Melamine Foams Decorated with In-Situ Synthesized Gold and Palladium Nanoparticles

A versatile and straightforward route to produce polymer foams with functional surface through their decoration with gold and palladium nanoparticles is proposed. Melamine foams, used as polymeric porous substrates, are first covered with a uniform coating of polydimethylsiloxane, thin enough to assure the preservation of their original porous structure. The polydimethylsiloxane layer allows the facile in-situ formation of metallic Au and Pd nanoparticles with sizes of tens of nanometers directly on the surface of the struts of the foam by the direct immersion of the foams into gold or palladium precursor solutions. The effect of the gold and palladium precursor concentration, as well as the reaction time with the foams, to the amount and sizes of the nanoparticles synthesized on the foams, was studied and the ideal conditions for an optimized functionalization were defined. Gold and palladium contents of about 1 wt.% were achieved, while the nanoparticles were proven to be stably adhered to the foam, avoiding potential risks related to their accidental release.

However, the direct use of free-standing NPs raises some concerns, mainly related to their efficiency, safety, and recovery [15,23,24]. On the one hand, free-standing NPs usually tend to aggregate, a fact that causes the decrease of their active surface area and the modification of their key physical properties (e.g., photocatalysis), therefore, their performance in some applications (e.g., water remediation) is compromised [15,23,24]. On the other hand, after their use on water treatment, sensoring, or catalysis procedures, time-consuming and complicated actions need to be adopted for their recovery, as their accidental release to the environment could be toxic for several organisms [25].

Functionalization of the ME Foams
Samples (2 × 1 × 1 cm 3 ) of the pristine ME foams were rinsed in ethanol, dried, and then immersed for 3 min into ethyl acetate solutions containing 1 v.% of the PDMS base and the curing agent in a weight ratio 10:1. Subsequently, the samples were extracted from the solution and cured for 3 h at 80 • C (these processing parameters were previously optimized, see Supplementary Information, Figure S1). Then, the obtained ME/PDMS foams were rinsed with ethanol to remove any unreacted PDMS. The in-situ synthesis of Au or Pd NPs on the ME/PDMS foams (2 × 1 × 1 cm 3 ) was achieved by immersing them on 20 mL ethanol solutions of HAuCl 4 or Na 2 PdCl 4 , respectively. Four different concentrations of each precursor were employed: 0.175, 0.350, 0.700, and 1.400 mg/mL of HAuCl 4 and 0.250, 0.500, 1.000, and 2.000 mg/mL of Na 2 PdCl 4 . These concentrations were determined to ensure a comparable availability of Au and Pd ions in the solutions. The immersed ME/PDMS samples were kept under shaking for different reaction times (ranging from 1 to 120 h). Then, the samples were extracted and subjected to five washing cycles. In each washing cycle, the samples were immersed in 20 mL of distilled water for 2 min under shaking. This washing procedure was required to ensure the complete removal of ethanol or NPs unreacted precursor from the foams, as well as of non-well attached NPs (see Supplementary Information, Figure S2).

Experimental Techniques
The porous structure of the ME foams, the surface and thickness of their struts, and the presence and size of Au and Pd NPs were studied by High-Resolution Scanning Electron Microscopy (HRSEM) using a JEOL JSM-7500La (Jeol, Tokyo, Japan) equipped with a cold field-emission gun (FEG), operating at 15 kV acceleration voltage. The thickness of the struts was measured at the intermediate point between vertices. The thickness distribution of the struts was determined from several micrographs of each foam, by analyzing at least 100 struts in each case using FIJI/ImageJ. [38] Micrographs were obtained using backscattered electrons, whereas energy-dispersive spectroscopy (EDS) was employed to study the distribution of the Au and Pd elements on the struts of the foams. The size distribution, average size, and standard deviation of the formed Au and Pd NPs for each preparation conditions were determined from several micrographs for each foam, by analyzing at least 100 NPs in each case using FIJI/ImageJ. [38] The PDMS content of the ME/PDMS foams was determined by weighing the samples before and after the formation of the PDMS coating. Moreover, the Au and Pd NPs content of the ME/PDMS/Au and ME/PDMS/Pd foams were evaluated using Inductive Couple Plasma-Optical Emission Spectroscopy (ICP-OES, iCAP 3600 spectrometer, Thermo Fisher Scientific, Waltham, MA, USA). A microwave digestion system (MARS Xpress, CEM, Matthews, NC, US) was employed to digest 5 mg of foams in 2.5 mL of nitric acid (70%, Sigma Aldrich, St. Louis, MO, US). The solid degradation reaction was performed at 180 • C for 15 min. Then, the samples were diluted in Milli-Q water up to Polymers 2020, 12, 934 4 of 13 25 mL and filtered through polytetrafluoroethylene (PTFE) syringe filters (15 mm, pore size 0.45 µm, Sartorius, Göttingen, Germany). The surface chemistry of the ME/PDMS/Au and ME/PDMS/Pd foams was studied by X-ray Photoelectron Spectroscopy (XPS) using an Axis Ultra DLD spectrometer from Kratos®(Manchester, UK) under 10 −9 mbar pressure and equipped with a monochromatic Al Kα source (photon energy = 1486.6 eV, emission current = 20 mA, and operation voltage = 15 kV). High-resolution spectra of these materials were obtained with a step of 0.1 eV and an analyzer pass energy of 10 eV. Surface charging was neutralized with low-energy electrons (4 eV), and the energy calibration was carried out by setting the C-C/C-H component of the C 1 s spectrum to a fixed binding energy value of 284.5 eV. Data analysis was performed with CasaXPS software.

ME/PDMS Foams
The efficiency of the proposed procedure to create a PDMS coating over the struts of the ME foams was studied. First, it was determined that the amount of PDMS transferred to the foam was 104 ± 13 wt.%, from which a stable PDMS coating of about 97 ± 13 wt.% remained after the ethanol rinsing. As previously proved in other polymer foams, the PDMS layer is expected to establish H-bonds with the polymer foam substrate [39]. This amount of PDMS was found to be enough to provide a thin homogeneous layer covering the struts of the ME foams, without inducing a significant thickness increase or clogging of the porous structure (Figure 1a,b). From the morphological analysis of the foams, the PDMS layer is expected to have a thickness below a micron, as no significant difference was found in terms of the struts thicknesses distribution (Figure 1a,b) or they average size (average values of 5.8 ± 2.3 and 5.3 ± 1.8 µm respectively, for ME and ME/PDMS foams). The EDS analysis of the surface of the struts shows that the Si signal, representative of the PDMS coating, is homogeneously distributed along the struts of the ME/PDMS foams (Figure 1c,d). From the corresponding EDS spectra of the foams, clear differences are shown, as in the case of ME/PDMS foams, the Si signal is present together with a significantly lower signal of the N of the ME substrate ( Figure 1e). Therefore, with this step, a stable PDMS layer homogeneously distributed on the ME foams is successfully formed, preserving the high surface area and porosity of the ME and providing a platform for the in-situ synthesis of noble metal nanoparticles.

ME/PDMS Foams
The efficiency of the proposed procedure to create a PDMS coating over the struts of the ME foams was studied. First, it was determined that the amount of PDMS transferred to the foam was 104 ± 13 wt.%, from which a stable PDMS coating of about 97 ± 13 wt.% remained after the ethanol rinsing. As previously proved in other polymer foams, the PDMS layer is expected to establish Hbonds with the polymer foam substrate [39]. This amount of PDMS was found to be enough to provide a thin homogeneous layer covering the struts of the ME foams, without inducing a significant thickness increase or clogging of the porous structure (Figure 1a,b). From the morphological analysis of the foams, the PDMS layer is expected to have a thickness below a micron, as no significant difference was found in terms of the struts thicknesses distribution (Figure 1a,b) or they average size (average values of 5.8 ± 2.3 and 5.3 ± 1.8 µ m respectively, for ME and ME/PDMS foams). The EDS analysis of the surface of the struts shows that the Si signal, representative of the PDMS coating, is homogeneously distributed along the struts of the ME/PDMS foams (Figure 1c,d). From the corresponding EDS spectra of the foams, clear differences are shown, as in the case of ME/PDMS foams, the Si signal is present together with a significantly lower signal of the N of the ME substrate ( Figure 1e). Therefore, with this step, a stable PDMS layer homogeneously distributed on the ME foams is successfully formed, preserving the high surface area and porosity of the ME and providing a platform for the in-situ synthesis of noble metal nanoparticles.

ME/PDMS/Au and ME/PDMS/Pd Foams
The treatment of ME/PDMS foams with the HAuCl4 solutions induced an ev color of the foams, from the light gray of ME/PDMS (Figure 2a,b) to red (Fi indication of the presence of Au NPs [33][34][35]. As shown by the photographs ( optical micrographs (Figure 2d), the coloration of the foams is homogeneous, wh that the expected presence of Au NPs is throughout the porous structure of the f hand, the treatment of the ME/PDMS foams with Na2PdCl4 solutions also induc with the ME/PDMS/Pd foams showing a uniform dark grey-blackish color (Figure presence of Pd NPs [36]. showing a homogeneous distribution of the Si corresponding to the PDMS (red). EDS spectra of ME and ME/PDMS foams (e).

ME/PDMS/Au and ME/PDMS/Pd Foams
The treatment of ME/PDMS foams with the HAuCl 4 solutions induced an evident change in the color of the foams, from the light gray of ME/PDMS (Figure 2a,b) to red (Figure 2c,d), a clear indication of the presence of Au NPs [33][34][35]. As shown by the photographs (Figure 2c) and the optical micrographs (Figure 2d), the coloration of the foams is homogeneous, which is an indication that the expected presence of Au NPs is throughout the porous structure of the foams. On the other hand, the treatment of the ME/PDMS foams with Na 2 PdCl 4 solutions also induced a color change, with the ME/PDMS/Pd foams showing a uniform dark grey-blackish color (Figure 2e,f), related to the presence of Pd NPs [36]. showing a homogeneous distribution of the Si corresponding to the PDMS (red). EDS spectra of ME and ME/PDMS foams (e).

ME/PDMS/Au and ME/PDMS/Pd Foams
The treatment of ME/PDMS foams with the HAuCl4 solutions induced an evident change in the color of the foams, from the light gray of ME/PDMS (Figure 2a No significant weight changes were found during the production of the ME/PDMS/Au and the ME/PDMS/Pd foams from the ME/PDMS foams, so moderate Au or Pd loads are expected on these samples. This amount of Au or Pd transferred to the foams, upon dipping to the precursor solutions of different concentrations and for different time intervals, was accurately determined by the ICP-OES analysis of the ME/PDMS/Au and ME/PDMS/Pd foams. After 48 h of immersion of the foams in solutions with HAuCl4 concentrations ranging from 0.175 to 1.400 mg/mL, the transferred Au to the foams was ranging from c.a. 0.6 to 1.0 wt.%. For the precursor concentration of 0.350 mg/mL, the Au transfer already reached c.a. 1 wt.%, and this concentration was chosen for the further analysis of the kinetics of the Au transfer onto the foams. Similarly, ME/PDMS/Pd foams obtained after 48 h of immersion in solutions with Na2PdCl4 concentrations ranging from 0.250 to 2.000 mg/mL showed Pd contents ranging from 0.8 to 1.2 wt.%. Additionally, an intermediate concentration of 0.500 mg/mL, which provides a Pd transfer c.a. 1.1 wt.%, was chosen to study the kinetics of the procedure. It should be noticed that this concentration is also equivalent to the selected Au concentration (i.e., 0.350 mg/mL) in terms of the noble metal ions availability, about 0.200 mg/mL. Thus, a direct comparison between the ME/PDMS/Au and ME/PDMS/Pd foams can be performed.
In particular, ME/PDMS/Au and ME/PDMS/Pd foams were produced upon the dipping of the ME/PDMS foams in the HAuCl4 or Na2PdCl4 solutions for time intervals ranging from 1 to 120 h. As shown in Figure  In particular, ME/PDMS/Au and ME/PDMS/Pd foams were produced upon the dipping of the ME/PDMS foams in the HAuCl 4 or Na 2 PdCl 4 solutions for time intervals ranging from 1 to 120 h. As shown in Figure 3, both the Au and Pd amounts transferred to the foam increased with the reaction time increase, starting from about 0.1 wt.% Au and 0.5 wt.% Pd after 1 h of reaction time and reaching a stable value of about 1 wt.% Au and 1.1 wt.% Pd after 48 h. According to these values, the Pd transfer seems to be faster in the beginning, reaching almost 50% of the final load in just 1 h, while the Au transfer reaches about 10% at the same time. However, after 15 h, the transferred amounts of both noble metals became similar. In particular, ME/PDMS/Au and ME/PDMS/Pd foams were produced upon the dipping of the ME/PDMS foams in the HAuCl4 or Na2PdCl4 solutions for time intervals ranging from 1 to 120 h. As shown in Figure 3, both the Au and Pd amounts transferred to the foam increased with the reaction time increase, starting from about 0.1 wt.% Au and 0.5 wt.% Pd after 1 h of reaction time and reaching a stable value of about 1 wt.% Au and 1.1 wt.% Pd after 48 h. According to these values, the Pd transfer seems to be faster in the beginning, reaching almost 50% of the final load in just 1 h, while the Au transfer reaches about 10% at the same time. However, after 15 h, the transferred amounts of both noble metals became similar. Previous studies have shown that the functional components of the PDMS are inducing the formation of noble metal NPs on their surface upon chemical reduction of the metallic precursors adsorbed. In particular, the noble metal nanoparticles formation is expected to happen in non-reacted cross-linking Si-H sites [10,30,34,35]. In order to prove, in the present case, the transformation of the adsorbed Au or Pd ions into NPs onto the surface of the foams in the presence of PDMS, detailed HRSEM and EDS studies were performed. As shown in Figure 4, both ME/PDMS/Au and ME/PDMS/Pd foams present small NPs and aggregates on the surface of their struts. Moreover, EDS mapping of the struts of the ME/PDMS/Au and ME/PDMS/Pd foams confirmed that both the small NPs and bigger aggregates are attributed to the presence of Au or Pd NPs (Figure 4c,d, respectively). The presence of Au or Pd NPs in their crystalline form was further confirmed by X-ray Diffraction analysis in all cases (XRD, see Supplementary Information, Figure S3). It should be mentioned that the presence of Au or Pd NPs is strictly related to the PDMS layer, as the EDS elemental maps obtained by HRSEM analysis of ME and ME/PDMS foams proved. In fact, the NPs are present only on the latter foams (see Supplementary Information, Figure S4). In contrast, pure ME foams subjected to the same dipping procedure presented no modifications on their surface.  Previous studies have shown that the functional components of the PDMS are inducing the formation of noble metal NPs on their surface upon chemical reduction of the metallic precursors adsorbed. In particular, the noble metal nanoparticles formation is expected to happen in non-reacted cross-linking Si-H sites [10,30,34,35]. In order to prove, in the present case, the transformation of the adsorbed Au or Pd ions into NPs onto the surface of the foams in the presence of PDMS, detailed HRSEM and EDS studies were performed. As shown in Figure 4, both ME/PDMS/Au and ME/PDMS/Pd foams present small NPs and aggregates on the surface of their struts. Moreover, EDS mapping of the struts of the ME/PDMS/Au and ME/PDMS/Pd foams confirmed that both the small NPs and bigger aggregates are attributed to the presence of Au or Pd NPs (Figure 4c,d, respectively). The presence of Au or Pd NPs in their crystalline form was further confirmed by X-ray Diffraction analysis in all cases (XRD, see Supplementary Information, Figure S3). It should be mentioned that the presence of Au or Pd NPs is strictly related to the PDMS layer, as the EDS elemental maps obtained by HRSEM analysis of ME and ME/PDMS foams proved. In fact, the NPs are present only on the latter foams (see Supplementary Information, Figure S4). In contrast, pure ME foams  Au NPs synthesized and not to their size increase. Longer reaction times provided a slightly larger average size of about 25 nm and a broadening of the size distribution with a noticeable rise in the presence of NPs over 30 nm ( Figure 5). Therefore, taking into account the results of the amount of Au transferred and the NPs sizes the optimal conditions to produce the ME/PDMS/Au foams are a HAuCl 4 concentration of 0.350 mg/mL and 48 h of reaction time, obtaining 1 wt.% of Au NPs with sizes about 18 nm synthesized on the foams. The achieved amount of Au NPs synthesized on the foams is comparable to the results obtained by Apyari et al. [33], about 0.9 wt.%, and significantly higher than those of Gupta and Kulkarni [35] (0.06 wt.%).
Polymers 2020, 12, x FOR PEER REVIEW 8 of 13 NPs ranging from 10 to 30 nm were found for reaction times between 1 and 48 h (Figures 5 and S6, see Supplementary Information). This result indicates that the significant increase of Au transferred to the foams occurring in that range, from 0.1 to 1 wt.% (Figure 3), is related to the rise of the number of Au NPs synthesized and not to their size increase. Longer reaction times provided a slightly larger average size of about 25 nm and a broadening of the size distribution with a noticeable rise in the presence of NPs over 30 nm ( Figure 5). Therefore, taking into account the results of the amount of Au transferred and the NPs sizes the optimal conditions to produce the ME/PDMS/Au foams are a HAuCl4 concentration of 0.350 mg/mL and 48 h of reaction time, obtaining 1 wt.% of Au NPs with sizes about 18 nm synthesized on the foams. The achieved amount of Au NPs synthesized on the foams is comparable to the results obtained by Apyari et al. [33], about 0.9 wt.%, and significantly On the contrary, the reaction time between the ME/PDMS foams and the 0.500 mg/mL Pd precursor solutions present a stronger influence on the Pd NPs size than in the case of Au NPs. Reaction times from 1 to 24 h provided NPs average sizes about 27 to 30 nm (Figures 6 and S7, see Supplementary Information), while longer times (48 and 120 h) produced Pd NPs with average sizes about 42 to 45 nm (Figures 6 and S7, see Supplementary Information). In all the cases, the NPs size distributions present similar widths, with no clear relationship with the reaction time. Accordingly, in this case, it is possible to obtain ME/PDMS/Pd foams with similar Pd loads of about 1.0 to 1.1 wt.% ( Figure 3) and different particle sizes, about 30 or 45 nm, by using reaction times of 24 or 48 h, respectively. The obtained results, in terms of the amount of Pd transferred to the foams, are lower than previous results reported in the literature using porous polymeric substrates (6.7 wt.% using Poly(styrene/DVB)/PolyHIPE and 3 wt.% using Polysulfone membranes) [36,37]. However, these previous approaches cannot be applied to commercially available substrates or ensure the presence of the Pd NPs on the surface of the struts of the porous substrates. On the contrary, the reaction time between the ME/PDMS foams and the 0.500 mg/mL Pd precursor solutions present a stronger influence on the Pd NPs size than in the case of Au NPs. Reaction times from 1 to 24 h provided NPs average sizes about 27 to 30 nm ( Figure 6 and Figure S7, see Supplementary Information), while longer times (48 and 120 h) produced Pd NPs with average sizes about 42 to 45 nm ( Figure 6 and Figure S7, see Supplementary Information). In all the cases, the NPs size distributions present similar widths, with no clear relationship with the reaction time. Accordingly, in this case, it is possible to obtain ME/PDMS/Pd foams with similar Pd loads of about 1.0 to 1.1 wt.% (Figure 3) and different particle sizes, about 30 or 45 nm, by using reaction times of 24 or 48 h, respectively. The obtained results, in terms of the amount of Pd transferred to the foams, are lower than previous results reported in the literature using porous polymeric substrates (6.7 wt.% using Poly(styrene/DVB)/PolyHIPE and 3 wt.% using Polysulfone membranes) [36,37]. However, these previous approaches cannot be applied to commercially available substrates or ensure the presence of the Pd NPs on the surface of the struts of the porous substrates. In addition, the proposed in-situ synthesis procedure ensures not only the presence of the Au or Pd NPs on the surface of the struts of the foams but also their stable anchoring, as the obtained nanocomposite foams do not release the NPs even if they are subjected to shaking in water during 24 h (see Supplementary Information, Figures S2 and S6). Therefore, these foams could be safely employed with any aim without risks of accidental release of the NPs to the environment.

XPS Study of the ME/PDMS/Au and ME/PDMS/Pd Foams
The surface of the obtained ME/PDMS/Au and ME/PDMS/Pd foams was further analyzed by XPS (Figure 7). After the treatment, the wide scan spectra confirm the presence of PDMS and of the metals on the surface of their struts for both foams. In fact, as shown in Figure 7a,b, the Si 2s and Si 2p peaks appear in both cases as well as the Au 4f and Pd 3d peaks for each type of foam. Highresolution XPS spectra of the Au and Pd peaks was performed to determine whether the Au and Pd signals come from metallic NPs or unreacted residual precursors (Figure 7c,d). On the one hand, the Au 4f spectrum of the ME/PDMS/Au foams was accurately fitted with the peaks corresponding to 4f7/2 and 4f5/2 of metallic gold (Au 0 ), with binding energies of 84.0 and 87.7 eV, proving the metallic character of the obtained Au NPs [40]. No residues from the Au precursor (i.e., Cl signal) were found on the foams (see Supplementary Information, Figure S8). On the other hand, the characteristic Pd 3d5/2 and Pd 3d3/2 peaks of metallic Pd particles appeared respectively about 335.5-335.9 and 340.8-341.2 eV. The obtained values seem to be slightly shifted from the position of metallic Pd (335.2 and 340.5 eV) [36]. This shift of the energies of the peaks could be related to the presence of not only metallic Pd, but also a small amount of some Pd compound. The potential presence of residues of the Pd precursor, Na2PdCl4, can be discarded as no Cl residues from the Pd precursor were found on the foams by XPS analysis (see Supplementary Information, Figure S8). On the contrary, according to In addition, the proposed in-situ synthesis procedure ensures not only the presence of the Au or Pd NPs on the surface of the struts of the foams but also their stable anchoring, as the obtained nanocomposite foams do not release the NPs even if they are subjected to shaking in water during 24 h (see Supplementary Information, Figures S2 and S6). Therefore, these foams could be safely employed with any aim without risks of accidental release of the NPs to the environment.

XPS Study of the ME/PDMS/Au and ME/PDMS/Pd Foams
The surface of the obtained ME/PDMS/Au and ME/PDMS/Pd foams was further analyzed by XPS (Figure 7). After the treatment, the wide scan spectra confirm the presence of PDMS and of the metals on the surface of their struts for both foams. In fact, as shown in Figure 7a,b, the Si 2s and Si 2p peaks appear in both cases as well as the Au 4f and Pd 3d peaks for each type of foam. High-resolution XPS spectra of the Au and Pd peaks was performed to determine whether the Au and Pd signals come from metallic NPs or unreacted residual precursors (Figure 7c,d). On the one hand, the Au 4f spectrum of the ME/PDMS/Au foams was accurately fitted with the peaks corresponding to 4f 7/2 and 4f 5/2 of metallic gold (Au 0 ), with binding energies of 84.0 and 87.7 eV, proving the metallic character of the obtained Au NPs [40]. No residues from the Au precursor (i.e., Cl signal) were found on the foams (see Supplementary Information, Figure S8). On the other hand, the characteristic Pd 3d 5/2 and Pd 3d 3/2 peaks of metallic Pd particles appeared respectively about 335.5-335.9 and 340.8-341.2 eV. The obtained values seem to be slightly shifted from the position of metallic Pd (335.2 and 340.5 eV) [36]. This shift of the energies of the peaks could be related to the presence of not only metallic Pd, but also a small amount of some Pd compound. The potential presence of residues of the Pd precursor, Na 2 PdCl 4 , can be discarded as no Cl residues from the Pd precursor were found on the foams by XPS analysis (see Supplementary Information, Figure S8). On the contrary, according to previous works, this effect can be related to the spontaneous formation of a thin oxide layer (PdO, with 3d 5/2 and 3d 3/2 energies of 336.8 and 342.1 eV) around the Pd NPs [36]. Although this outer oxide layer could negatively affect the performance of the Pd NPs on different applications, it has been previously reported that the NPs are still active on catalysis procedures [36].
Polymers 2020, 12, x FOR PEER REVIEW 10 of 13 previous works, this effect can be related to the spontaneous formation of a thin oxide layer (PdO, with 3d5/2 and 3d3/2 energies of 336.8 and 342.1 eV) around the Pd NPs [36]. Although this outer oxide layer could negatively affect the performance of the Pd NPs on different applications, it has been previously reported that the NPs are still active on catalysis procedures [36]. Therefore, the proposed approach proved to be a suitable route to obtain nanocomposite polymer foams with a large number of noble metal NPs (up to 10 14 particles/m 2 , according to SEM observations ( Figures 5 and 6)), specifically located and stably anchored on the struts of the foams. Additionally, this versatile route can be applied to commercially available porous substrates other than ME foams (see Supplementary Information, Figure S9), without modifying their porous structure, proving the applicability and scalability of this approach, as well as the possibility to be used in diverse applications such as water treatment, catalysis, or sensoring.

Conclusions
This work proposed a facile two-step route to produce polymer foams decorated with Au or Pd nanoparticles (ME/PDMS/Au or ME/PDMS/Pd foams). The metallic nanoparticles were formed insitu onto the surface of the struts of the foams, which in a previous step were covered by a thin layer of PDMS. The presence of Au and Pd nanoparticles, with mean sizes respectively of about 18 and 30 to 45 nm, over this PDMS layer, was confirmed by electron microscopy studies. In addition, the metallic character of the noble metal nanoparticles was demonstrated by XRD and XPS, although the Therefore, the proposed approach proved to be a suitable route to obtain nanocomposite polymer foams with a large number of noble metal NPs (up to 10 14 particles/m 2 , according to SEM observations ( Figures 5 and 6)), specifically located and stably anchored on the struts of the foams. Additionally, this versatile route can be applied to commercially available porous substrates other than ME foams (see Supplementary Information, Figure S9), without modifying their porous structure, proving the applicability and scalability of this approach, as well as the possibility to be used in diverse applications such as water treatment, catalysis, or sensoring.

Conclusions
This work proposed a facile two-step route to produce polymer foams decorated with Au or Pd nanoparticles (ME/PDMS/Au or ME/PDMS/Pd foams). The metallic nanoparticles were formed in-situ onto the surface of the struts of the foams, which in a previous step were covered by a thin layer of PDMS. The presence of Au and Pd nanoparticles, with mean sizes respectively of about 18 and 30 to 45 nm, over this PDMS layer, was confirmed by electron microscopy studies. In addition, the metallic character of the noble metal nanoparticles was demonstrated by XRD and XPS, although the Pd nanoparticles seem to show a thin oxidized outer layer. It was proved that 0.350 mg/mL and 0.500 mg/mL solutions, respectively of the Au and Pd precursors, provide enough availability of metal ions for the formation of the nanoparticles, making it possible to control the amount of metal nanoparticles synthesized by adjusting the reaction times between 1 and 48 h. Maximum Au and Pd loads reached on the ME/PDMS/Au and ME/PDMS/Pd foams were about 1 and 1.2 wt.%, respectively.
Moreover, the adhesion of the Au and Pd nanoparticles to the foams was proved to be stable. Thus, the proposed approach is a promising and versatile route to produce nanocomposite polymer foams with metallic Au or Pd nanoparticles specifically located on the outer surfaces of the struts of the foams. Therefore, the obtained nanocomposite foams can be suitable and safe materials on several applications on which Au and Pd have shown remarkable performances, such as water treatment, catalysis, and sensoring.