNbO5, and

3

images (b) and (c) which were air-dried samples of the RuPOX material. HTiNbO5 was derived

4

by cation exchange of KTiNbO5 and is known to be a crystalline material,30 which like

5

KTiNbO5, has a densely stacked layered structure. However, the new RuPOX material has a low

6

density and a random pore structure, with individual sheets randomly arranged, thus generating

7

the material’s porosity.30,43 The thickness of the individual sheets is approximately 1 nm,41,54,55

8

and the pore walls are composed of just a few restacked sheets, making the pore walls

9

very thin. The diameter of the clusters of the precipitated sheets in the RuPOX material is tens to

10

hundreds of microns, as seen with optical microscopy (See Supporting Information, Fig. S1).

11 25 ACS Paragon Plus Environment


Page 26 of 42

1

(a)

Foliated crystal habit

40 µm

(b)

Randomly distributed nanosheets

40 µm


Page 27 of 42 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


(c)

Restacked nanosheets form the pore walls

5 µm

Figure 4. Scanning electron microscopy of (a) HTiNbO5, (b) RuPOX, and (c) RuPOX at higher magnification showing the disordered nature of the structure.

1

However, the nanometer thickness of the pore walls gives the composite the characteristics of a

2

nanomaterial.30,43

3

diminished, due to the crushing intermolecular forces imposed by water as it dries.

Having been dried in air before imaging, the porosity of RuPOX is

4

Figure 5a is a transmission electron microscopy (TEM) image of the colloidal nanosheets

5

derived from HTiNbO5, showing that the compound was completely exfoliated, resulting in a

6

colloid of single (TiNbO5¯ )n nanosheets. The sizes of the sheets typically range from about 0.1

7

µm to 10 µm in diameter, which is similar to the sizes of the crystalline particles of the parent

8

HTiNbO5 powder.54 Figure 5b is a TEM image of a RuPOX material, showing the disorder of

9

the re-stacked and overlapping (TiNbO5¯ )n sheets. The thin darker lines seen on the sheets in

10

Fig. 5b are from the wrinkles and folds in the sheets. As discussed, the Ru(bpy)32+ molecules are

11



Page 28 of 42

1

(a)

(b)

Figure 5. Transmission electron microscopy of (a) TBA(TiNbO5) colloid, and (b) RuPOX. Black scale bars are 100 nm.


Page 29 of 42 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


1

sandwiched in between the nanosheets, electrostatically binding the solid together. TEM images

2

of the thicker regions of the RuPOX material were too opaque and did not reveal any

3

morphology. The morphology of RuPOX was revealed more clearly in the SEM data (Fig. 4b

4

and 4c).

5 6

3.4. Epifluorescence

7

An epifluorescence optical microscopy image of an air-dried RuPOX sample is shown in

8

Fig. 6. As suggested by the elemental analysis and X-ray data discussions above and the high

9

levels of Ru(bpy)32+ contained within the RuPOX samples, the Ru(bpy)32+ dye molecules are

10

located in between the sheets and are likely distributed evenly everywhere in single layers

11

between sheets.

12

excitation wavelength of 390 nm. The contrast and brightness of the image in Fig. 3 has been

13

adjusted to darken the brightest areas of the image in order to enhance the textural details,

14

however, this also darkened the less bright areas of the sample. Light emission, though uneven,

15

appears to emanate from all areas of the RuPOX material, suggesting that the Ru(bpy)32+

16

molecules are distributed everywhere. However, the edges of the restacked sheets can be seen

17

fluorescing more brightly than the center areas of the particles. The quenching of electronic

18

excited states in the Ru(bpy)32+ molecules by electron transfer into a semicinductor’s conduction

19

band is a well established phenomena.8,56 Other non-semiconductor materials loaded with

20

Ru(bpy)32+ would not necessarily exhibit similar quenching, and thus may fluoresce very

21

brightly and evenly in fluorescent microscopy.57 There are reports of metal organic frameworks

22

(MOFs) and other materials incorporating Ru(bpy)32+. However, significant differences exist

23

between the MOFs and the RuPOX material. The semiconductor nature of the metal oxide frame

The Ru(bpy)32+ fluoresces at around 600 nm,24 under the microscope’s



Page 30 of 42

1

in the RuPOX material provides a quenching mechanism for the excited state of the sensitizer,

2

via electron transfer into the conduction band.

3

Differences in the local environemnt surrounding the individual Ru(bpy)32+ molecules

4

may cause differences in the fluorescence quenching efficiency within the RuPOX sample, thus

5

uneven emission may be expected. For example, Ru(bpy)32+ molecules attached near the edges

6

may not be as efficiently quenched, as compared to the molecules that are deeper inside the solid,

Figure 6. Epifluorescence microscope image of an air-dried RuPOX sample (white scale bar is 5 microns), The inset image shows an enlargement of a particle with brighter edges.

7

which are completely covered by sheets, where the Ru(bpy)32+ may have more effective contact

8

with the metal oxide. Therefore, the edges may produce less quenching and brighter emissions.

9

The metal oxide sheets, which are a high dielectric material, may also act as a waveguides to the

10

emitted light, much like a fiber optic guide channels light to the ends of the fiber. This could 30 ACS Paragon Plus Environment

Page 31 of 42 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


1

enhance the brightness at the edges where the light is then transmitted out of the sample.

2

Although uneven distribution of Ru(bpy)32+ in the RuPOX could be considered a possible

3

explanation for the observed uneveness of fluorescence emission, the consistent pattern of bright

4

edges and darker centers throughout the sample, combined with the X-ray data and data showing

5

the high loading of dye suggests that this was not the case. It appears that the Ru(bpy)32+

6

molcules are indeed evenly distributed throughout the RuPOX samples, and that they are

7

sandwiched in single layers, between the nanosheets of the exfoliated metal oxide.

8 9

4. Conclusion

10

A new material that incorporates an exceptionally high loading of Ru(bpy)32+ into a

11

porous titanium niobium mixed oxide (RuPOX) was synthesized. In an aqueous solution-phase

12

mixing process, Ru(bpy)32+ was successfully used to aggregate colloids of lamellar anionic metal

13

oxide nanosheets, derived from the exfoliation of KTiNbO5. The resulting aggregate was found

14

to be a porous metal oxide composite material, with Ru(bpy)32+ imbedded in between

15

overlapping oxide nanosheets, forming the pore walls of the solid. Based on the electrostatic

16

stoichiometry, the new material contains 52.5 % of the theoretical maximum of Ru(bpy)32+ in its

17

structure, and using the criteria of only external surface area coverage, the new material contains

18

212.5 % of the theoretical maximum.

19

contained inside the new RuPOX materials, and this is equivalent to 295.6 mg/g. As calculated,

20

due to steric limitations and charge density considerations, these measured loadings are likely the

21

maximum that is achievable. Based on steric factors and charge densities, calculations suggest

22

that the measured loading exceeds the theoretical maximum. The new RuPOX composites are

23

stable after thorough rinsing, and the Ru(bpy)32+ molecules do not leach out into solution, and

Results show that 518.9 µmol/g of Ru(bpy)32+ are



Page 32 of 42

1

leaching did not occur after it was tested as a photocatalyst to degrade an organic dye. The

2

resulting porous solids were highly colored and contained an exceptionally high loading of the

3

Ru(bpy)32+ sensitizer. The UV-Vis-DRS absorption spectrum of the RuPOX material shows a

4

slighly expanded absorption range in the visible and peak red shifting, as compared to

5

Ru(bpy)32+ in solution. The new RuPOX material is a promising candidate for use as visible

6

light photocatalysts and other solar conversion technologies.

7

adaptable and could be used to investigate other sensitizers and layered semiconductors.

The synthetic approach is

8 9 10

Associated Content Supporting Information

11

A supporting information file is available, containing the following. S1 - Calculations on

12

the dye content analysis via electrostatic balance; S2 - Dye content analysis calculations

13

via 2-D surface charge densities; Figure S1 (a) Optical microscopy image of powdered

14

HTiNbO5 particles in water, (b) Optical microscopy image of the RuPOX material in

15

water; Figure S2 (a) UV-Vis absorption spectrum of the supernatant liquid of the

16

RuPOX in contact with acidic water solution for 2 days; (b) UV-Vis absorption spectrum

17

of the supernatant liquid of the well rinsed RuPOX in contact with DI water for months;

18

Figure S3 (a) Fourier transform infrared spectroscopy spectrum of dry powdered

19

tris(2,2'-bipyridyl)ruthenium(II) dichloride, and (b) An air-dried powder sample RuPOX ;

20

Figure S4 - Thermogravimetric analysis of HTiNbO5 and RuPOX; S3 and Figure S5 -

21

Visible light photocatalytic activity of the new RuPOX composite for the decolorization

22

of a dye in solution.


Page 33 of 42 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

1


Acknowledgements

2

The authors would like to thank Dr. Jorge Gardea-Torresdey, and Dr. Mahesh Narayan

3

from the University of Texas at El Paso for the use of some of their laboratory instrumentation.

4

A special thanks to Dr. Jose Hernandez and Dr. Cyren Rico for their generous help. Some of the

5

instruments used in this work were made available from the following grants: NSF CHE-

6

0840525, and the STARs program of the University of Texas System. Epifluorescence analysis

7

was supported by the National Institutes on Minority Health and Health Disparities (NIMHD)

8

grant #2G12MD007592, a component of the National Institutes of Health (NIH) at the Analytical

9

Cytology Core Facility at this institution. TEM analysis was supported through the Border

10

Biomedical Research Center (NIH-BBRC) grant #5G12MD007592.

11 12



Page 34 of 42

1

References

2

(1) Youngblood, W. J.; Lee, S. H. A.; Maeda, K., Mallouk, T. E. Visible Light Water Splitting

3

Using Dye-Sensitized Oxide Semiconductors; Acc. Chem. Res., 2009, 42, 1966-1973.

4

(2) Rao, C. N. R.; Lingampalli, S. R. Generation of Hydrogen by Visible Light-Induced Water

5

Splitting with the Use of Semiconductors and Dyes; Small., 2016, 12, 16-23.

6

(3) Kumar, S. G.; Rao, K. K. Comparison of Modification Strategies Towards Enhanced Charge

7

Carrier Separation and Photocatalytic Degradation Activity of Metal Oxide Semiconductors

8

(TiO2, WO3 and ZnO); Appl. Surface Sci., 2017, 391, 124-148.

9

(4) Liu, X.; Iocozzia, J.; Wang, Y.; Cui, X.; Chen, Y.; Zhao, S.; Li, Z.; Lin, Z. Noble Metal-

10

Metal Oxide Nanohybrids with Tailored Nanostructures for Efficient Solar Energy Conversion,

11

Photocatalysis and Environmental Remediation; Ener. Environ. Sci., 2017, 10, 402-434.

12

(5) Lee, C.P.; Lin, R.Y.Y.; Lin, L.Y.; Li, C.T.; Chu, T.C.; Sun, S.S.; Lin, J.T.; Ho, K.C. Recent

13

Progress in Organic Sensitizers for Dye-Sensitized Solar Cells; RSC Adv., 2015, 5, 23810-

14

23825.

15

(6) Higashino, T.; Imahori, H. Porphyrins as Excellent Dyes for Dye-Sensitized Solar Cells:

16

Recent Developments and Insights; Dalton. Trans., 2015, 44, 448-463.

17

(7) Furlong, D. N.; Wells, D.; Sasse, W. H. Colloidal Semiconductors in Systems for the

18

Sacrificial Photolysis of Water: Sensitization of Titanium Dioxide by Adsorption of Ruthenium

19

Complexes; J. Phys. Chem., 1986, 90, 1107–1115.


Page 35 of 42 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


1

(8) Kim, Y.I.; Atherton, S.J.; Brigham, E.S.; Mallouk, T.E. Sensitized layered metal oxide

2

semiconductor particles for photochemical hydrogen evolution from nonsacrificial electron

3

donors. J. Phys. Chem. 1993 97, 11802-11810.

4

(9) Rillema, D. P.; Edwards, A. K.; Perine, S. C.; Crumbliss, A. L. Electrochemistry and

5

Photocurrents of the Tris(bipyridine)ruthenium(II) and Methyl Viologen Cations Immobilized in

6

Carrageenan Hydrogel; Inorg. Chem., 1991, 30, 4421–4425.

7

(10) (a) Maeda, K.; Eguchi, M.; Youngblood, W.J.; Mallouk, T.E. Niobium oxide nanoscrolls as

8

building blocks for dye-sensitized hydrogen production from water under visible light

9

irradiation; Chem. Mater. 2008, 20, 6770-6778.

(b) Xiaohu Zhang, Tianyou Peng,and

10

Shuaishuai Song; Recent advances in dye-sensitized semiconductor systems for photocatalytic

11

hydrogen production; J.Mater.Chem.A, 2016, 4, 2365

12

(11) Bae, E.; Choi, W.; Park, J.; Shin, H. S.; Kim, S. B.; Lee, J. S. Effects of Surface Anchoring

13

Groups (Carboxylate vs Phosphonate) in Ruthenium-Complex-Sensitized TiO2 on Visible Light

14

Reactivity in Aqueous Suspensions; J. Phys. Chem. B., 2004, 108, 14093-14101.

15

(12) Brown, D. G.; Schauer, P. A.; Borau-Garcia, J.; Fancy, B. R.; Berlinguette, C. P.

16

Stabilization of Ruthenium Sensitizers to TiO2 Surfaces through Cooperative Anchoring Groups;

17

J. Am. Chem. Soc., 2013, 135, 1692–1695.

18

(13) Zhang, L.; Cole, J. M. Anchoring Groups for Dye-Sensitized Solar Cells; ACS Appl. Mater.

19

Interfaces., 2015, 7, 3427−3455.

20

(14) Swetha, T.; Mondal, I.; Bhanuprakash, K.; Pal, U., Singh, S. P. First Study on Phosphonite-

21

Coordinated Ruthenium Sensitizers for Efficient Photocatalytic Hydrogen Evolution; ACS Appl.

22

Mater. Interfaces., 2015, 7, 19635-19642.



Page 36 of 42

1

(15) Grätzel, M. Recent Advances in Sensitized Mesoscopic Solar Cells; Acc. Chem. Res., 2009,

2

42, 1788–1798.

3

(16) Wee, K. R.; Brennaman, M. K.; Alibabaei, L.; Farnum, B. H.; Sherman, B.; Lapides, A. M.;

4

Meyer, T. J. Stabilization of Ruthenium(II) Polypyridyl Chromophores on Nanoparticle Metal-

5

Oxide Electrodes in Water by Hydrophobic PMMA Overlayers; J. Am. Chem. Soc., 2014, 136,

6

13514–13517.

7

(17) Hanson, K.; Brennaman, M. K.; Luo, H.; Glasson, C. R.; Concepcion, J. J.; Song, W.;

8

Meyer, T. J. Photostability of Phosphonate-Derivatized, RuII Polypyridyl Complexes on Metal

9

Oxide Surfaces; ACS Appl. Mater. Inter., 2012, 4, 1462-1469.

10

(18) Maeda, K.; Eguchi, M.; Lee, S.H.A.; Youngblood, W.J.; Hata, H.; Mallouk, T.E.

11

Photocatalytic hydrogen evolution from hexaniobate nanoscrolls and calcium niobate nanosheets

12

sensitized by ruthenium (II) bipyridyl complexes; J. Phys. Chem. B., 2009, 113, 7962-7969.

13

(19) Nakato, T.; Kusunoki, K.; Yoshizawa, K.; Kuroda, K.; Kaneko, M. Photoluminescence of

14

Tris(2,2’-bipyridine)ruthenium(II) Ions Intercalated in Layered Niobates and Titanates: Effect of

15

Interlayer Structure on Host-Guest and Guest-Guest Interactions; J. Phys. Chem., 1995, 99,

16

17896-17905.

17

(20) Yao, K.; Nishimura, S.; Imai, Y.; Wang, H.; Ma, T.; Abe, E.; Tateyama, H.; Yamagishi, A.

18

Spectroscopic and Photoelectrochemical Differences Between Racemic and Enantiomeric

19

[Ru(phen)3]2+ Ions Intercalated into Layered Niobate K4Nb6O17; J. Electroanal. Chem. 2001,

20

510, 144–148.

21

(21) Ma, J.; Wu, J.; Gu, J., Liu, L.; Zhang, D.; Xu, X.; Yang, X.; Tong, Z. Fabrication and

22

Spectroscopic, Electrochemical and Catalytic Properties of a New Intercalation Compound of


Page 37 of 42 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


1

K4Nb6O17 with Cationic Cobalt Porphyrin; J. of Molecular Cat. A Chemical, 2012, 357, 95–

2

100.

3

(22) Lee, S.H.; Park, Y.; Wee, K.R.; Son, H.J.; Cho, D.W.; Pac, C.; Choi, W. Kang, S.O.

4

Significance of Hydrophilic Characters of Organic Dyes in Visible-Light Hydrogen Generation

5

Based on TiO2; Org. Lett., 2010, 12, 460–463.

6

(23) Kim, W.; Tachikawa, T.; Majima, T.; Li, C.; Kim, H. J.; Choi, W. Tin-Porphyrin Sensitized

7

TiO2 for the Production of H2 Under Visible Light; Energy. Environ. Sci., 2010, 3 1789-1795.

8

(24) Kalyanasundaram, K.; Grätzel, M. Applications of Functionalized Transition Metal

9

Complexes in Photonic and Optoelectronic Devices; Coord. Chem. Rev., 1998, 177, 347–414.

10

(25) Unal, U.; Matsumoto, Y.; Tanaka, N.; Kimura, Y.; Tamoto, N. Electrostatic Self-Assembly

11

Deposition of Titanate (IV) Layered Oxides Intercalated with Transition Metal Complexes and

12

their Electrochemical Properties; J. Phys. Chem. B, 2003, 107, 12680-12689.

13

(26) Abe, R.; Shinohara, K.; Tanaka, A.; Hara, M.; Kondo, J. N.; Domen, K. Preparation of

14

Porous Niobium Oxides by Soft-Chemical Process and their Photocatalytic Activity; Chem.

15

Mater., 1997, 9, 2179-2184.

16

(27) Nakato, T.; Miyamoto, N. Sol–Gel Transition of Nanosheet Colloids of Layered Niobate

17

K4Nb6O17; J. Mater. Chem. 2002, 12, 1245–1246.

18

(28) Wang, Z. S.; Sasaki, T.; Muramatsu, M.; Ebina, Y.; Tanaka, T.; Wang, L.; Watanabe, M.

19

Self-Assembled Multilayers of Titania Nanoparticles and Nanosheets with Polyelectrolytes;

20

Chem. Mater. 2003, 15, 807-812.

21

(29) Unal, U.; Matsumoto, Y.; Tamoto, N.; Koinuma, M.; Machida, M.; Izawa, K. Visible Light

22

Photoelectrochemical Activity of K4Nb6O17 Intercalated with Photoactive Complexes by

23

Electrostatic Self-Assembly Deposition; J. Solid State Chem., 2006, 179, 33-40.



Page 38 of 42

1

(30) Saupe, G. B.; Zhao, Y.; Bang, J.; Yesu, N. R.; Carballo, G. A.; Ordonez, R.; Bubphamala,

2

T. Evaluation of a New Porous Titanium-Niobium Mixed Oxide for Photocatalytic Water

3

Decontamination; Microchem. J., 2005, 81, 156-162.

4

(31) Ebina, Y.; Sakai, N.; Sasaki, T. Photocatalyst of Lamellar Aggregates of RuOx-Loaded

5

Perovskite Nanosheets for Overall Water Splitting; J. Phys. Chem. B 2005, 109, 17212-17216.

6

(32) Li, L.; Ma, R.; Ebina, Y.; Fukuda, K.; Takada, K.; Sasaki, T. Layer-by-Layer Assembly and

7

Spontaneous Flocculation of Oppositely Charged Oxide and Hydroxide Nanosheets into

8

Inorganic Sandwich Layered Materials; J. Am. Chem. Soc., 2007, 129, 8000–8007.

9

(33) Gunjakar, J. L.; Kim, T. W.; Kim, H. N.; Kim, I. Y.; Hwang, S. J. Mesoporous Layer-by-

10

Layer Ordered Nanohybrids of Layered Double Hydroxide and Layered Metal Oxide: Highly

11

Active Visible Light Photocatalysts with Improved Chemical Stability; J. Am. Chem. Soc. 2011,

12

133, 14998–15007.

13

(34) Ebina, Y.; Sasaki, T.; Harada, M.; Watanabe, M. Restacked Perovskite Nanosheets and their

14

Pt-Loaded Materials as Photocatalysts; Chem. Mater. 2002, 14, 4390-4395.

15

(35) Ma, J.; Wu, J.; Zheng, J.; Liu, L.; Zhang, D.; Xu, X.; Yang, X.; Tong, Z. Synthesis,

16

Characterization and Electrochemical Behavior of Cationic Iron Porphyrin Intercalated into

17

Layered Niobate. Microporous Mesoporous Mater. 2012, 151, 325-329.

18

(36) Kawashima, M.; Mori, K.; Aoyama, J.; Yamashita, H. Synthesis and Characterization of Ir

19

and Rh Complexes Supported on Layered K4Nb6O17 as a Heterogeneous Photocatalyst for

20

Visible-Light-Induced Hydrogen Evolution; Bull. Chem. Soc. Jpn. 2014, 87, 874-881.

21

(37) Bizeto, M. A.; Constantino, V. R. Porphyrin Inclusion into Hexaniobate Nanoscrolls;

22

Micropor. and Mesopor. Mater; 2005, 83, 212–218.


Page 39 of 42 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


1

(38) Wadsley, A.D. Alkali titanoniobates. The crystal structures of KTiNbO5 and KTi3NbO9.

2

Acta Crystallographica. 1964, 17, 623-628.

3

(39) Rebbah, H.; Desgardin, G.; Raveau, B. Les Oxydes ATiMO5: Echangeurs Cationiques.

4

Mater. Res. Bull., 1979, 14, 1125-1131.

5

(40) Keller, S. W.; Kim, H. N.; Mallouk, T. E. Layer-by-Layer Assembly of Intercalation

6

Compounds and Heterostructures on Surfaces: Toward Molecular “Beaker" Epitaxy; J. Am.

7

Chem. Soc. 1994, 116, 8817-8818.

8

(41) Fang, M.; Kim, C.H.; Mallouk, T.E. Dielectric properties of the lamellar niobates and

9

titanoniobates AM2Nb3O10 and ATiNbO5 (A= H, K, M= Ca, Pb), and their condensation

10

products Ca4Nb6O19 and Ti2Nb2O9. Chem. Mater. 1999, 11, 1519-1525.

11

(42) Zhang, O.; Lin, B.Z.; Chen, Y.L.; Gao, B.F.; Fu, L.M.; Li, B. Electrochemical and

12

photoelectrochemical characteristics of TiNbO5 nanosheet electrode. Electrochimica Acta. 2012,

13

81, 74-82.

14

(43) Masud, S.; Zarei, M.; Lopez, M. L.; Gardea-Torresdey, J.; Ramana, C. V.; Saupe, G. B.

15

Photoreduction of Metallic Co-catalysts onto Novel Semiconducting Metal Oxides; Mater. Sci.

16

& Eng. B 2010, 174, 66–70.

17

(44) Sasaki, T.; Watanabe, M.; Hashizume, H.; Yamada, H.; Nakazawa, H. Macromolecule-like

18

Aspects for a Colloidal Suspension of an Exfoliated Titanate. Pairwise Association of

19

Nanosheets and Dynamic Reassembling Process Initiated from It; J. Am. Chem. Soc. 1996, 118,

20

8329–8335.

21

(45) Kobayashi, Y.; Hata, H.; Salama, M.; Mallouk, T.E. Scrolled sheet precursor route to

22

niobium and tantalum oxide nanotubes. Nano lett. 2007, 7, 2142-2145.



Page 40 of 42

1

(46) Brennaman, M.K.; Patrocinio, A.O.T.; Song, W.; Jurss, J.W.; Concepcion, J.J.; Hoertz,

2

P.G.; Traub, M.C.; Murakami Iha, N.Y.; Meyer, T.J. Interfacial Electron Transfer Dynamics

3

Following Laser Flash Photolysis of [Ru(bpy)2((4, 4′‐PO3H2)2bpy)]2+ in TiO2 Nanoparticle Films

4

in Aqueous Environments; ChemSusChem, 2011, 4, 216-227.

5

(47) Bujdák, J.; Janek, M.; Madejová, J.; Komadel, P.. Influence of the layer charge density of

6

smectites on the interaction with methylene blue. J. of the Chem. Society, Faraday

7

Transactions, 1998, 94, 3487-3492.

8

(48)

9

Education, 2016.

Chemistry, by Raymond Chang and Kenneth Goldsby, 12th Edition, McGraw Hill

10

(49) Rudzinski, W. E.; Bard, A. J. Clay Modified Electrodes: Part VI. Aluminum and Silicon

11

Pillared Clay-Modified Electrodes. J. Electroanal. Chem. and Interfacial. Electrochem. 1986,

12

199, 323-340.

13

(50) Sheridan, M.V.; Sherman, B.D.; Marquard, S.L.; Fang, Z.; Ashford, D.L.; Wee, K.R.; Gold,

14

A.S.; Alibabaei, L.; Rudd, J.A.; Coggins, M.K.; Meyer, T.J., Electron Transfer Mediator Effects

15

in Water Oxidation Catalysis by Solution and Surface-Bound Ruthenium Bpy-Dicarboxylate

16

Complexes. J. Phys. Chem. 2015, 119, 25420− 25428.

17

(51) Rawling, T.; Austin, C.; Buchholz, F.; Colbran, S. B.; McDonagh, A. M. Ruthenium

18

Phthalocyanine-Bipyridyl Dyads as Sensitisers for Dye-Sensitised Solar Cells: Dye Coverage

19

versus Molecular Efficiency. Inorg. Chem. 2009, 48, 3215-3227.

20

(52) Kalyanasundaram, K. Photophysics, Photochemistry and Solar Energy Conversion with

21

Tris(bipyridyl)ruthenium(II) and its Analogues; Coord. Chem. Rev. 1982, 46, 159-244.

22

in Aqueous Environments; ChemSusChem, 2011, 4 216-227.


Page 41 of 42 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


1

(53) Martí, A. A.; Colón, J. L. Photophysical Characterization of the Interactions among

2

Tris(2,2′-bipyridyl)ruthenium(II) Complexes Ion-Exchanged within Zirconium Phosphate. Inorg.

3

Chem., 2010, 49, 7298–7303.

4

(54) Du, G.H.; Yu, Y.; Chen, Q.; Wang, R.H.; Zhou, W.; Peng, L.M. Exfoliating KTiNbO5

5

particles into nanosheets. Chem. Phy. Lett. 2003, 377, 445-448.

6

(55) Fang, M.; Kim, C.H.; Saupe, G.B.; Kim, H.N.; Waraksa, C.C.; Miwa, T.; Fujishima, A.;

7

Mallouk, T.E. Layer-by-layer growth and condensation reactions of niobate and titanoniobate

8

thin films. Chem. Mater. 1999, 11, 1526-1532.

9

(56) Kamat, P.V. Photoelectrochemistry in particulate systems. 9. Photosensitized reduction in a

10

colloidal titania system using anthracene-9-carboxylate as the sensitizer. J. Phys. Chem., 1989,

11

93, 859-864.

12

(57) Lin, X.; Luo, F.; Zheng, L.; Gao, G.; Chi, Y. Fast, sensitive, and selective ion-triggered

13

disassembly and release based on tris(bipyridine)ruthenium(II)-functionalized metal–organic

14

frameworks. Analytical chemistry, 2015, 87, 4864-4870.

15

(58) Saupe, G.B.; Mallouk, T.E.; Kim, W.; Schmehl, R.H. Visible light photolysis of hydrogen

16

iodide using sensitized layered metal oxide semiconductors: the role of surface chemical

17

modification in controlling back electron transfer reactions. J. Phys. Chem. B. 1997, 101, 2508-

18

2513.

19 20 21 22 23



Page 42 of 42

1 2

For table of content use:

3 4 5

6 7


Exceptional Sensitizer Dye Loading via a New Porous Titanium

Recommend Documents