Probing the Surface Reactivity of Nanocrystals by the Catalytic Degradation of Organic Dyes: The Effect of Size, Surface Chemistry and Composition

We herein present a comprehensive study on how the catalytic performance and reusability of Au nanocrystals (NCs) are affected by systematic variations of crystal size, surface coating and composition. The reduction of different organic dyes (4-Nitrophenol, Rhodamine B and Methylene blue) by borohydride ions were used as model catalytic reactions. The catalytic performance of the Au NCs ranged between 3.6 to 110 nm was found to be dependent on crystal size, indicating that Au surface atoms have a distinct size-dependent reactivity in this reaction. Similarly, catalytic performance was found to be strongly dependent on the nature of the coating molecule, obtaining lower catalytic activities for molecules strongly bounded to Au surface. Finally, catalytic performance was found to be dependent on the chemical composition of the NC (Au, Ag, Pt) and the model dye used as a testing system, with the highest degradation rate found for Methylene blue, followed by 4-Nitrophenol and Rhodamine B. We believe that this study provides a better understanding of the catalytic performance of Au NCs upon controlled modifications of structural and morphological parameters, and working environment that can be used to facilitate the selection of the optimum NC size, coating molecules and evaluation systems for a particular study of interest. Page 1 of 23 Journal of Materials Chemistry A Jo ur na lo fM at er ia ls C he m is tr y A A cc ep te d M an us cr ip t Pu bl is he d on 1 5 M ay 2 01 7. D ow nl oa de d by U ni ve rs ita t A ut on om a de B ar ce lo na o n 16 /0 5/ 20 17 1 1: 28 :4 1. View Article Online DOI: 10.1039/C7TA01328K

direct route is a stepwise hydrogenation process involving the formation of two intermediate products, nitrosobenzene and phenylhydroxylamine.In the condensation route, intermediate product was azobenzene. 2 When formed, 4-AP detaches from the metal NCs surface, the next cycle of new catalytic reduction can be triggered again.
Degradation of MB to Leuco MB and RhB to Leuco RhB.The catalytic reduction of RhB and MB can be also explained by an electrochemical mechanism, where the metal NCs serve as an electron relay system for the oxidant and reductant species.First BH 4 − ions and MB or RhB molecules are adsorbed together onto the surface of the metal NCs.Then electron transfer takes place between the dyes and BH 4 − through particle surface.After receiving the electrons, the dye molecules are reduced to Leuco MB or Leuco RhB.

Figure S3
. Comparison of the ratio of catalytically active Au atoms for 2, 5, 10, 15, 20 and 30 nm NCs.The ratio of Au active atoms if found to be 55.83%, 21.97%, 15.95%, 10.63%, 7.48% and 6.67%, respectively.Corresponding HRTEM experimental images for all the simulated models are also presented.[6] On the basis of above presented HRTEM observations, we propose a model of spheroidal decahedron shape particles, which is present in all samples.We must say here, that for bigger nanoparticles the {111} external facets tend to be sharper and the nanoparticles have a morphology closer to the ideal decahedron, losing the spheroidal shape.Regular decahedron shape consists of the merging of 5 {111} faceted tetrahedral. 5,7,8 Yt, the theoretical angle between to (111) planes is 70.53 o and combination of 5 tetrahedrons to form a decahedron results in a 7.35 o gap 7 , which must be filled by some form of internal strain such as dislocations and other structural defects. 6,8 ohnson et al. 8 reports that elastic anisotropy, results in an internal lattice rotation of 4.3 o , combined with about 0.6 o shear-strain for each tetrahedron accommodate this 7.35 o gap in their experimental decahedron shaped Au nanoparticle.
Here, we assume a homogeneous distribution of these strains to form a regular decahedron out of 5 {111} faceted tetrahedra with a rotation angle of 72 o .In Figure S9 we have modeled a regular Au decahedron with a size of 10 nm. Figure S6 shows the stages of this modeling process.However, as can be seen in the above presented HRTEM images, such a model cannot be representative for the present samples.Most of the reported Au decahedral shapes having sizes smaller than 30 nm reveals similar spheroidal behavior [5][6][7] , however, modeling and simulation studies are conducted as if they were perfect decahedrons. 7,9  to above presented discrepancy between experimentally observed nanoparticles and suggested regular decahedral models, we propose a spheroidal decahedron model, which contains 5 {111} facets but has a spheroidal like structure.Figure S10 shows the steps of the spheroidal decahedron modeling process.First, we have created a {111} faceted segment over a tetrahedron.Then, we merge 5 of them with a sequential rotations of 72 o forming a 5 twinned spheroidal decahedral structure.We extended the spheroidal decahedral model and added a shell in order to calculate the ratio of catalytically active surface atoms over volume atoms.Figure S11 shows the catalytically active surface atoms in red and volume atoms in yellow.For the case of 10 nm spheroidal decahedron, total number of atoms is found to be 30245, where 4272 of the total atoms are in the surface.Percent ratio of catalytically active surface Au atoms is 14.12% for this model.
In addition to 10 nm model, we have modeled particles with different sizes such as 1 nm, 2 nm, 4 nm, 6 nm, 8 nm, 12 nm and 15 nm in order to compare the variation of the ratio of surface active Au atoms with size (Fig. S12).

Figure S2 .
Figure S2.TEM images analysis of Au NPs shown in Figure 1.Au seeds diameter increase from ~3.6 to ~110 nm after different growth steps.At least 1000 NPs were counted for each size.

Figure S5 .
Figure S5.CTAB degradation of 4-Nitrophenol.The catalytic performance of CTAB was evaluated by added an aqueous solution of CTAB molecules (0.1 mL, 10 mM) into a solution of 4-NP and NaBH4, obtaining that CTAB molecules are able to rapidly degrade 4-NP by themselves.Similarly, the drop observed at short reaction times for CTAB-coated Au NCs can be attributed to the free of bound CTAB molecules.

Figure S6 .
Figure S6.Optical interference between 4-Nitrophenol, Rhodamine B and Methylene blue and Au NCs of 10 nm (A), 50 nm (B), Ag NCs (C) and Pt NCs (D).In the case of Au and Ag spectra were normalized to the maximum absorption of the dyes.

Figure S9 .
Figure S9.Modeling a regular Au decahedron with a size of 10 nm: (i) creation of a {111} faceted tetrahedron, (ii)-(iii) combination of 5 individual tetrahedra with sequential rotations of 72 o .

Figure S10 .
Figure S10.Modeling a spheroidal Au decahedron with a diameter of 10 nm: (i) creation of a {111} faceted spheroidal subunit, (ii)-(iii) combination of 5 individual subunits with sequential rotations of 72 o .

Figure S11 .
Figure S11.A spheroidal Au decahedron with a diameter of 10 nm, showing the surface active atoms in red.

Figure S12 .
Figure S12.Comparison of the ratio of catalytically active Au atoms with increasing particle diameters for different Au nanostructures such as sphere, spheroidal decahedron, cuboctahedron and cube.The results plotted here have been summarized in an animated video movie that can be reached by following the link in ref 10 .