Next Article in Journal
Advancements in Serine Protease Inhibitors: From Mechanistic Insights to Clinical Applications
Next Article in Special Issue
Preparation and Electrocatalytic Properties of One-Dimensional Nanorod-Shaped N, S Co-Doped Bimetallic Catalysts of FeCuS-N-C
Previous Article in Journal
Influence of Dopants on Pt/Al2O3-Based Monolithic Catalysts for Autothermal Oxidative Coupling of Methane
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Catalysts
Volume 14
Issue 11
10.3390/catal14110786
catalysts-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Modification of High-Surface-Area Carbons Using Self-Limited Atomic Layer Deposition

by
Mengjie Fan
,
Kai Shen
,
Raymond J. Gorte
and
John M. Vohs
*
Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, PA 19104, USA
*
Author to whom correspondence should be addressed.
Catalysts 2024, 14(11), 786; https://doi.org/10.3390/catal14110786
Submission received: 30 September 2024 / Revised: 29 October 2024 / Accepted: 1 November 2024 / Published: 5 November 2024
(This article belongs to the Special Issue Carbon-Based Catalysts and Electrocatalysts: Synthesis and Applications)
Browse Figures
Versions Notes

Abstract

:
This study explores the application of Atomic Layer Deposition (ALD) to functionalize high-surface-area carbon supports with metal and metal oxide films and particles for applications in catalysis and electrocatalysis. The work reported here demonstrates that, through careful choice of precursors and absorption and reaction conditions, self-limited ALD growth on a high-surface-area carbon support can be achieved. Specific examples presented include the growth of conformal films of ZrO2 and SnO2 and the deposition of Ga2O3 and Pt particles on a carbon black support with a surface area of 250 m2·g−1. A novel strategy for controlling the Pt weight loading and producing sub-nanometer Pt particles on a carbon support using a single ALD cycle is also presented.

Graphical Abstract

1. Introduction

Carbon in its various forms, such as carbon black, graphite, carbon nanotubes, and activated carbon, has a range of properties that make it a useful support material for heterogeneous catalysts [1]. These include a high surface area, good thermal properties, and chemical stability. Additionally, its high electrical conductivity and stability in both basic and acidic environments have led to its wide use as an electrode and catalyst support material in electrochemical devices, including batteries [2], supercapacitors [3], fuel cells [4], and sensors [5]. Specific examples of this include the use of Pt supported on carbon as the anode in low-temperature polymer-exchange membrane fuel cells (PEMFC) [6,7], and IrO2 supported on carbon nanotubes as the catalyst for the oxygen evolution reaction in water electrolysis cells [8].
All these applications of carbon as a support material require either surface functionalization or the dispersion of required catalytic components on its surface. These are hindered by the inert nature of carbon, which interacts only weakly with supported metals and metal oxides. As a result, the catalytic species deposited by traditional methods, such as impregnation, ion exchange, and precipitation, tend to be in the form of particles with relatively low dispersion. This is particularly problematic for many highly catalytic materials, such as Pt, Pd, and IrO2, where high dispersion is required in order for the electrodes to be cost-effective [9,10,11,12,13]. Introducing oxygen-containing functional groups on the carbon surface via pretreatment with a strong acid, such as HNO3 and H2SO4, which provide nucleation sites for deposited catalytic species is one approach that has been used to enhance catalyst dispersion when using these traditional synthesis methods [14,15], but this tends to still produce surfaces with a low density of catalytic sites.
Recently, ALD has also received much attention as a method for depositing catalytic species on carbon surfaces. In a conventional ALD process, a carrier gas is used to expose the growth substrate to a gaseous precursor for the material to be deposited (typically an organometallic compound) under conditions where only a single monolayer of the precursor adsorbs. Following adsorption of the precursor, the sample is exposed to an oxidizing or reducing environment to remove the ligands on the adsorbed metal center. This cycle is then repeated as many times as necessary to achieve the desired surface coverage or film thickness. For example, Pt can be deposited using cycles consisting of exposure to Pt(acac)2 (acac = acetylacetonate), followed by an oxidation or reduction treatment to remove the acac ligands [16,17]. The adsorption of a single monolayer of the precursor per cycle makes ALD growth self-limiting, and the amount of metal or metal oxide deposited is determined by the size and number of ligands on the metal center. ALD has the advantage of providing precise control over the structure of the deposited material and, in many cases, can produce conformal films of both metals and metal oxides on a range of substrates [18,19,20,21].
The formation of conformal films on carbon supports using ALD, however, presents several challenges. First, for high-surface-area carbons, such as those used in catalytic applications, diffusion of the precursor into the nanopores of the carbon can be limiting and, if not accounted for, can result in less than saturated coverage of the adsorbed precursor, causing inhomogeneous growth and low per-cycle growth rates. This is especially problematic when the carbon support is exposed to the precursor using a carrier gas at atmospheric pressure [22]. For example, Fiorentino et al. used ALD to deposit Al2O3 and TiN on carbon nanotube bundles and found that only 25% of the surface of the nanotubes was coated, which they attributed to diffusion-limited penetration of the ALD precursors deep into the nanotube bundle [23]. Second, the adsorbed precursor molecules typically interact weakly with carbon surfaces. This requires the temperature used during precursor exposure to be chosen carefully to ensure that a saturated monolayer of the adsorbed precursor is formed.
While numerous papers can be found in the literature in which ALD has been used to deposit catalytic materials on carbon supports [24,25], many of these studies show unusually low per-cycle deposition rates, indicating that conventional self-limited ALD growth was not achieved. While the materials produced in these studies may still have quite interesting structures and useful catalytic properties, it may be advantageous to obtain true self-limited ALD growth on high-surface-area carbon support materials for certain catalytic applications. True, self-limited growth would provide a more efficient means to functionalize the entire carbon surface, control the size of deposited metal particles, and produce ultrathin conformal films whose catalytic properties could be more easily optimized than those of the bare carbon surface.
The goal of this study was to explore the use of ALD to modify carbon surfaces under conditions where self-limited growth was obtained via reaction of an adsorbed monolayer of the gaseous precursor, as demonstrated via quantitative analysis of the ALD deposition rates. Examples that will be presented include ALD growth of films of SnO2 and ZrO2, and the ALD deposition of Ga2O3 and Pt particles. A strategy for controlling the Pt particle size using ALD is also presented.

2. Results and Discussion

2.1. ALD of Metal Oxide Films on Carbon

Our initial studies focused on the growth of conformal oxide films on a high-surface-area carbon black support. The specific oxide films grown were ZrO2, SnO2, and Ga2O3. Tin oxide and zirconium oxide were chosen because their organometallic precursors, SnCl4 and tetrakis(dimethylamido)zirconium (TDMAZ), can be oxidized at the ALD growth temperature via exposure to H2O. For Ga2O3, Ga(TMHD)3 (TMHD = 2,2,6,6-tetramethyl-3,5-heptanedionate) was used as the precursor, and a high-temperature (673 K) air oxidation step was used to remove the ligands and form the oxide. This precursor was chosen because TMHD is a common ligand used in many organometallic ALD precursors.
To demonstrate that self-limited growth was obtained, the sample mass as a function of the number of ALD cycles was measured for the growth of the ZrO2, SnO2, and Ga2O3 films, as displayed in Figure 1. Gravimetrically measured deposition rates, in units of the number of metal atoms deposited per m2 per ALD cycle for the three oxides, are also listed in Table 1. Figure 1 shows that, in all three cases, the growth curves were nearly linear, which is consistent with self-limited growth; however, as shown in Table 1, the amount deposited per cycle varied somewhat between the oxides.
For the ZrO2 film, which was grown using TDMAZ as the precursor with adsorption and oxidation steps at 453 K, a ZrO2 weight loading of 34% was achieved after only five ALD cycles, which corresponds to a deposition rate of 1.3 × 1018 Zr atoms·m−2·cycle−1 (Table 1). It is useful to compare this deposition rate on carbon with what has been reported previously for the ALD growth of ZrO2 films using the same precursor on other supports. While values for the ZrO2 deposition rates vary somewhat from study to study, reported values are typically between 0.1 and 0.2 nm·cycle−1 [26,27,28,29,30,31]. For example, Shim et al. report a growth rate of 0.15 nm·cycle−1 on a Si3N4-coated planar Si(100) substrate at 523 K using TDMAZ. This corresponds to a metal atom deposition rate of 4.2 × 1018 Zr atoms·m−2·cycle−1, which is comparable to that obtained here for growth on carbon black.
It is also useful to estimate how many ALD cycles would be required to grow a monolayer conformal film of ZrO2 on the carbon support. The density of Zr atoms on an ideal ZrO2(100) surface of cubic zirconia is 7.5 × 1018 Zr atoms·m−2. Assuming that a monolayer ZrO2 film on the carbon support would have a similar Zr atom density, for the measured Zr atom deposition rate, approximately six ALD cycles would be required to completely cover the support. This value seems reasonable when one accounts for the large size of the dimethylamino ligands on the Zr precursor.
As shown in Figure 1b, while a linear growth curve was obtained for SnO2 deposition at 423 K using water as the oxidant, the increase in mass per ALD cycle was substantially less than that for ZrO2 deposition. For SnO2, 23 ALD cycles produced a 21 wt% loading of the oxide. This mass change corresponds to a Sn atom deposition rate of 1.0 × 1017 Sn atoms·m−2·cycle−1. Using the SnO2(100) surface to estimate the density of Sn atoms in a monolayer SnO2 film gives a value of 4.4 × 1018 Sn atoms·m−2. Together, these values indicate that ~40 ALD cycles would be required to coat the surface of the carbon support with a monolayer film of SnO2.
Unlike that for ZrO2 ALD, the SnO2 deposition rate is substantially lower than what has been reported in the literature for SnO2 ALD on other supports. For example, Sundqvist et al. [32] report that, when using a SnI4 precursor to grow SnO2 on a single-crystal α-Al2O3(012) surface, the growth rate was ~0.1 nm·cycle−1. This corresponds to a Sn atom deposition rate of 2.8 × 1018 Sn atoms·m−2·cycle−1, which is over an order of magnitude higher than what was obtained here for growth on carbon. This difference in the deposition rate on carbon and an oxide surface is likely a result of how Sn halide precursors interact with the two surfaces. On alumina, SnI4 (which is expected to have a reactivity similar to the SnCl4 used here) would adsorb dissociatively to form strongly chemisorbed species, resulting in a relatively stable monolayer of the adsorbed precursor. In contrast, SnI4 (or SnCl4) will likely not dissociate and will only exhibit van der Waals interactions with the unreactive carbon surface. This makes it harder to maintain saturation coverage of the precursor and may result in some of it desorbing intact during the water oxidation step. While the situation may be similar for the TDMAZ precursor used for ZrO2 growth, the large dimethylamino ligands in this molecule would result in stronger van der Waals interactions with the surface compared to SnCl4, thereby producing a more stable adsorbed precursor layer.
The TMHD ligands in the Ga(TMHD)3 precursor used for ALD deposition of Ga2O3 are significantly less reactive than those in the Sn and Zr precursors used in this study and cannot be removed by exposure to H2O at the growth temperature. Thus, for ALD growth of Ga2O3, a different procedure was used. As described in Section 3, each ALD cycle for Ga2O3 deposition consisted of exposing the evacuated carbon support to Ga(TMHD)3 at 523 K, followed by heating the sample to 673 K, and then exposing it to air for 10 min. As shown in Figure 1c, this procedure also produced a linear growth curve with a 16 wt% loading of Ga2O3 obtained after 10 ALD cycles. This corresponds to a Ga deposition rate of 4.0 × 1017 Ga atoms·m−2·cycle−1. This deposition rate is essentially the same as that reported by Ramachandran et al. for ALD growth of Ga2O3 on an oxidized Si wafer using Ga(TMHD)3 [33]. It is again useful to use the measured Ga atom deposition rate to estimate how many ALD cycles would be required to form a complete monolayer Ga2O3 film on the carbon support. Using the density of Ga atoms on the β-Ga2O3(100) surface (5.6 × 1018 Ga atoms·m−2) to define a monolayer, we estimate that ~14 ALD Ga cycles would be required.
STEM, TEM, and EDS elemental scans were used to characterize the structure of the deposited oxide films on the carbon black support. Figure 2 displays these images for a 5 ALD cycle ZrO2/C sample. The large TEM image in this figure shows that the carbon black support is composed of carbon particles that are 30–80 nm in diameter. The right side of the figure shows C, O, and Zr EDS elemental scans that were obtained from the designated region in the smaller STEM image. The EDS data show a strong correlation between C, O, and Zr elemental scans. This demonstrates that the ZrO2 film covers nearly all the surface of the carbon support. The conformal nature of the ZrO2 film is further confirmed by the slightly higher intensity in the O and Zr signals around the edges of the carbon particles. In these positions, the electron beam passes through a cross-section of the film, which produces a higher signal than when the beam is more perpendicular to the film, which is the case when it is positioned in the middle of one of the carbon particles. It is also noteworthy that five ZrO2 ALD cycles were sufficient to completely cover the carbon surface, consistent with the analysis of the growth rate data presented above, which indicated that approximately six cycles would be required to completely coat the surface.
To evaluate the thermal stability of the ALD-generated ZrO2 thin film on carbon, the same five ALD cycle ZrO2/C sample was treated in air at 623 K for 10 min, followed by structural characterization using TEM. As shown in Figure 3, distinct dark spots are visible in the TEM image, indicating that the heat treatment at 623 K promoted partial agglomeration of the film into particles that are ~2 nm in diameter. The EDS mapping on the right shows the distribution of C, O, and Zr signals. While some particle agglomeration is evident in the TEM image, the overall Zr signal remains well aligned with C and O, suggesting that the majority of the ZrO2 remains uniformly distributed on the carbon black surface, with only a small portion aggregating into particles. For applications in electrodes for electrochemical devices, carbon is not likely to be exposed to such high-temperature conditions. Therefore, the mild particle agglomeration observed here after heating to 623 K would not pose a significant issue for these applications.
A similar set of TEM/EDS images for a 23 ALD cycle SnO2/C sample is presented in Figure 4. The Sn and O elemental EDS scans obtained from this sample show a film-like structure for the SnO2, but the Sn EDS signal intensity is not uniform over the carbon surface, indicating that some islanding has occurred. From the growth rate analysis presented above, it is estimated that ~40 ALD cycles would be required to form a SnO2 film that completely covers the surface of the support. Based on this estimate, the SnO2 film in the 23 ALD cycle SnO2/C sample should cover approximately half of the surface, which seems reasonably consistent with the TEM/EDS results.
For the 23 ALD cycle SnO2/C sample, we also investigated the thermal stability of the SnO2 film by heating it to 623 K (i.e., 200 K higher than the growth temperature) for 10 min. TEM/EDS scans obtained from this sample are presented in Figure 5. Similar to the case of ZrO2, the TEM image contains features that suggest some particle formation may have occurred, but this is not readily evident in the EDS scans, suggesting that much of the film nature of the SnO2 has been maintained. This result, therefore, demonstrates that the ALD-deposited SnO2 has high thermal stability and mostly maintains its film-like structure at elevated temperatures.
TEM/EDS images for a 10 ALD cycle Ga2O3/C sample are displayed in Figure 6. Based on the growth rate analysis presented above, the surface of this sample should be mostly coated with a Ga2O3 film. However, in contrast to the ZrO2 and SnO2 samples, the TEM and EDS images for this sample clearly show that particle formation has occurred and that the deposited Ga2O3 is in the form of 6–7 nm nanoparticles distributed evenly across the surface of the carbon black, leaving large regions of the carbon surface uncovered. This difference is likely related to the different conditions required to remove the ligands and oxidize the metal center for the different precursors. For the Zr and Sn precursors, this was accomplished by exposing the sample to water vapor at the growth temperatures (423 K or 453 K). In contrast, for Ga2O3 growth, the Ga(TMHD)3 precursor was adsorbed at 523 K and then oxidized in air at 673 K. Since the growth rate data indicate that a saturation monolayer coverage of the Ga precursor was obtained during each ALD cycle, it appears that agglomeration and particle formation must have occurred either while heating to 673 K or during oxidation at this temperature.

2.2. ALD of Pt on Carbon

In addition to using ALD for depositing metal oxides on carbon black, we explored the use of ALD to deposit catalytic metals, such as Pt. Since the most common way to deposit metals on high-surface-area supports for catalytic applications is via wet infiltration of an aqueous metal salt solution, followed by drying and reduction, we synthesized a 10 wt% Pt/C sample using this method to provide a base case for comparison to samples synthesized using ALD. STEM and EDS images of this infiltrated sample are displayed in Figure 7a. A histogram showing the particle size distribution obtained from the STEM image is also included in the figure. These data show that the sample produced by infiltration resulted in a bimodal particle size distribution, with the majority of the particles between 2 and 7 nm in diameter, and a few larger particles between 9 and 10 nm in diameter. It is important to note that, while there are more smaller particles, the majority of the Pt is contained in the larger particles. An XRD pattern for this sample, displayed in Figure 8, contains broad peaks at 18° and 25° 2θ due to the carbon black substrate and narrower peaks at 39.7°, 46.2°, and 68° 2θ due to diffraction from the [111], [200], and [220] planes of Pt, respectively. Using the Scherrer equation and the full width at half maximum (FWHM) of the Pt peaks, the Pt particle size in this sample is estimated to be 11 nm, which again demonstrates that the majority of the Pt is in the larger particles.
Figure 7b displays STEM and EDS images and a particle size histogram for the Pt/C sample prepared using a single Pt ALD cycle. Since Pt is an excellent oxidation catalyst, for the Pt(acac)2 precursor, the ligands cannot be removed by oxidation, because this also results in combustion of the carbon support. It was found, however, that the acac ligands could be removed by heating in H2 at 623 K. The Pt weight loading for this sample was 10.4 wt%, which corresponds to a Pt deposition rate of 1.3 × 1018 Pt atoms·m−2·cycle−1, similar to the Zr deposition rate during ZrO2 ALD growth and consistent with the adsorption and reaction of a saturated monolayer of the Pt(acac)2 precursor.
The STEM image shows that, even though a monolayer of the precursor was formed during the ALD adsorption step, Pt particle formation occurs during the reduction step. This is not surprising since Pt atoms are expected to have high mobility on the C surface at the reduction temperature of 623 K. While both the infiltrated and one ALD cycle samples contain ~10 wt% Pt and were reduced in H2 at 623 K, the particle size distributions for the two samples differ significantly. The histogram shows that a unimodal particle distribution was obtained for the ALD Pt/C sample, with an average particle size of 2.8 nm. The XRD pattern in Figure 8 shows that the Pt diffraction peaks for this sample are also quite broad, with the Scherrer equation giving an average particle size of 3.6 nm. Note that this value is biased toward the larger particles since they contain a larger fraction of the Pt. These results demonstrate that, with carefully chosen conditions, a single Pt ALD cycle can be used to produce high coverage of 2–3 nm Pt nanoparticles on a carbon support.
The thermal stability of the Pt particles in the ALD Pt/C sample was also investigated by heating the reduced ALD Pt/C sample in H2 to 773 K for 60 min. A STEM image, EDS map, and particle histogram obtained after this treatment are displayed in Figure 7c. The XRD pattern for this sample is also displayed in Figure 8. These data show that the additional heat treatment resulted in a narrowing of the particle size distribution and an increase in the average particle size from 2.8 nm to 3.5 nm, as determined from the histogram data (XRD gives a value of 3.7 nm for this sample). A close examination of the STEM images shows that the primary change upon heating to 773 K was the coalescence of the smallest Pt nanoparticles into the larger 3–4 nm particles, consistent with an Ostwald ripening type process. Even taking into account this sintering, it is noteworthy that ALD Pt deposition followed by heating to high temperature produces a carbon sample with a high coverage of relatively uniform 3–4 nm Pt particles.
To explore the possibility that ALD could be used to produce even smaller Pt particles on the carbon support, we modified the ALD procedure by diluting the Pt(acac)2 precursor with liquid acetylacetone (Hacac) and exposing the carbon support to the vapor from this mixture during the ALD adsorption step. The idea here is that Hacac will compete for adsorption sites with Pt(acac)2. Due to the similarity of the acac ligands and Hacac, they will both be removed by the 623 K H2 reduction treatment. The Pt(acac)2 and Hacac will have different adsorption energies on the surface, but the relative coverage of each in the adsorbed monolayer may be controllable by varying their ratio in the precursor solution. Using this approach, Pt/C samples were prepared using precursor solutions with Pt(acac)2:Hacac ratios of 0.01 and 0.001, resulting in samples with 2 wt% Pt and 0.2 wt% Pt, respectively. STEM images for these samples are presented in Figure 9 and show that the approach was successful in producing smaller Pt particles with a relatively uniform size distribution.
The image of the Pt/C sample prepared using one Pt ALD cycle with a precursor solution containing a Pt(acac)2:Hacac ratio of 0.01 in Figure 9a shows that, after reduction in H2 at 623 K, the carbon support is sparsely covered with Pt particles that are 1 to 2 nm in diameter, significantly smaller than those obtained using pure Pt(acac)2 as the precursor. Decreasing the Pt(acac)2:Hacac ratio to 0.001 produced a sample covered with Pt particles < 1 nm in diameter, as shown in Figure 9b. Since these particles are stable against sintering at temperatures up to 623 K (i.e., the reduction temperature), this synthesis approach provides a relatively easy method to produce very highly dispersed Pt particles on a high surface area carbon support for use in a range of catalytic applications, including those in electrochemistry.
Since alloying with a second metal is often used to alter both the stability and catalytic properties of noble metal catalysts, we also explored the use of ALD to synthesize PtSn alloy particles on the carbon black support. In this work, we used one Pt ALD cycle to deposit Pt on a carbon black support that had previously been coated with a SnO2 film using ALD. This SnO2/C sample had a 21 wt% loading of SnO2 and had the same structure as that shown in Figure 4. As before, the final step of the Pt ALD process was reduction in H2 at 623 K for 1 h. In this case, one ALD cycle using Pt(acac)2 led to a weight loading of 4.4 wt%. The decreased growth rate compared to the Pt/C sample highlights the presence of the SnO2 film and its impact on the adsorption mechanism of Pt(acac)2. STEM and EDS images of this sample are displayed in Figure 10. The STEM image shows that the carbon support is covered by metal particles with an average diameter of 2.8 nm. The EDS images reveal that Sn did not remain in the form of a conformal film of SnO2 on the carbon support but had become reduced during the 623 K H2 reduction step and then migrated to the Pt resulting in the formation of PtSn alloy particles, as evidence by the correspondence between the Pt and Sn signals. The XRD pattern for this sample also showed broadening of the Pt diffraction peaks and the appearance of new peaks at 30° and 43° two theta, which are characteristic of a PtSn alloy. While we did not further investigate using ALD to produce PtSn alloy particles, the Pt:Sn ratio in the particles could easily be controlled by varying the amount of SnO2 deposited on the support prior to Pt ALD.

3. Materials and Methods

Carbon black (Vulcan XC-72R) with a surface area of 250 m−2·g was used as the support material in this study. To remove any adsorbed species, prior to metal or metal oxide deposition, the carbon was pre-treated by heating it to 773 K. ALD deposition of Pt, Ga2O3, SnO2, and ZrO2 onto the carbon support was performed using a purpose-built adsorption apparatus that has been described in detail previously [34,35]. A key feature of this system was that it allowed the sample to be evacuated prior to exposure to the precursor vapor and did not use a carrier gas—features that are needed to limit pore diffusional limitations and achieve conformal growth on high surface area materials [22]. A typical ALD cycle consisted of (1) evacuating the substrate and heating it to the desired deposition temperature, (2) exposing the evacuated substrate to the precursor vapor, (3) evacuating to remove excess precursor vapor, and (4) an oxidation or reduction treatment to remove the ligands on the adsorbed precursor. Films of SnO2 and ZrO2 were grown with the carbon support material held at 423 K using SnCl4 (98%, Sigma-Aldrich, St. Louis, MO, USA) and at 473 K using [(CH3)2N]4Zr (tetrakis(dimethylamino)zirconium, TDMAZ, 99%, Strem, Newburyport, MA, USA) as the precursors. These temperatures were chosen because they were found to be low enough to prevent chemical vapor deposition and high enough to allow for the adsorption of a precursor monolayer, not the formation of adsorbed multilayers. The chloride and dimethylamino ligands could be removed and the metal center oxidized by exposure to water vapor with the sample held at the growth temperature. For Ga2O3 deposition, Ga(TMHD)3 (TMHD = 2,2,6,6-tetramethyl-3,5-heptanedionate, 99%, Strem) was used as the precursor, and the substrate was held at 523 K during the exposure. This higher adsorption temperature was required here to prevent the formation of multilayers of the adsorbed Ga(TMHD)3. The TMHD ligand was removed by oxidizing in air at 673 K for 10 min at the end of each deposition cycle.
Deposition of Pt on carbon black was achieved using Pt(acac)2 (acac = acetylacetonate, 98%, Strem) as the precursor at 443 K, and the ligands were removed by reduction in flowing 5% H2-He at 623 K for 1 h. In some cases, the Pt(acac)2 precursor was mixed with Hacac (acetylacetone, 99%, Sigma-Aldrich), and the vapor form of the mixture was exposed to the carbon support during ALD growth. To provide a base case Pt/C sample for comparison, a Pt/C sample containing 10 wt% Pt was also prepared using conventional wet impregnation. The impregnation solution was prepared by dissolving the required amount of tetraammineplatinum(II) nitrate (Pt(NH3)4(NO3)2, 99.99%, Alfa Aesar, Haverhill, MA, USA) in a solution containing 75 vol% water and 25 vol% ethanol. The carbon support powder was then added to the solution, and the mixture was stirred for several minutes. This mixture was then left in an oven at 343 K overnight to evaporate the solvents. The dried powder was then reduced in a flowing stream of 5% H2 in He at 623 K for 1 h.
The surface area of the carbon black substrate was determined using BET analysis of N2 adsorption isotherm at 77 K, which was measured using a home-built adsorption apparatus. A Rigaku MiniFlex diffractometer with a Cu-Kα source was used to perform X-ray diffraction (XRD). A JEOL NEOARM was used to collect transmission electron microscopy (TEM) and energy dispersive X-ray spectra (EDS) images, and a JEOL F200 was used to collect transmission electron microscopy (TEM) images. Sample preparation for STEM/TEM analysis consisted of suspending the sample in ethanol and sonicating to achieve high dispersion. The solution was then added dropwise onto a carbon-coated copper TEM grid (Electron Microscopy Sciences, Hatfield, PA, USA).

4. Summary

In this study, we have demonstrated that through careful choice of adsorption and reaction conditions, and by using a vacuum ALD apparatus that helps reduce diffusional limitations, self-limiting ALD growth of a range of materials on high surface area carbon supports can be obtained. Our findings further demonstrate that ALD can provide a high degree of control over both the amount and the structure of the catalytic components that are deposited.
Quantitative analysis demonstrated that for most of the films grown in this study, the number of metal atoms deposited per ALD cycle was consistent with the adsorption and subsequent reaction of an adsorbed monolayer of the precursor. For the ALD deposition of ZrO2 and SnO2 using TDMAZ and SnCl4 precursors, respectively, the growth of conformal films on the carbon support was achieved. These films exhibited good thermal stability, undergoing only modest agglomeration upon heating to 623 K. In contrast, during ALD growth of Ga2O3, the higher oxidation temperature, 673 K, required to remove the ligands on the Ga(TMHD)3 precursor resulted in the formation of Ga2O3 particles.
For Pt ALD, it was found that reduction in H2 was required to remove the ligands on the Pt(acac)2 precursor. Due to Pt’s high catalytic activity, using an oxidation treatment for this purpose results in combustion of the carbon support. A single Pt ALD cycle followed by heating to 773 K in H2 was found to produce a surface covered with a high concentration of Pt particles and a relatively narrow particle size distribution with an average size of 3.5 nm. Furthermore, it was shown that both the amount of Pt deposited and the average Pt particle size could be controlled during ALD deposition by diluting the Pt(acac)2 in Hacac and using the vapor from this mixture as the ALD precursor. Using this method, it was demonstrated that Pt particles with an average size of less than 1 nm in diameter could be synthesized on a high surface area carbon support using a single ALD cycle. Finally, it was demonstrated that by depositing Pt on top of a preformed SnO2 film followed by reduction in H2, PtSn alloy particles were produced on the carbon support.

Author Contributions

Conceptualization: M.F., R.J.G. and J.M.V.; Investigation: M.F. and K.S.; Project Administration: R.J.G. and J.M.V.; Writing—original draft: M.F.; Writing—review and editing M.F., R.J.G. and J.M.V. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the National Science Foundation grant no. DMREF-2323701. Some of the work was performed at the Singh Center for Nanotechnology at the University of Pennsylvania, a member of the National Nanotechnology Coordinated Infrastructure (NNCI) network, which is supported by the National Science Foundation (GrantNNCI-2025608).

Data Availability Statement

All data are contained in the article.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Flandrois, S.; Simon, B. Carbon Materials for Lithium-Ion Rechargeable Batteries. Carbon 1999, 37, 165–180. [Google Scholar] [CrossRef]
  2. Yuan, S.; Lai, Q.; Duan, X.; Wang, Q. Carbon-Based Materials as Anode Materials for Lithium-Ion Batteries and Lithium-Ion Capacitors: A Review. J. Energy Storage 2023, 61, 106716. [Google Scholar] [CrossRef]
  3. Lobato-Peralta, D.R.; Okoye, P.U.; Alegre, C. A Review on Carbon Materials for Electrochemical Energy Storage Applications: State of the Art, Implementation, and Synergy with Metallic Compounds for Supercapacitor and Battery Electrodes. J. Power Sources 2024, 617, 235140. [Google Scholar] [CrossRef]
  4. Jaleh, B.; Nasrollahzadeh, M.; Eslamipanah, M.; Nasri, A.; Shabanlou, E.; Manwar, N.R.; Zboril, R.; Fornasiero, P.; Gawande, M.B. The Role of Carbon-Based Materials for Fuel Cells Performance. Carbon 2022, 198, 301–352. [Google Scholar] [CrossRef]
  5. Gibi, C.; Liu, C.-H.; Barton, S.C.; Anandan, S.; Wu, J.J. Carbon Materials for Electrochemical Sensing Application—A Mini Review. J. Taiwan Inst. Chem. Eng. 2024, 154, 105071. [Google Scholar] [CrossRef]
  6. Antolini, E. Carbon Supports for Low-Temperature Fuel Cell Catalysts. Appl. Catal. B Environ. 2009, 88, 1–24. [Google Scholar] [CrossRef]
  7. Trogadas, P.; Fuller, T.F.; Strasser, P. Carbon as Catalyst and Support for Electrochemical Energy Conversion. Carbon 2014, 75, 5–42. [Google Scholar] [CrossRef]
  8. Badam, R.; Hara, M.; Huang, H.-H.; Yoshimura, M. Synthesis and Electrochemical Analysis of Novel IrO2 Nanoparticle Catalysts Supported on Carbon Nanotube for Oxygen Evolution Reaction. Int. J. Hydrogen Energy 2018, 43, 18095–18104. [Google Scholar] [CrossRef]
  9. Olowoyo, J.O.; Gharahshiran, V.S.; Zeng, Y.; Zhao, Y.; Zheng, Y. Atomic/Molecular Layer Deposition Strategies for Enhanced CO2 Capture, Utilisation and Storage Materials. Chem. Soc. Rev. 2024, 53, 5428–5488. [Google Scholar] [CrossRef]
  10. Kinoshita, K. Particle Size Effects for Oxygen Reduction on Highly Dispersed Platinum in Acid Electrolytes. J. Electrochem. Soc. 1990, 137, 845. [Google Scholar] [CrossRef]
  11. Mukerjee, S.; McBreen, J. Effect of Particle Size on the Electrocatalysis by Carbon-Supported Pt Electrocatalysts: An in Situ XAS Investigation. J. Electroanal. Chem. 1998, 448, 163–171. [Google Scholar] [CrossRef]
  12. Zhou, W.; Li, M.; Ding, O.L.; Chan, S.H.; Zhang, L.; Xue, Y. Pd Particle Size Effects on Oxygen Electrochemical Reduction. Int. J. Hydrogen Energy 2014, 39, 6433–6442. [Google Scholar] [CrossRef]
  13. Abbott, D.F.; Lebedev, D.; Waltar, K.; Povia, M.; Nachtegaal, M.; Fabbri, E.; Copéret, C.; Schmidt, T.J. Iridium Oxide for the Oxygen Evolution Reaction: Correlation between Particle Size, Morphology, and the Surface Hydroxo Layer from Operando XAS. Chem. Mater. 2016, 28, 6591–6604. [Google Scholar] [CrossRef]
  14. Gopiraman, M.; Ganesh Babu, S.; Khatri, Z.; Kai, W.; Kim, Y.A.; Endo, M.; Karvembu, R.; Kim, I.S. An Efficient, Reusable Copper-Oxide/Carbon-Nanotube Catalyst for N-Arylation of Imidazole. Carbon 2013, 62, 135–148. [Google Scholar] [CrossRef]
  15. Shen, J.; Huang, W.; Wu, L.; Hu, Y.; Ye, M. Study on Amino-Functionalized Multiwalled Carbon Nanotubes. Mater. Sci. Eng. A 2007, 464, 151–156. [Google Scholar] [CrossRef]
  16. Shen, K.; Lee, S.; Kwon, O.; Fan, M.; Gorte, R.J.; Vohs, J.M. Effect of Perovskite-Fluorite Phase Transition on the Catalytic Activity of Pt/CeFeOx. ACS Catal. 2023, 13, 11144–11152. [Google Scholar] [CrossRef]
  17. Mao, X.; Foucher, A.C.; Montini, T.; Stach, E.A.; Fornasiero, P.; Gorte, R.J. Epitaxial and Strong Support Interactions between Pt and LaFeO3 Films Stabilize Pt Dispersion. J. Am. Chem. Soc. 2020, 142, 10373–10382. [Google Scholar] [CrossRef]
  18. Leskelä, M.; Ritala, M. Atomic Layer Deposition (ALD): From Precursors to Thin Film Structures. Thin Solid Film. 2002, 409, 138–146. [Google Scholar] [CrossRef]
  19. Knisley, T.J.; Kalutarage, L.C.; Winter, C.H. Precursors and Chemistry for the Atomic Layer Deposition of Metallic First Row Transition Metal Films. Coord. Chem. Rev. 2013, 257, 3222–3231. [Google Scholar] [CrossRef]
  20. Johnson, R.W.; Hultqvist, A.; Bent, S.F. A Brief Review of Atomic Layer Deposition: From Fundamentals to Applications. Mater. Today 2014, 17, 236–246. [Google Scholar] [CrossRef]
  21. Ponraj, J.S.; Attolini, G.; Bosi, M. Review on Atomic Layer Deposition and Applications of Oxide Thin Films. Crit. Rev. Solid State Mater. Sci. 2013, 38, 203–233. [Google Scholar] [CrossRef]
  22. Onn, T.M.; Küngas, R.; Fornasiero, P.; Huang, K.; Gorte, R.J. Atomic Layer Deposition on Porous Materials: Problems with Conventional Approaches to Catalyst and Fuel Cell Electrode Preparation. Inorganics 2018, 6, 34. [Google Scholar] [CrossRef]
  23. Fiorentino, G.; Vollebregt, S.; Tichelaar, F.D.; Ishihara, R.; Sarro, P.M. Impact of the Atomic Layer Deposition Precursors Diffusion on Solid-State Carbon Nanotube Based Supercapacitors Performances. Nanotechnology 2015, 26, 064002. [Google Scholar] [CrossRef] [PubMed]
  24. Marichy, C.; Pinna, N. Carbon-Nanostructures Coated/Decorated by Atomic Layer Deposition: Growth and Applications. Coord. Chem. Rev. 2013, 257, 3232–3253. [Google Scholar] [CrossRef]
  25. Guan, C.; Wang, J. Recent Development of Advanced Electrode Materials by Atomic Layer Deposition for Electrochemical Energy Storage. Adv. Sci. 2016, 3, 1500405. [Google Scholar] [CrossRef]
  26. Rasteiro, L.F.; Motin, M.A.; Vieira, L.H.; Assaf, E.M.; Zaera, F. Growth of ZrO2 Films on Mesoporous Silica Sieve via Atomic Layer Deposition. Thin Solid Film. 2023, 768, 139716. [Google Scholar] [CrossRef]
  27. Niinistö, J.; Kukli, K.; Kariniemi, M.; Ritala, M.; Leskelä, M.; Blasco, N.; Pinchart, A.; Lachaud, C.; Laaroussi, N.; Wang, Z.; et al. Novel Mixed Alkylamido-Cyclopentadienyl Precursors for ALD of ZrO2 Thin Films. J. Mater. Chem. 2008, 18, 5243–5247. [Google Scholar] [CrossRef]
  28. Shim, J.H.; Chao, C.-C.; Huang, H.; Prinz, F.B. Atomic Layer Deposition of Yttria-Stabilized Zirconia for Solid Oxide Fuel Cells. Chem. Mater. 2007, 19, 3850–3854. [Google Scholar] [CrossRef]
  29. Sempel, J.D.; Kaariainen, M.-L.; Colleran, T.A.; Lifschitz, A.M.; George, S.M. Ultrathin ZrO2 Thickness Control on TiO2/ZrO2 Core/Shell Nanoparticles Using ZrO2 Atomic Layer Deposition and Etching. J. Vac. Sci. Technol. A 2024, 42, 052604. [Google Scholar] [CrossRef]
  30. Fan, M.; Ji, Y.; Lawal, A.; Abdelrahman, O.A.; Gorte, R.J.; Vohs, J.M. Site and Structural Requirements for the Dehydra-Decyclization of Cyclic Ethers on ZrO2. Catalysts 2022, 12, 902. [Google Scholar] [CrossRef]
  31. Fan, M.; Wang, C.-Y.; Gorte, R.J.; Vohs, J.M. Synthesis of Lewis Acid Sites in SBA-15 and Amorphous Silica Using Vapor Phase Infiltration of Zr and Sn. Catal. Lett. 2024, 154, 2725–2734. [Google Scholar] [CrossRef]
  32. Sundqvist, J.; Lu, J.; Ottosson, M.; Hårsta, A. Growth of SnO2 Thin Films by Atomic Layer Deposition and Chemical Vapour Deposition: A Comparative Study. Thin Solid Film. 2006, 514, 63–68. [Google Scholar] [CrossRef]
  33. Ramachandran, R.K.; Dendooven, J.; Botterman, J.; Pulinthanathu Sree, S.; Poelman, D.; Martens, J.A.; Poelman, H.; Detavernier, C. Plasma Enhanced Atomic Layer Deposition of Ga2O3 Thin Films. J. Mater. Chem. A 2014, 2, 19232–19238. [Google Scholar] [CrossRef]
  34. Onn, T.M.; Arroyo-Ramirez, L.; Monai, M.; Oh, T.-S.; Talati, M.; Fornasiero, P.; Gorte, R.J.; Khader, M.M. Modification of Pd/CeO2 Catalyst by Atomic Layer Deposition of ZrO2. Appl. Catal. B Environ. 2016, 197, 280–285. [Google Scholar] [CrossRef]
  35. Onn, T.M.; Zhang, S.; Arroyo-Ramirez, L.; Xia, Y.; Wang, C.; Pan, X.; Graham, G.W.; Gorte, R.J. High-Surface-Area Ceria Prepared by ALD on Al2O3 Support. Appl. Catal. B Environ. 2017, 201, 430–437. [Google Scholar] [CrossRef]
Catalysts 14 00786 g001
Figure 1. % Weight loading as a function of ALD cycles for (a) ZrO2, (b) SnO2, and (c) Ga2O3 on a carbon black support with an initial surface area of 250 m2/g.
Figure 1. % Weight loading as a function of ALD cycles for (a) ZrO2, (b) SnO2, and (c) Ga2O3 on a carbon black support with an initial surface area of 250 m2/g.
Catalysts 14 00786 g001
Catalysts 14 00786 g002
Figure 2. TEM (large image), STEM, and EDS images of a five ALD cycle ZrO2 on carbon black support after 453 K oxidation with water vapor. The area of the elemental EDS images corresponds to the red box in the STEM image.
Figure 2. TEM (large image), STEM, and EDS images of a five ALD cycle ZrO2 on carbon black support after 453 K oxidation with water vapor. The area of the elemental EDS images corresponds to the red box in the STEM image.
Catalysts 14 00786 g002
Catalysts 14 00786 g003
Figure 3. TEM (large image), STEM, and EDS images of a five ALD cycle ZrO2 on carbon black support after 623 K oxidation in air. The area of the elemental EDS images corresponds to the red box in the STEM image.
Figure 3. TEM (large image), STEM, and EDS images of a five ALD cycle ZrO2 on carbon black support after 623 K oxidation in air. The area of the elemental EDS images corresponds to the red box in the STEM image.
Catalysts 14 00786 g003
Catalysts 14 00786 g004
Figure 4. TEM (large image), STEM, and EDS images of a 23 ALD cycle SnO2 on carbon black support after 423 K oxidation with water. The area of the elemental EDS images corresponds to the red box in the STEM image.
Figure 4. TEM (large image), STEM, and EDS images of a 23 ALD cycle SnO2 on carbon black support after 423 K oxidation with water. The area of the elemental EDS images corresponds to the red box in the STEM image.
Catalysts 14 00786 g004
Catalysts 14 00786 g005
Figure 5. TEM (large image), STEM, and EDS images of a 23 ALD cycle SnO2 on carbon black support after 623 K oxidation in air. The area of the elemental EDS images corresponds to the red box in the STEM image.
Figure 5. TEM (large image), STEM, and EDS images of a 23 ALD cycle SnO2 on carbon black support after 623 K oxidation in air. The area of the elemental EDS images corresponds to the red box in the STEM image.
Catalysts 14 00786 g005
Catalysts 14 00786 g006
Figure 6. TEM (large image), STEM, and EDS images of a 10 ALD cycle Ga2O3 on carbon black support after 673 K oxidation. The area of the elemental EDS images corresponds to the red box in the STEM image.
Figure 6. TEM (large image), STEM, and EDS images of a 10 ALD cycle Ga2O3 on carbon black support after 673 K oxidation. The area of the elemental EDS images corresponds to the red box in the STEM image.
Catalysts 14 00786 g006
Catalysts 14 00786 g007
Figure 7. STEM and EDS mapping of (a) impregnated 10 wt% Pt on carbon black support after 623 K He-H2 reduction (average particle size 4.6 nm); (b) one ALD cycle of Pt on carbon black support (10.4 wt%) after 623 K He-H2 reduction (average particle size 2.8 nm); and (c) one ALD cycle of Pt on carbon black support (10.4 wt%) after 773 K He-H2 reduction (average particle size 3.5 nm). The area of the Pt EDS images corresponds to the red box in the STEM images.
Figure 7. STEM and EDS mapping of (a) impregnated 10 wt% Pt on carbon black support after 623 K He-H2 reduction (average particle size 4.6 nm); (b) one ALD cycle of Pt on carbon black support (10.4 wt%) after 623 K He-H2 reduction (average particle size 2.8 nm); and (c) one ALD cycle of Pt on carbon black support (10.4 wt%) after 773 K He-H2 reduction (average particle size 3.5 nm). The area of the Pt EDS images corresponds to the red box in the STEM images.
Catalysts 14 00786 g007
Catalysts 14 00786 g008
Figure 8. XRD of (a) impregnated 10 wt% Pt on carbon black support after 623 K He-H2 reduction (average particle size 11 nm), (b) one ALD cycle of Pt on carbon black support after 623 K He-H2 reduction (average particle size 3.6 nm); and (c) one ALD cycle of Pt on carbon black support after 773 K He-H2 reduction (average particle size 3.7 nm). The diamonds designate the peaks that are characteristic of Pt. All of the patterns have the same vertical scale.
Figure 8. XRD of (a) impregnated 10 wt% Pt on carbon black support after 623 K He-H2 reduction (average particle size 11 nm), (b) one ALD cycle of Pt on carbon black support after 623 K He-H2 reduction (average particle size 3.6 nm); and (c) one ALD cycle of Pt on carbon black support after 773 K He-H2 reduction (average particle size 3.7 nm). The diamonds designate the peaks that are characteristic of Pt. All of the patterns have the same vertical scale.
Catalysts 14 00786 g008
Catalysts 14 00786 g009
Figure 9. STEM image of a one ALD cycle of Pt on carbon black support with Hacac dilution targeting (a) 2 wt% Pt loading (estimated particle size 1.3 nm) and (b) 0.2 wt% Pt loading. Both samples were reduced in a continuous flow of He-H2 at 623 K for 1 h (estimated particle size 0.8 nm). The bright features in each image correspond to Pt particles.
Figure 9. STEM image of a one ALD cycle of Pt on carbon black support with Hacac dilution targeting (a) 2 wt% Pt loading (estimated particle size 1.3 nm) and (b) 0.2 wt% Pt loading. Both samples were reduced in a continuous flow of He-H2 at 623 K for 1 h (estimated particle size 0.8 nm). The bright features in each image correspond to Pt particles.
Catalysts 14 00786 g009
Catalysts 14 00786 g010
Figure 10. STEM and EDS images of Pt-SnO2 on carbon black support after reduction in a continuous flow of He-H2 at 623 K for 1 h. The area of the elemental EDS images corresponds to the red box in the STEM image.
Figure 10. STEM and EDS images of Pt-SnO2 on carbon black support after reduction in a continuous flow of He-H2 at 623 K for 1 h. The area of the elemental EDS images corresponds to the red box in the STEM image.
Catalysts 14 00786 g010
Table 1. Summary of ALD procedure and growth rate.
Table 1. Summary of ALD procedure and growth rate.
ComponentLigandGrowth Rate on Carbon per Cycle
ZrO2TDMA1.3 × 1018 atoms/m2
SnO2Chloride1.0 × 1017 atoms/m2
Ga2O3TMHD4.0 × 1017 atoms/m2
Ptacac1.3 × 1018 atoms/m2
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Fan, M.; Shen, K.; Gorte, R.J.; Vohs, J.M. Modification of High-Surface-Area Carbons Using Self-Limited Atomic Layer Deposition. Catalysts 2024, 14, 786. https://doi.org/10.3390/catal14110786

AMA Style

Fan M, Shen K, Gorte RJ, Vohs JM. Modification of High-Surface-Area Carbons Using Self-Limited Atomic Layer Deposition. Catalysts. 2024; 14(11):786. https://doi.org/10.3390/catal14110786

Chicago/Turabian Style

Fan, Mengjie, Kai Shen, Raymond J. Gorte, and John M. Vohs. 2024. "Modification of High-Surface-Area Carbons Using Self-Limited Atomic Layer Deposition" Catalysts 14, no. 11: 786. https://doi.org/10.3390/catal14110786

APA Style

Fan, M., Shen, K., Gorte, R. J., & Vohs, J. M. (2024). Modification of High-Surface-Area Carbons Using Self-Limited Atomic Layer Deposition. Catalysts, 14(11), 786. https://doi.org/10.3390/catal14110786

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop