Biomaterials 48 (2015) 37e44

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Mechanism of pH-switchable peroxidase and catalase-like activities of gold, silver, platinum and palladium Junnan Li a, Wenqi Liu b, Xiaochun Wu b, *, Xingfa Gao a, * a

CAS Key Laboratory for Biomedical Effects of Nanomaterials and Nanosafety, Institute of High Energy Physics, Chinese Academy of Sciences, Beijing 100049, China b CAS Key Laboratory of Standardization and Measurement for Nanotechnology, National Center for Nanoscience and Technology of China, Beijing 100190, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 12 August 2014 Accepted 20 January 2015 Available online 9 February 2015

Despite being increasingly used as artificial enzymes, little has been known for the origin of the pHswitchable peroxidase-like and catalase-like activities of metals. Using calculations and experiments, we report the mechanisms for both activities and their pH-switchability for metals Au, Ag, Pd and Pt. The calculations suggest that both activities are intrinsic properties of metals, regardless of the surfaces and intersections of facets exposed to environments. The pre-adsorbed OH groups on the surfaces, which are only favorably formed in basic conditions, trigger the switch between both activities and render the pHswitchability. The adsorption energies between H2O2 and metals can be used as convenient descriptors to predict the relative enzyme-like activities of the metals with similar surface morphologies. The results agree with the enzyme-mimic activities that have been experimentally reported for Au, Ag, Pt and predict that Pd should have the similar properties. The prediction, as well as the predicted activity order for the four metals, has been verified by the experimental tests. The results thus provide an in-depth insight into the peroxidase-like and catalase-like activities of the metals and will guide the de novo design, synthesis and application of artificial enzymes based on inorganic materials. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Catalysts Density-functional calc. Metals Reaction mechanisms Surface

1. Introduction Materials in the nanometer size range have fundamentally new properties that are not owned as they are in large bulks. This makes it possible to develop novel applications using nanotechnologies. Recently, various nano-structured materials have been reported to possess intrinsic enzyme mimetic activities similar to those found in natural peroxidases or catalases. These enzyme mimics can be categorized into three classes: carbon, metal, and metal oxide nanomaterials [1]. As for metal oxides, Yan and co-workers have reported that Fe3O4 magnetic nanoparticles (NPs) possess intrinsic peroxidase-like activity when the pH of solution is 3.5, and Gu and coworkers have reported that these particles alternatively possess catalase-like activity in a neutral pH condition, though they are conventionally thought to be biologically inert [2,3]. Therefore, Fe3O4 NPs can be used as peroxidase mimics to detect H2O2 and thrombin. Since this pioneering work, other nanomaterials with

* Corresponding authors. E-mail addresses: [email protected] (X. Wu), [email protected] (X. Gao). http://dx.doi.org/10.1016/j.biomaterials.2015.01.012 0142-9612/© 2015 Elsevier Ltd. All rights reserved.

similar properties have been rapidly developed. The synthesized FeS nanosheets display high peroxidase-like activity and good stability, which is thus used as an enzyme mimic for the development of biocatalysts and amperometric biosensors [4]. V2O5 nanowires with a variable length between 500 nm and a few micrometers and a width of 100 nm present an intrinsic catalytic activity towards classical peroxidase substrates in slightly acidic environments (pH ¼ 4.0), such as 2,2-azino-bis (3ethylbenzothiazoline-6-sulfonic acid) (ABTS) and 3,3,5,5tetramethylbenzidine (TMB), in the presence of H2O2 [5]. As for carbon nanomaterials, carboxyl-modified graphene oxide (GOCOOH) can serve as an effective peroxidase mimic, which catalyzes the reaction of peroxidase substrate TMB in the presence of H2O2 to produce a blue color. Similar to horseradish peroxidase (HRP), the catalytic activity of the GOCOOH is dependent on pH, temperature and H2O2 concentration [6]. Single-wall carbon nanotubes (SWNTs) also possess the intrinsic peroxidase-like activity, which can be used in the label-free colorimetric detection system for disease-associated single nucleotide polymorphism (SNPs) in human DNA [7]. For the exploitation of new enzyme-like functions of known nanomaterials, the noble metal NPs, including

38

J. Li et al. / Biomaterials 48 (2015) 37e44

Ag, Au and their shell alloys like Au@Pt and AgAu@Pt have attracted more attention, because they exhibit multiple enzyme-mimetic properties [8e15]. Interestingly, these multiple enzyme-mimetic activities have been found to be pH-switchable. For instance, Yin and co-workers have reported that Ag and Au NPs can catalyze the rapid decomposition of H2O2. The decomposition of H2O2 is accompanied by the formation of OH radicals at lower pH conditions and O2 at higher pH conditions, showing a pH-switchable peroxidase-like and catalase-like activities. When pH ¼ 7.4, these particles also present the SOD-like activity [11,12]. These functions, along with their good biocompatibility, have made Au and Ag NPs promising candidates for use in biosensors. Because tumor issues usually have a more acidic environment than normal tissues, they might also be beneficial for intelligent cancer treatments, which produce OH radicals to kill tumor cells in tumor tissues and reduce the H2O2 level to relieve the oxidative stress in normal tissues. On the other hand, bimetallic nanostructures have been demonstrated to exhibit improved catalytic performance, because of the synergistic and electronic effects [14]. Some of the Ag-based bimetallic nanostructures with proper Ag ratios exhibit stable peroxidase-like activity at pH ¼ 4.5, which can be further tailored by alloy composition and solution pH value [9]. Because of their low costs, tunable compositions and structures, and high stabilities, NPs consisting of novel bimetallic alloys serve as a promising candidate of mimetic enzymes and may have a wide range of new applications, such as immunoassay, biocatalyst, and environmental monitoring. Despite being increasingly used as enzyme mimics in various fields, little has been known for the chemistry substantializing the experimental observations and the corresponding practical applications [16e18]. It is highly desirable to investigate the mechanisms underlying the enzyme-like activities and especially the pHswitchability for the above mentioned materials. Calculations have been indispensable to unravel reaction mechanisms at atomistic levels [19e22]. Here, using calculations and experiments, we study the mechanisms for the pH-switchable peroxidase-like and catalase-like activities of Au, Ag, Pt and Pd nanomaterials (Fig.1). The calculations suggest that both activities are intrinsic properties of the metals, regardless of the surfaces exposed to environments. The intersections of the facets are also considered using the Au(211) surface, which consists of repeated steps. The results agree well with the enzyme-mimic activities that have been experimentally reported for Au, Ag, Pt and predict that Pd should have the similar properties. The prediction, as well as the predicted activity order for the metals, has been verified by the experimental tests. The results thus provide an in-depth insight into the enzymatic activities of metals and will guide the de novo design, synthesis and application of artificial enzymes based on inorganic materials.

2. Methods 2.1. Calculations All geometry optimizations and energy calculations were performed with periodic slab models using the VASP code [23e25]. The generalized gradient approximation (GGA) was used with the PerdeweBurkeeErnzerhof functional (PBE) [26]. The optimizations were performed using an energy cut-off for the plane-wave basis set of 400 eV, and a second-order MethfesselePaxton smearing of 0.2 eV. The Au(111) surface was modeled by a four-layered slab in (111) direction, and a p(3  3) unit cell in the lateral directions. For the Au(110) and Au(211) surfaces including repeated steps, eight-layer slabs with (2  2) unit cell were used for the Au(110) surface, and six-layer slabs with (4  2) unit cell for Au(211), respectively. There was a vacuum space with a height of 15 Å in the vertical direction of slab. The MonkhorstePack scheme with 3  3  1, 2  2  1 and 2  1  1 k-point samplings were used for (3  3), (2  2) and (4  2) unit cells, respectively. During geometry optimizations, the top two layers for Au(111), the top four layers for Au(110) and the top three layers for Au(211) were fully relaxed, respectively, and the remaining bottom layers were frozen at the corresponding bulk face-centered-cubic (fcc) lattice positions. All the electronic structures were converged to 106eV and the geometries were optimized until the forces were smaller in magnitude than 0.02 eV/Å. Our optimized lattice constants for Au surfaces were in agreement with the experimental values [27]. For most structures, a spin-unpolarized scheme was used. However, for those with the presence of isolated O2 species, a spin-polarized scheme was used. Our test calculations showed that both schemes obtained almost the same energetic value for geometries without O2. The Pd(111), Pt(111) and Ag(111) surfaces were modeled using the similar method as that for Au(111) surface. The optimized lattice constants were also in agreement with the respective experimental values [28]. The adsorption energies were calculated using the following equation: Eads ¼ Eslabþmol  ðEslab þ Emol Þ where Eslabþmol denotes the total energy of the Au slab with adsorbates on it, and Eslab and Emol are energies of the isolated slab and adsorbate, respectively. The climbing image implementation of the nudge-elastic band (CI-NEB) and dimer methods were used for locating the minimum energy path between energy minima, as associated with the saddle point search [29]. The convergence was at a reduced force threshold of 0.02 eV/Å [30]. Vibrational frequency for each of the transition states was calculated within the harmonic approximation, and only one imaginary frequency was found along the reaction coordinate [31]. The solvent effect on the surface reaction was evaluated using a water cluster consisting of n (n ¼ 1e3) water molecules, and the surrounding water was modeled using the hexagonally-packed ice-like water adsorbed on the metal surface [32e34]. The corresponding adsorption energies were also calculated by equation (1), in which Eslabþmol represents the total energy of reactants and water layers adsorbed on the metal surface, the Eslab is the energy of the metal surface with adsorbed water layers, and Emol is energy of the isolated adsorbate. The electric field calculations were performed using standard features provided within VASP. 2.2. Experiments 2.2.1. Chemicals Chlorauric acid (HAuCl43H2O), potassium(Ⅱ) tetrachloroplatinate (K2PtCl4), palladium chloride (PdCl2), silver nitrate (AgNO3), Sodium borohydride (NaBH4), cetyltrimethylammonium bromide (CTAB), L-ascorbic acid (AA), poly(sodiumpstyrensulfonate) (PSS, mw 70 000), o-phenylenediamine (OPD), 30% H2O2 were all purchased from Alfa Aesar and used as received. Sulfuric acid (H2SO4), sodium hydroxide (NaOH) were purchased from Beijing Chemical Reagent Company. Milli-Q water (18 MU cm) was used for all solution preparation. 2.2.2. Synthesis of gold nanorods The nanorods were synthesized by using a seed-mediated method. CTABcapped seeds were prepared by chemical reduction of HAuCl4 with NaBH4. A freshly prepared, ice-cold NaBH4 solution (0.01 M, 0.6 mL) was injected into the mixture of CTAB (0.1 M, 7.5 mL), HAuCl4 (24 mM, 0.1 mL), and water (1.8 mL). The mixture was kept stirring for 3 min and aged in a 30  C water bath for 2e5 h before use. The growth solution was made by the sequential addition of CTAB (0.1 M, 100 mL), HAuCl4 (24 mM, 2.04 mL), AgNO3 (0.1 M, 105 mL), and AA (0.1 M, 552 mL) solutions. 120 mL seed solution was injected into the growth solution to initiate the growth of GNRs. The resultant reaction solution was gently mixed by inversion and then left undisturbed. After 12 h, AA (0.1 M, 55.2 mL) was added twice with 40 min interval. The reaction mixture was reacted for 24 h. The nanorods were centrifuged twice (9200 rpm, 10 min) and redispersed either in 50 mL deionized water or 50 mL 0.1 M CTAB aqueous solution.

Fig. 1. The pH-switchable peroxidase-like and catalase-like activities of Au, Ag, Pt and Pd metals, in which the metals catalyze the oxidation of organic substrates and the decomposition of H2O2 to give O2, respectively.

2.2.3. Coating of Ag, Pd and Pt on the GNRs For the synthesis of Au@Pd NRs: 2 mL 0.1 M CTAB aqueous solution and 1 mL purified Au NR dispersed in 0.1 M CTAB were first mixed together. Then, 65 mL of 2.5 mM K2PdCl4 solution and 15 mL of 1 M H2SO4, 15 mL AA (0.1 M) was added. The mixture was shaken vigorously and placed in a 30  C water bath for 3 h. The

J. Li et al. / Biomaterials 48 (2015) 37e44 products were purified by centrifugation (12,000 rpm 5 min). The precipitate was redispersed in 100 (l deionized water). For the synthesis of Au@Pt NRs, 5 mL 10 mM CTAB was first added in 1 mL purified Au NRs suspension and then the mixture was diluted by adding 1 mL deionized water. After that, 75 mL 2 mM K2PtCl4 solution and 15 mL 0.1 M AA was added. The mixture was shaken vigorously and then kept in 30  C water bath for 30 min. The products were purified by centrifugation (12,000 rpm 5 min). The precipitates were then redispersed in 100 mL deionized water. For the synthesis of Au@Ag NRs, 1 mL purified Au NRs was mixed with CTAB (1 mL, 0.1 M), 1 mL H2O, AgNO3 (15 mL, 10 mM), NaOH (300 mL, 0.2 M), AA (15 mL, 0.1 M). The calculated Ag/Au molar ratio is 0.3. Then the mixture was kept in a 30  C water bath for 12 h. The products were purified by centrifugation (12,000 rpm 5 min). The precipitate was redispersed in 100 mL deionized water. 2.2.4. Coating nanorods with PSS The CTAB-coated NRs suspension (1.0 mL) was centrifuged at 12,000 rpm for 5 min. The precipitate was then redispersed in 1.0 mL water, and the concentration of CTAB was about 1 mM. Then 50 mL 20 mg mL1 PSS solution (containing 60 mM NaCl) was added. The suspension was placed in a 30  C water bath for at least 3 h. After that, one time centrifugation (12,000 rpm 5 min) was executed to remove the excessive PSS and the precipitate was redispersed in water for further use. 2.2.5. Enzyme mimic activity Peroxidase-like activity. 10 mL 0.1 M OPD, 50 mL 5 nM NRs and 50 mL 1 M H2O2 were added in 2.39 mL 0.1 M PBS buffers with different pH values. In the case of Au@Pt NRs, 5 mL 5 nM NRs was used instead. The reaction kinetics was monitored with 1 min interval by recording extinction spectra in a scanning kinetics mode. Unless indicated, reaction temperature is 30  C. Catalase-like activity. 50 mL 5 nM NRs were diluted into 2.4 mL 0.1 M PBS buffers with different pH values first. Then, 50 mL 1 M H2O2 were added. The reaction kinetics was monitored with 1 min interval by recording the extinction spectra in a scanning kinetics mode. 2.2.6. Characterizations Extinction spectra were obtained on a Cary 50 UV/vis/NIR absorption spectrometer. TEM images were taken from a Tecnai G2 20 S-TWIN TEM with accelerating voltage of 200 kV. Scan TEM (STEM) and energy-dispersive X-ray analysis (EDX) element mappings were conducted under a higheangle annular dark field (HAADF) mode using a Tecnai G2 F20 U-Twin microscope. Dissolved oxygen was measured by Mettler Toledo SG9-ELK dissolved oxygen meter.

3. Results and discussion

39

Au, at different pH conditions [35]. The structure of Au(111) surface is shown in Fig. 2. The calculations suggested that the H2O2 favorably adsorb on the top of Au (111) surface at a neutral condition (for other less stable adsorbing structures, see part 3 of SI). The adsorption energy is 0.11 eV. Because the calculated adsorption energy for H2O on the Au surface is 0.07 eV, which is more positive than that between H2O2 and the surface, H2O2 will occupy the Au surface more thermodynamically favorably than H2O does. The adsorption energy of H2O monomer on Au(111) was calculated to be 0.105 eV, previously [36]. Hence, despite the abundance of H2O in aqueous solutions, water molecules will not prevent the Au surfaces from H2O2 adsorption. In acidic and basic conditions, the Hþ and OH ions will adsorb on the Au surfaces [37,38]. The adsorption of Hþ on the surface in low-pH conditions is easy because of the intrinsic negative charges carried on metal surfaces in aqueous conditions [39]. The adsorption of OH on Au with the presence of NaOH has been experimentally verified [40]. The adsorption energy for H on the Au surface is 3.30 eV. That for H2O2 on the surface with the presence of a pre-adsorbed H is 3.38 eV, which is slightly more positive than the sum for the individual adsorption of H and H2O2 on the surface (3.41 eV). This means that the pre-adsorbed H slightly weakens the H2O2eAu interactions. Thus H2O2 will avoid the preadsorbed H when adsorbing on the Au surfaces in acidic conditions. However, with a pre-adsorbed OH, the adsorption of H2O2 on the surface becomes stronger, for which the adsorption energy is 2.07 eV. This co-adsorption energy is more negative than the sum of the individual adsorption energies for H2O2 and OH on the surface (0.11 and 1.55 eV). This means that the pre-adsorbed OH augments the interacting force between H2O2 and Au surfaces. Similar cooperative adsorption of CO with NO2 and CO with OH on Au(111) surface have been reported before [40,41]. Thus H2O2 favors the vicinities of pre-adsorbed OH when adsorbing on Au surfaces in basic conditions.

3.1. The adsorption characters of H2O2 molecule on the Au(111) surface

3.2. The decomposition reactions of H2O2 on the Au surfaces in different pH conditions

We firstly studied the stable configuration of a H2O2 molecule adsorbed on the Au(111) surface, which is the most stable facet of

We then studied the decomposition reactions of H2O2 on the Au(111) surface in a neutral condition. As shown in Fig. 3A, the

Fig. 2. Structures of Au(111), Au(110), Au(211), Ag(111), Pt(111) and Pd(111) surfaces.

40

J. Li et al. / Biomaterials 48 (2015) 37e44

adsorbed H2O2 molecule can undergo two types of decompositions: acid-like decomposition and base-like decomposition (for structures, see part 4 of SI).

Acid  like decomposition : H2 O2 4 H* þ HO*2

(1)

Base  like decomposition : H2 O2 4 H2 O* þ O*

(2)

Asterisk (*) is used to mark species adsorbed on metal surfaces. According to the acid-like pathway, the H2O2 molecule firstly breaks the OeH bond to give H* and HO*2 , which has an energy barrier of 1.60 eV. The subsequent decomposition of HO*2 can lead to the formation of H* and O2 , and alternatively the formation of O* and OH*. However, both subsequent decompositions have high energy barriers, 4.00 and 2.99 eV, respectively (Fig. 3A). According to the base-like decomposition pathway, the H2O2 firstly breaks the OeO to give two OH*, which has a smaller energy barrier of 0.57 eV. The subsequent reactions lead to the formation of H2O* and O*, or alternatively the formation of H*, O* and OH*. Fig. 3A apparently shows that the base-like decomposition yielding H2O* and O* is the most favorable decomposition pathway at neutral conditions and that the acid-like decomposition hardly occurs because of its high energy barrier. However, the nucleation of two O* has a high energy barrier of 1.42 eV (see part 5 of SI), which suggests the formation of O2 in neutral conditions are hard to occur. To understand the origin of the pH-dependency in the decomposition reaction of H2O2 on the Au(111) surface, we studied the decomposition pathways with a pre-adsorbed H and OH, which represent the acidic and basic conditions, respectively. With a preadsorbed H, H2O2 still prefers the base-like decomposition as in neutral conditions (see Fig. 3B and part 4 of SI). The pre-adsorbed H can combine with the OH obtained from the H2O2, giving rise to a H2O* and a OH* as products (Eq. (3)).

H2 O2 þ H* 4H2 O* þ OH*

(3)

However, this reaction has an energy barrier of 0.61 eV (Fig. 3B), which is higher than that for the direct decomposition of H2 O2 to 2OH* (0.57 eV). This means the pre-adsorbed H produces little influence on the H2O2 decomposition. In acidic conditions, the H2O2 still prefers the same decomposition pathway as that in neutral conditions, which gives rise to OH* and subsequently H2O* and O* as products. The generated O* is highly oxidizing and will readily

abstract hydrogen atoms from organic substrates. These results suggest that the Au surface will act as a peroxidase in acidic conditions, which catalyzes the oxidation of organic substrates in the presence of H2O2. This agrees with the experimentally observed peroxidase activity of Au materials in acidic conditions [11]. In sharp contrast, our calculations suggested that the preadsorbed OH originated from the basic environment greatly influences the decomposition of H2O2 on the Au surface. As shown in Fig. 3C, with a pre-adsorbed OH, the H2 O2 prefers the acid-like decomposition, passing the decomposed H to the pre-adsorbed OH* to yield HO*2 and H2O* (Eq. (4)). This is sharply different from the case of the acid condition, in which the H2O2 dislikes the acidlike decomposition. Subsequently, the generated HO*2 will pass the H to another H2 O2, leaving a O2 and converting the H2 O2 to H2O* and OH* (Eq. (5)).

H2 O2 þ OH* 4H2 O* þ HO2

(4)

H2 O2 þ HO2 4H2 O* þ OH* þ O2

(5)

The energy barrier of the first step is only 0.13 eV (Fig. 3C) and thus will rapidly occur. The second step will also easily occur because of the small energy barrier, 0.80 eV (Fig. 3C). In the reactions of Eq. (4) and Eq. (5), the H2O2 serves as reducing and oxidizing agents, respectively.

2H2 O2 42H2 O* þ O2

(6)

Eq. (6) is the overall reaction of Eq. (4) and Eq. (5), in which the pre-adsorbed OH formally disappears. Therefore, Eq. (4) and Eq. (5) form a catalytic cycle, through which two H2 O2 molecules are decomposed to two H2O* and one O2 (Eq. (6)). Compared with the preference of base-like decomposition for H2O2 at acid conditions, the pre-adsorbed OH at basic conditions plays a key role in triggering the acid-like decomposition and finally the formation of O2 . The pre-adsorbed OH group on the Au surface is the active site catalyzing the conversion from H2O2 to H2O and O2. This substantializes the observed production of oxygen, namely the catalase activity of Au in basic conditions [11,12]. Unlike the Au(111) surface, which is in general flat, the Au(211) and Au(110) surfaces are composed by repeated steps (Fig. 2) and are thermodynamically less stable. To comprehensively understand the peroxidase and catalase activities of gold, we studied the

Fig. 3. Calculated reaction energy profiles for H2O2 decomposition on the Au(111) surface in neutral (A), acidic (B) and basic (C) conditions. (Unit: eV).

J. Li et al. / Biomaterials 48 (2015) 37e44

adsorptions and decompositions of H2O2 on the Au(211) and Au(110) surfaces in different pH conditions. Especially, the Au(211) surface can be regarded as the structure consisting of three-atomwide terraces of (111) orientation and a monatomic step with a (100) orientation, which thereby models the highly faceted metal NPs and NRs consisting of intersections of different facets [33,34,42]. Fig. 4 illustrates the key structures involved in these reactions. For comparison, Fig. 4 also shows those involved in the reactions on the Au(111) surface. As shown in Fig. 4, the decompositions of H2O2 on the three surfaces have common features under both acidic and basic conditions, although differences also exist (for details see part 6 of SI). In an acidic condition, a H2O2 adsorbs on each of the three surfaces and undergoes the base-like decomposition, which gives rise to a H2O* and a O* with a similar rate-determining transition state (RDTS). In a basic condition, a H2O2 adsorbs in the vicinity of a pre-adsorbed OH on each of the three surfaces and then undergoes the acid-like decomposition, giving rise to a H2O* and a HO*2 . The HO*2 subsequently reacts with another H2O2 to yield a O2 , a OH* and a H2O*, which corresponds to the rate-determining step with a similar RDTS. Table 1 summarizes the corresponding adsorption energies, activation energies of the rate-determining steps and overall reaction energies. According to Table 1, the energy barriers for H2O2 decompositions at either acidic or basic condition are less than 0.65 eV on the Au(211), Au(110) and Au(111) surfaces; Au(111) is the least reactive with the highest barriers. The larger reactivity of (110) facet than that of (111) agrees with the recent experimental studies of oxidation reactions on Au and Pd NPs [39,40,43]. This suggests the pH-switchable peroxidase and catalase activities are intrinsic for Au, regardless of the Au surfaces or intersections of the facets exposed to environments. The robust pHswitchable peroxidase and catalase-like activities of gold are in line with the previous finding that gold is not very structure sensitive as electrocatalyst for the oxidation of CO [44].

3.3. The decomposition reactions of H2O2 on Ag(111), Pd(111) and Pt(111) surfaces in different pH conditions To explore whether the above mechanisms hold for other metals, we studied the decomposition reactions of H2O2 on the

41

Table 1 Adsorption energies (Eads), activation energy barriers of rate-determining steps (Eact) and overall reaction energies (Er) for H2O2 decompositions on metal surfaces at acidic and basic conditions (Unit: eV). Acidic condition H2 O2 ¼ H2 O* þ O*

Au(211) Au(110) Au(111) Ag(111) Pt(111) Pd(111) a b

Basic condition 2H2 O2 ¼ O*2 þ 2H2 O*

Eads,1a

Eact

Er

Eads,2b

Eact

Er

0.15 0.33 0.11 0.16 0.26 0.32

0.33 0.49 0.58 0.41 0.18 0.16

1.78 1.35 1.08 1.83 1.91 2.35

2.69 2.54 2.07 2.37 2.48 2.87

0.24 0.04 0.65 0.49 0.27 0.23

2.28 2.55 1.28 0.43 1.03 0.49

Adsorption energies of H2O2. Co-adsorption energies of H2O2 and OH.

Ag(111), Pd(111) and Pt(111) surfaces at acidic and basic conditions, respectively. The calculations suggest that the H2O2 decomposition mechanisms on these surfaces are similar to that on the Au(111) surface. This can be ascribed to the morphological similarity of these four (111) surfaces (Fig. 2). The energetic data involved in these reactions are given in Table 1 (for details see part 7 of SI). According to Table 1, the activation energies of the ratedetermining steps decrease in the order of Au(111), Ag(111), Pt(111) and Pd(111) at both conditions. Fig. 5A plots the relationships between the adsorption energies and the activation energies. In either condition, the activation energies scale approximately linearly with the adsorption energies on the metal's (111) surface. The adsorption affinity order Au < Pt < Pd for H2O2 on the (111) surfaces is consistent with what has been reported for H2O2 on the (111) surfaces [45,46,39]. Therefore, the pH-switchable enzymatic activities are also intrinsic for Ag, Pd and Pt. Practically, the adsorption energies between H2O2 and metals can be used as convenient descriptors to estimate the relative enzymatic activities of the metals with similar surface structures, as has been suggested for the other catalytic activities of metals [47e50]. To consider the possible influences of solvent on the reactions, we studied the decomposition of H2O2 on the Au (111) surface with H2O molecules (see part 8 of SI). For simplicity, we only considered the reactions at neutral conditions. With n H2O, the H2O2 firstly

Fig. 4. Initial adsorbing structures (IAS), rate-determining transition states (RDTS) and products for H2O2 decompositions on the surfaces of Au(211) (top), Au(110) (middle) and Au(111) (bottom).

42

J. Li et al. / Biomaterials 48 (2015) 37e44

Fig. 5. pH-switchable enzyme-mimic activities of metals. A) Relationships between adsorption energies (Eads) and activation energies (Eact) for H2O2 decompositions on metal's (111) surfaces at acidic (left) and basic (right) conditions. B) TEM images of the NRs. Enzyme-mimic activities of Au@Pd NRs (0.1 nM) at 0.1 M PBS buffers with different pH values for peroxidase with C) 20 mM H2O2 þ 0.4 mM OPD and for catalase with D) 20 mM H2O2, respectively. Peroxidase-like activity at pH ¼ 4.5 E) and catalase-like activity at pH ¼ 7.4 F) for the metals. Unless indicated, reaction temperature in C), D), E) and F) is 30  C. G and H) STEM-HAADF images and STEM-EDX element maps of Au, Pd, and Ag for Au@Ag NRs (G) and Au@Pd NRs (H), respectively.

J. Li et al. / Biomaterials 48 (2015) 37e44

forms the H2O2-(H2O)n cluster on the surface through hydrogen bonds [51,52]. Then, the decomposition occurs in the cluster, in which the H2O serve as protolytes [53,54]. For each n among 1, 2 and 3, the base-like decomposition was preferred, which is the same as that in gas phase. However, the energy barrier varies. For n ¼ 1, 2 and 3, the energy barriers are 0.89, 0.55 and 0.15 eV, respectively (0.57 eV for gas phase). Namely, water will greatly facilitate the reaction, in line with the vital role of moisture found for reactions occurred on Au surfaces [18,55,56]. This suggests that the presence of water molecules alters the decomposition rates of H2O2 but do not change the favorable decomposition pathways. The promotion of reactions on metal surfaces by water has been reported previously [33,34,42,43]. Thus, the main mechanisms obtained from the gas phase calculations are qualitatively valid for the H2O2 decompositions in aqueous conditions. 3.4. Experiments of Ag, Au, Pd and Pt nanoparticles display enzymatic activities in different pH conditions The above calculated results substantialize the enzymatic activities that have been reported for Au, Ag and Pt. The predicted mechanisms have been illustrated in Fig. 6. For example, Au NRs coated with a shell of Pt (Au@Pt NRs) readily catalyze the oxidation of o-phenylenediamine (OPD) and 3,30 ,5,5’-tetramethylbenzidine (TMB) with the presence of H2O2 at acidic conditions. OPD and TMB are typical horseradish peroxidase (HRP) substrates, suggesting the peroxidase-like activity of Au@Pt [10]. Instead, Au@Pt NRs assist the decomposition of H2O2 to release gaseous O2 in basic solutions, suggesting the catalase-like activity of Au@Pt [10]. Such pHswitchable enzymatic activity of Au@Pt agrees well with the calculations, which predict that H2O2 molecules on the Pt surfaces undergo the basic-like (acid-like) decompositions at acidic (basic) conditions, finally yielding the highly oxidative O* (the adsorbed O2) as products. Analogously, the pH-switchable enzymatic activities of Ag NPs [11] and Au NPs [12] can also be interpreted with the calculated results. The calculations predict that Pd has the similar enzyme-mimic functions. Furthermore, the activities of Pd should be comparable to those of Pt and even larger than those of Au and Ag (Fig. 5A). To verify this prediction, we synthesized Au@Pd NRs and measured their enzyme functions at different pH values (see part 2 of SI). For comparison, NRs of Au, Au@Ag and Au@Pt were also synthesized and tested. In the experiment, the peroxidase-like activity positively correlates with the absorbance at the oxidized OPD (OPDox) peak for OPD as reduction substrate; the catalase-like activity negatively correlates with the absorbance of H2O2 at 240 nm (A240 nm) (see part 10 of SI). Au@Pd has the largest peroxidase-like activity at pH 4.5 (Fig. 5C). In sharp contrast, it has the largest catalase-like activity at pH 11 (Fig. 5D). Both activities rapidly decrease and increase in the order of pH 4.5, 7.4, 9.0 and 11.0, which

43

well agrees with the predicted properties for Pd in acidic and basic conditions, respectively. Noteworthily, at a strong acidic condition with pH 2, the expected peroxidase-like activity of Au@Pd NRs was quite low (Fig. 5C). This is ascribed to that the oxidation of OPD does not favor a strong acidic environment [10]. In Fig. 5E and F, the activities of Au@Pd are compared with those of Au, Au@Ag, and Au@Pt. Both activities of Au@Pd and Au@Pt are larger than those of Au and Au@Ag, in agreement with the prediction. The larger activity of Pt than Au has been reported [18]. Seemingly different from the prediction, Au@Pd and Au@Ag are less active than Au@Pt and Au at 30  C respectively. However, the larger activities of Au@Pt are due to the larger surface exposure of the Pt shell. Pt prefers the island-growth mode on the Au rods whereas Ag and Pd could grow epitaxially on the Au rods (Fig. 5B). The coreeshell structure is verified by STEM-EDX element maps for Au@Ag and Au@Pd NRs (Fig. 5G and H). Au@Ag is less active than Au possibly due to the partial surface passivation by Ag oxidation. Indeed, at pH 7.4, Ag is gradually etched away by H2O2 from Au as seen from the red-shift of longitudinal surface plasmon peak (Fig. S11C). Therefore, considering the differences in the surface morphologies of the samples, the experimental results are in good agreement with and thus strongly verify the computational predictions. 4. Conclusion In summary, the peroxidase-like activities of metals (Au, Ag, Pd and Pt) manifested in low-pH conditions are ascribed to the baselike decompositions of H2O2 on the metal surfaces, and the catalase-like activities manifested in high-pH conditions are ascribed to the acid-like decompositions of H2O2 on the surfaces (Fig. 6). The pre-adsorbed OH groups on the surfaces, which are only favorably formed in basic conditions, trigger the switch between both activities and render the pH-switchability. The enzymatic activities are intrinsic properties of the metals, regardless of the surfaces or intersections of the facets exposed to environments. The adsorption energies between H2O2 and metals can be used as convenient descriptors to predict the relative enzymatic activities of the metals with similar surface morphologies. The results will guide the de novo design, synthesis and application of artificial enzymes based on metals. Acknowledgments This work was supported by the CAS Hundreds Elite Program, NSFC Project (21373226) and MOST 973 program (2012CB934001, 2011CB933400, 2011CB933101). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.biomaterials.2015.01.012. References

Fig. 6. The predicted mechanisms for the pH-switchable peroxidase-like and catalaselike activities of the metals.

[1] Lin Y, Ren J, Qu X. Catalytically active nanomaterials: a promising candidate for artificial enzymes. Acc Chem Res 2014;47:1097e105. [2] Gao L, Zhuang J, Nie L, Zhang J, Zhang Y, Gu N, et al. Intrinsic peroxidase-like activity of ferromagnetic nanoparticles. Nat Nanotechnol 2007;2:577e83. [3] Chen Z, Yin J-J, Zhou Y-T, Zhang Y, Song L, Song M, et al. Dual enzyme-like activities of iron oxide nanoparticles and their implication for diminishing cytotoxicity. Acs Nano 2012;6:4001e12. [4] Dai Z, Liu S, Bao J, Ju H. Nanostructured FeS as a mimic peroxidase for biocatalysis and biosensing. Chemistry 2009;15:4321e6.  R, Nata lio F, Humanes M, Leppin J, Heinze K, Wever R, et al. V2O5 [5] Andre nanowires with an intrinsic peroxidase-like activity. Adv Funct Mater 2011;21:501e9. [6] Song Y, Qu K, Zhao C, Ren J, Qu XG. Graphene oxide: intrinsic peroxidase catalytic activity and its application to glucose detection. Adv Mater 2010;22:2206e10.

44

J. Li et al. / Biomaterials 48 (2015) 37e44

[7] Song YJ, Wang XH, Zhao C, Qu KG, Ren JS, Qu XG. Label-Free colorimetric detection of single nucleotide polymorphism by using single-walled carbon nanotube intrinsic peroxidase-like activity. Chem Eur J 2010;16:3617e21. [8] Pirmohamed T, Dowding JM, Singh S, Wasserman B, Heckert E, Karakoti AS, et al. Nanoceria exhibit redox state-dependent catalase mimetic activity. Chem Commun (Camb) 2010;46:2736e8. [9] He W, Wu X, Liu J, Hu X, Zhang K, Hou S, et al. Design of AgM Bimetallic alloy nanostructures (M¼Au, Pd, Pt) with tunable morphology and peroxidase-like activity. Chem Mater 2010;22:2988e94. [10] He W, Liu Y, Yuan J, Yin JJ, Wu X, Hu X, et al. Au@Pt nanostructures as oxidase and peroxidase mimetics for use in immunoassays. Biomaterials 2011;32: 1139e47. [11] He W, Zhou YT, Wamer WG, Boudreau MD, Yin JJ. Mechanisms of the pH dependent generation of hydroxyl radicals and oxygen induced by Ag nanoparticles. Biomaterials 2012;33:7547e55. [12] He W, Zhou YT, Wamer WG, Hu X, Wu X, Zheng Z, et al. Intrinsic catalytic activity of Au nanoparticles with respect to hydrogen peroxide decomposition and superoxide scavenging. Biomaterials 2013;34:765e73. [13] Hu X, Saran A, Hou S, Wen T, Ji Y, Liu W, et al. Au@PtAg core/shell nanorods: tailoring enzyme-like activities via alloying. RSC Adv 2013;3:6095. [14] Liu J, Hu X, Hou S, Wen T, Liu W, Zhu X, et al. Au@Pt core/shell nanorods with peroxidase and ascorbate oxidase-like activities for improved detection of glucose. Sensors Actuat B: Chem 2012;166e167:708e14. [15] Wang XX, Wu Q, Shan Z, Huang QM. BSA-stabilized Au clusters as peroxidase mimetics for use in xanthine detection. Biosens Bioelectron 2011;26:3614e9. [16] Jv Y, Li B, Cao R. Positively-charged gold nanoparticles as peroxidase mimic and their application in hydrogen peroxide and glucose detection. Chem Commun (Camb) 2010;46:8017e9. [17] Slocik JM, Govorov AO, Naik RR. Photoactivated biotemplated nanoparticles as an enzyme mimic. Angew Chem Int Ed Engl 2008;47:5335e9. [18] Chen T, Hu Y, Cen Y, Chu X, Lu Y. A dual-emission fluorescent nanocomplex of gold-cluster-decorated silica particles for live cell imaging of highly reactive oxygen species. J Am Chem Soc 2013;135:11595e602. [19] Zope BN, Hibbitts DD, Neurock M, Davis RJ. Reactivity of the gold/water interface during selective oxidation catalysis. Science 2010;330:74e8. [20] Pal R, Wang LM, Pei Y, Wang LS, Zeng XC. Unraveling the mechanisms of O2 activation by size-selected gold clusters: transition from superoxo to peroxo chemisorption. J Am Chem Soc 2012;134:9438e45. [21] Yang B, Gong XQ, Wang HF, Cao XM, Rooney JJ, Hu P. Evidence to challenge the universality of the Horiuti-Polanyi mechanism for hydrogenation in heterogeneous catalysis: origin and trend of the preference of a non-HoriutiPolanyi mechanism. J Am Chem Soc 2013;135:15244e50. [22] Liu C, Tan Y, Lin S, Li H, Wu X, Li L, et al. CO self-promoting oxidation on nanosized gold clusters: triangular Au3 active site and CO induced O-O scission. J Am Chem Soc 2013;135:2583e95. [23] Methfessel M, Paxton A. High-precision sampling for Brillouin-zone integration in metals. Phys Chem B 1989;40:3616e21. €chl PE. Improved tetrahedron method for Brillouin-zone integrations. Phys [24] Blo Chem B 1994;49:16223e33. [25] Kresse G, Furthmüller J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys Chem B 1996;54:11169. [26] Perdew JP, Burke K, Ernzerhof M. Generalized gradient approximation made simple. Phys Chem Lett 1996;77:3865e8. [27] Sandy A, Mochrie S, Zehner D, Huang K, Gibbs D. Structure and phases of the Au(111) surface: X-ray-scattering measurements. Phys Rev B 1991;43: 4667e87. [28] Wan J, Fan YL, Gong DW, Shen SG, Fan XQ. Surface relaxation and stress of fcc metals: Cu, Ag, Au, Ni, Pd, Pt, Al and Pb. Model Simul Mater Sci Eng 1999;7: 189e206. [29] Henkelman G, Uberuaga BP, Jonsson H. A climbing image nudged elastic band method for finding saddle points and minimum energy paths. J Chem Phys 2000;113:9901e4. [30] Sheppard D, Terrell R, Henkelman G. Optimization methods for finding minimum energy paths. J Chem Phys 2008:128. [31] Henkelman G, Jonsson H. A dimer method for finding saddle points on high dimensional potential surfaces using only first derivatives. J Chem Phys 1999;111:7010e22.

[32] Henderson MA. The interaction of water with solid surfaces: fundamental aspects revisited. Surf Sci Rep 2002;46:1e308. [33] Gao Y, Zeng XC. Water-promoted O-2 dissociation on small-sized anionic gold clusters. Acs Catal 2012;2:2614e21. [34] Kulkarni AD, Pathak RK, Bartolotti LJ. Structures, energetics, and vibrational spectra of H2O2 center dot center dot center dot(H2O)(n), n¼1-6 clusters: ab initio quantum chemical investigations. J Phys Chem A 2005;109:4583e90. [35] Saha K, Agasti SS, Kim C, Li X, Rotello VM. Gold nanoparticles in chemical and biological sensing. Chem Rev 2012;112:2739e79. [36] Meng S, Wang E, Gao S. Water adsorption on metal surfaces: a general picture from density functional theory studies. Phys Rev B 2004;69:195404. [37] Zhou YH, Fuentes-Hernandez C, Shim J, Meyer J, Giordano AJ, Li H, et al. A universal method to produce low-work function electrodes for organic electronics. Science 2012;336:327e32. [38] Kwon Y, Lai SCS, Rodriguez P, Koper MTM. Electrocatalytic oxidation of alcohols on gold in alkaline media: base or gold catalysis? J Am Chem Soc 2011;133:6914e7. [39] Rodriguez P, Kwon Y, Koper MTM. The promoting effect of adsorbed carbon monoxide on the oxidation of alcohols on a gold catalyst. Nat Chem 2012;4: 177e82. [40] Koverga AA, Frank S, Koper MTM. Density functional theory study of electric field effects on CO and OH adsorption and co-adsorption on gold surfaces. Electrochim Acta 2013;101:244e53. [41] Zhang TF, Liu ZP, Driver SM, Pratt SJ, Jenkins SJ, King DA. Stabilizing CO on Au with NO2: electronegative species as promoters on coinage metals? Phys Rev Lett 2005:95. [42] Buszek RJ, Torrent-Sucarrat M, Anglada JM, Francisco JS. Effects of a single water molecule on the OH þ H2O2 reaction. J Phys Chem A 2012;116:5821e9. [43] Ojifinni RA, Gong J, Froemming NS, Flaherty DW, Pan M, Henkelman G, et al. Carbonate formation and decomposition on atomic oxygen precovered Au(111). J Am Chem Soc 2008;130:6801e12. [44] Kim J, Samano E, Koel BE. CO adsorption and reaction on clean and oxygencovered Au(211) surfaces. J Phys Chem B 2006;110:17512e7. [45] Blakely DW, Somorjai GA. The stability and structure of high miller index platinum crystal surfaces in vacuum and in the presence of adsorbed carbon and oxygen. Surf Sci 1977;65:419e42. [46] Long R, Mao K, Ye X, Yan W, Huang Y, Wang J, et al. Surface facet of palladium nanocrystals: a key parameter to the activation of molecular oxygen for organic catalysis and cancer treatment. J Am Chem Soc 2013;135:3200e7. [47] Hou S, Hu X, Wen T, Liu W, Wu XC. Core-shell noble metal nanostructures templated by gold nanorods. Adv Mater 2013;25:3857e62. [48] Oo MKK, Yang YM, Hu Y, Gomez M, Du H, Wang HJ. Gold nanoparticleenhanced and size-dependent generation of reactive oxygen species from protoporphyrin IX. Acs Nano 2012;6:1939e47. [49] Wang X, Orikasa Y, Takesue Y, Inoue H, Nakamura M, Minato T, et al. Quantitating the lattice strain dependence of monolayer Pt shell activity toward oxygen reduction. J Am Chem Soc 2013;135:5938e41. [50] Siahrostami S, Verdaguer-Casadevall A, Karamad M, Deiana D, Malacrida P, Wickman B, et al. Enabling direct H2O2 production through rational electrocatalyst design. Nat Mater 2013;12:1137e43. [51] Robertson WH, Diken EG, Price EA, Shin JW, Johnson MA. Spectroscopic determination of the OH-solvation shell in the OH-(H2O)n clusters. Science 2003;299:1367e72. [52] Shin JW, Hammer NI, Diken EG, Johnson MA, Walters RS, Jaeger TD, et al. Infrared signature of structures associated with the Hþ(H2O)n (n ¼ 6 to 27) clusters. Science 2004;304:1137e40. [53] Shang C, Liu ZP. Origin and activity of gold nanoparticles as aerobic oxidation catalysts in aqueous solution. J Am Chem Soc 2011;133:9938e47. [54] Wei GF, Fang YH, Liu ZP. First principles Tafel kinetics for resolving key parameters in optimizing oxygen electrocatalytic reduction catalyst. J Phys Chem C 2012;116:12696e705.  M, Okumura M, Tsubota S, Haruta M. Vital role of moisture in the cat[55] Date alytic activity of supported gold nanoparticles. Angew Chem Int Ed 2004;43: 2129e32. [56] Sanchez-Castillo MA, Couto C, Kim WB, Dumesic J. Gold-nanotube membranes for the oxidation of CO at gasewater interfaces. Angew Chem Int Ed 2004;43: 1140e2.

Mechanism of pH-switchable peroxidase and catalase-like activities of gold, silver, platinum and palladium.

Despite being increasingly used as artificial enzymes, little has been known for the origin of the pH-switchable peroxidase-like and catalase-like act...
2MB Sizes 0 Downloads 8 Views