PRIORITY CLAIM

This application claims priority to U.S. Provisional Application No. 62/426,858, filed November 28, 2016, which is incorporated by reference in its entirety.

TECHNICAL FIELD

This invention relates to oxidation catalysts.

BACKGROUND

In newly built houses, formaldehyde (HCHO) generated from construction and furnishing materials is one of the major indoor pollutants. Various approaches have been made to eliminate HCHO including thermal and photo- catalytic oxidation, plasma decomposition and adsorption. Catalytic oxidation is an attractive option owing to the production of less toxic C0 ₂ and H ₂ 0 products and low cost of operation. While oxide-supported noble metal catalysts (Pt, Pd, or Au) have been shown to be active at room temperature, the cost of such catalysts prohibits their wide spread use.

SUMMARY

In one aspect, a method of oxidizing an organic contaminant in a gas can include providing a catalyst, exposing the catalyst to the gas including the organic contaminant and oxygen without adding heat, and, at selected times, exposing the catalyst to water, oxygen or heat, or combinations thereof, for a regeneration period.

In another aspect, a method of regenerating a catalyst for oxidizing an organic contaminant in a gas can include exposing a catalyst that has reduced activity relative to an unused portion of the catalyst to water, oxygen or heat, or combinations thereof, for a regeneration period.

In another aspect, a system for oxidizing an organic contaminant in a gas can include a catalyst and a regeneration system, wherein the regeneration system exposes a catalyst that has reduced activity relative to an unused portion of the catalyst to water or heat, or both, for a regeneration period. In certain embodiments, the method can include exposing the catalyst to water for the regeneration period. In certain embodiments, the method can include exposing the catalyst to heat for the regeneration period. In certain embodiments, the method can include exposing the catalyst to water and heat for the regeneration period. In certain embodiments, the method can include exposing the catalyst to oxygen for the regeneration period, optionally, with the heating or the water or both.

In certain embodiments, the organic contaminant can be formaldehyde, acetaldehyde, methanol or formic acid. For example, the organic contaminant can be formaldehyde.

In certain embodiments, the catalyst can include manganese oxide.

In certain embodiments, the selected time can be between 30 and 60 minutes after exposing the catalyst to the gas.

In certain embodiments, exposing the catalyst to the gas including the organic

contaminant and oxygen without adding heat can be repeated after the regeneration period.

In certain embodiments, exposing the catalyst to water or heat, or both, for a regeneration period is repeated after the repeated exposing the catalyst to the gas including the organic contaminant and oxygen without adding heat.

In certain embodiments, the regeneration period can be between about 1 and 10 minutes.

Other aspects, embodiments, and features will be apparent from the following description, the drawings, and the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 depicts activities of classes of catalysts.

FIG. 2 depicts an example of a flow reactor.

FIG. 3 A is graph depicting catalyst activity.

FIG. 3B is graph depicting catalyst activity.

FIG. 4A is graph depicting catalyst activity.

FIG. 4B is a graph depicting formaldehyde decomposition with and without water.

FIG. 5A is graph depicting catalyst activity.

FIG. 5B is graph depicting catalyst activity.

FIG. 6A is graph depicting a series of DRIFT spectra as a function of time. FIG. 6B is graph depicting spectra of a carbonate species (top spectrum) and adsorbed water (bottom spectrum).

FIG. 6C is schematic showing different reaction intermediates.

FIG. 7 is graph depicting a series of DRIFT spectra as a function of time.

FIG. 8A is a graph depicting formation of stable intermediates and treatment with oxygen.

FIG. 8B is a graph depicting a reaction barrier.

FIG. 9 is graph depicting formate production.

FIG. 1 OA is a graph depicting oxidation of surface intermediates to C0 ₂ .

FIG. 10B is graph depicting a series of DRIFT spectra as a function of temperature.

FIG. 11 A is graph depicting an XPS spectrum of a surface.

FIG. 1 IB is graph depicting an XPS spectrum of a surface.

FIG. 12A is graph depicting desorption of HCHO.

FIG. 12B is graph depicting a series of DRIFT spectra.

FIG. 13 is graph depicting recovery of surface sites.

FIG. 14 is three graphs depicting a series of DRIFT spectra as a function of time.

FIG. 15 is a graph showing decomposition pathway for carbonates species via competitive adsorption.

FIG. 16A is a graph depicting a series of DRIFT spectra as a function of time.

FIG. 16B is a graph depicting a series of DRIFT spectra as a function of time.

FIG. 17 is graph depicting catalyst activity.

DETAILED DESCRIPTION

Formaldehyde is a component of indoor air that has been known to cause respiratory issues and is a potential carcinogenic. Various adsorbent, catalytic, and photo-catalytic strategies have been devised to eliminate formaldehyde at room temperature. Nobel metal catalysts on supported oxide such as Pd/Ti0 ₂ , Ag/HMO, Pt/Ti0 ₂ have been used for oxidation of

formaldehyde to C0 ₂ at room temperature. The high cost of Platinum Group Metal (PGM) catalysts limits the widespread use of these materials for commercial purposes. While various oxides of manganese and cobalt have been explored as an alternative to the PGM catalysts, these catalysts are primarily active in the 75-125 degree C range. See, FIG. 1, which is a comparative plot of activities of the current classes of PGM catalysts (orange) and Manganese Oxide catalysts (blue). The PGM catalysts are active at room temperature while Manganese Oxide catalyst show limited TOF at 100 degrees C. Transition metal-oxide catalysts like Mn0 ₂ and C0 ₃ O ₄ have been found to be active for HCHO oxidation. Apart from being inexpensive, the oxide catalysts also have wider operational temperature windows. However, the activity of oxide catalysts is much lower compared to that of the noble metal catalysts as seen in FIG. 1. Composite oxides, alkali metal addition, and transition metal doping are some commonly used strategies to improve the activity. In this regard, Mn0 ₂ catalysts offer flexibility in terms of controlling the crystal structure, tunnel effects, and morphology. CO/C0 ₂ are the common end products after successive oxygenation and dehydrogenation steps accompanied by water formation.

Room temperature operation of both the PGM and manganese oxides catalysts is limited by poisoning of the surface by reaction intermediates that lead to deactivation. Previous studies have shown the formation of dioxymethylene (DOM), formate, and carbonate species being the reason for deactivation at room temperature. However, most of the studies in literature have focused on inlet formaldehyde concentrations ranging from few tens to hundreds of ppm. On the other hand, typical indoor air formaldehyde levels are in the range of 0.2-0.5 ppm. Such a large spread in the inlet concentration can lead to different reaction kinetics and observance of various intermediates.

Further, water vapor has been shown to enhance the kinetics of C0 ₂ production at low relative humidity. In the field of heterogeneous catalysis, it has often been suggested that the formation of active 'hydroxyl species' is responsible for this enhancement. However, there is no consensus on the nature of the active species for C0 ₂ formation at room temperature and the cause of poisoning at room temperature. The reaction mechanism and the cause for deactivation on manganese oxide catalyst are still unclear.

Herein, the reaction mechanism has been explored at different surface coverage in order to understand the coverage effects. The nature of active species was further evaluated by understanding the role of adsorbed 0 ₂ and water vapor in the gas stream. Finally, different strategies are discussed for the regeneration of a poisoned surface.

A method of oxidizing an organic contaminant in a gas can include providing a catalyst, exposing the catalyst to the gas including the organic contaminant and oxygen without adding heat, and at selected times, exposing the catalyst to water, oxygen or heat, or combinations thereof, for a regeneration period. The catalyst can deactivate during use. Thus, at some time during use, regeneration can be effective to increase the activity of the catalyst. For example, at a time during catalyst use, the catalyst efficiency can drop by 5%, 10%, 20%, 30%, 40%, 50% or more. This decrease in activity can be monitored to trigger a regeneration cycle, or the timing can be designated through a study of the catalyst and the surrounding system to establish a designated time for regeneration, which can be a selected time. The selected time for

regeneration can be, for example, 30 to 60 minutes after the catalyst use started. Shorter or longer selected times can be appropriate depending on the catalyst and organic contaminant load. For example, the regeneration time can be 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours or longer. In other examples, the regeneration time can be 30 minutes, 20 minutes, 15 minutes, 10 minutes, 5 minutes or less. The exposure of the catalyst to water or heat, or both, for the regeneration period can reactivate the deactivated catalyst surface, thereby improving the ability of the catalyst to oxidizing the organic contaminant. In certain circumstances, the method can include exposing the catalyst to water for the regeneration period, exposing the catalyst to heat for the regeneration period, or exposing the catalyst to water and heat for the regeneration period. Optionally, oxygen can be provided to the catalyst before, during or after exposure to the water and/or the heat.

The gas containing the organic contaminant can be air or a carrier gas such as nitrogen, argon, or helium, with added oxygen. The added oxygen can be provided at a concentration of up to 5%, up to 10%, up to 20%, up to 30% or up to 50%.

The regeneration period is the period of time that is needed to bring the catalyst activity level for oxidizing the organic contaminant back to an activity level of the catalysis prior to use. For example, the regeneration period can be greater than 1 minute, greater than 2 minutes, greater than 5 minutes, greater than 10 minutes, greater than 15 minutes, greater than 20 minutes, greater than 30 minutes, or greater than 60 minutes. The catalyst can be active for at least 40 to 60 minutes prior to regeneration.

The exposure to water can include exposing the surface to water vapor in a gas. The exposure to water can be simultaneous with the gas with the organic contaminant or can be in a second gas flow with air or another non-reactive carrier gas. The water concentration can be 0.1%, 0.5%, 1.0%, 1.5%, 2%, 3%, 4% or 5%. The catalyst can be effective at room temperature, thus, the exposure to heat can include heating the catalyst to a temperature above room temperature. This regeneration temperature can be 30 to 125 degrees C, preferably 40 to 100 degrees C, for example, 50 to 90 degrees C or around 60 degrees C.

The organic contaminant can be formaldehyde, acetaldehyde, methanol or formic acid.

For example, the organic contaminant can be formaldehyde. The content of organic contaminant in the gas can be 1 ppm to 1,000 ppm, for example, 10 ppm to 100 ppm.

The catalyst can include manganese oxide, titanium dioxide, or gold-titanium dioxide. For example, the catalyst can include amorphous manganese oxide, manganese-cerium oxide mixture, alpha-manganese oxide, beta-manganese oxide, gamma-manganese oxide, or delta- manganese oxide.

The regeneration cycle can be repeated in a periodic manner (i.e., every 30 minutes, every 60 minutes, every 90 minutes) during catalyst use. Alternatively, regeneration can be affected as needed.

In one aspect, a system for oxidizing an organic contaminant in a gas can include a catalyst and a regeneration system that implements the regeneration steps described above during catalyst use. The regeneration system can be operated manually or in an automated manner.

Examples

Catalysts Powders

Partially amorphous manganese oxide and manganese-cerium oxide mixture catalysts were obtained from BASF Corporation.

Activity Measurements

Activity measurements were conducted in a plug-flow reactor. The catalyst was loaded into a quartz tube (6mm id) with silica as a support. A total flow rate of 250 ml min ^"1 was used for all the plug flow reactor experiments and the flow rate of individual gases, 50ppm HCHO (50ppm/ N ₂ ), CO (50ppm/N ₂ ), C0 ₂ (50ppm/N ₂ ), 0 ₂ (25%0 ₂ / N ₂ ) and N ₂ gases was controlled using different sets of mass flow controller. A water bubble was attached to the N ₂ line to introduce water vapor. The gas composition was analyzed using a FTIR gas cell attached to a Bruker Vertex 70 instrument. 64 Scans were collected at a resolution of 4cm ^"1 . See, FIG. 2, which depicts a schematic of the experiment setup used in this study, which included a plug flow reactor equipped with mass flow controllers was used for testing the catalyst with the gas concentrations being monitored by a FTIR gas cell.

Diffuse Reflectance Infrared Transmission Spectroscopy (DRIFTS) measurements were carried out using an in-situ Praying Mantis setup. Amorphous manganese oxide powder and KBr were mixed in different ratios in order to access different surface coverage. At each coverage, the role of adsorbed 0 ₂ and co-adsorbed H ₂ 0 was evaluated by purging the catalyst with HCHO and 0 ₂ under dry and wet conditions for 2hr. For these measurements, a background spectrum was taken under a flow of N ₂ at room temperature. 128 scans with a resolution of 1 cm ^"1 were obtained.

Computational Methods included density functional theory (DFT) calculations that were performed using the vasp ab-initio package using the PBE functional and PAW projectors. The cutoff energy was chosen as 500 eV. A 6x6x1 Monkhorst-Pack73 k-point sampling for the super cell studied was used. The Gibbs free energies of reaction intermediates were computed after correcting for ZPE and vibrational entropy. For the surface, the contributions of all atoms beyond the fully reduced surface to the ZPE and vibrational entropy terms were considered.

Formaldehyde can be eliminated at room temperature. The amorphous Mn0 ₂ catalyst showed the highest sustained conversion to C0 ₂ , 1.4 ppm of C0 ₂ at the end of 120 min as seen in FIG. 3A. The Mn0 ₂ -Ce0 ₂ showed larger activity in the first 20 min but the activity drop as the experiments progressed. Interestingly, complete elimination of formaldehyde was observed for all the catalyst in this duration as the remaining HCHO was adsorbed onto the catalyst surface. For C0 ₂ production, the unsaturated catalyst surface was heated and the resulting light off curve was plotted as a function of temperature. In the investigated temperature regimes, the amorphous sample was the most active, with the α-Μη0 ₂ showing comparable activity (FIG. 3B). Based on the results obtained in FIGS. 3 A-3B, amorphous Mn0 ₂ was chosen as the model system to investigate the reaction pathway. All the results discussed henceforth shall be pertaining to that of amorphous MnO _x . Referring to FIGS. 3A-3B, comparative behavior of C0 ₂ conversion is shown at room temperature (FIG. 3 A) and at higher temperature (FIG. 3B). Circles indicate C0 ₂ content while the diamond symbols are for HCHO. The surface was unsaturated when it was heated up. Under the investigated reaction conditions, it was seen that formaldehyde is eliminated instantaneously. Upon exposure of water, the C0 ₂ production was enhanced (FIG. 4A) and the breakthrough time (i.e., the time it takes for a detectable amount of the undesired chemical to pass the catalyst) was higher than in the dry conditions. The role of water in improving the catalyst lifetime has been shown in a number of earlier studies. However, the deactivation of sample could not be avoided at this coverage. Referring to FIGS. 4A-4B, the promotional role of water increasing the C0 ₂ conversion at room temperature and in prolonging the lifetime of the catalyst at room temperature is shown under the following reaction conditions: lOOmg Μηθχ, 250ml min ^"1 , 20ppmHCHO/10%O ₂ (grey) and 20ppmHCHO/10%O ₂ /1.8%H ₂ O (black) and (b) 50mg, 250ml min ^"1 , 20ppmHCHO/10%O ₂ (grey) and

20ppmHCHO/10%O ₂ /l .8%H ₂ 0 (black). Changing the loading of the catalyst can alter the breakthrough time, as shown, for example, in FIG. 5A. The C0 ₂ generated also increases with time and drops as the sample is saturated, as shown, for example, in FIG. 5B. Referring to FIGS. 5 A-5B, the graphs depict the change in the breakthrough time as a function of catalyst loading (FIG. 5 A) and evolution of HCHO and C0 ₂ as a function of exposure to 20ppm-10%O ₂ (FIG. 5B).

In order to understand the deactivation process, reaction intermediates were identified using DRIFTS. FIG. 6 A depicts the evolution of DRIFT spectra as a function of time after exposure to 20 ppm HCHO/10 0 ₂ at room temperature. The peaks at 2896, 2852, 1561, 1382, 1342 cm ^"1 were assigned to bi-dentate formate species. The band of peaks in the range of 1000- 1200 cm ^"1 ranges were assigned to the dioxymethylene species (DOM). The assignment of the peaks at 1420 and 1320 is unclear, as these peaks have been assigned to monodentate carbonate, hydrogen bicarbonate and monodentate formate species. In order to clarify these peak

assignments, the sample was exposed to C0 ₂ . Based on the spectra obtained after C0 ₂ interaction in FIG. 6B, which shows formation of carbonate species by water after C0 ₂ treatment (top spectrum) and adsorbed water (bottom spectrum) at room temperature, the peaks at 1520 and 1320 have been assigned to monodentate carbonate species. The peaks at 1639 cm ^"1 and the broader band at 3500 cm ^"1 were assigned to the OH stretching and HOH bending of water based on the adsorption of water. The identified reaction intermediates are shown in FIG. 6C.

The reaction mechanism was evaluated at three different 'flow rate/catalyst loading' in order to access different surface coverages. Table 1 summaries the three reaction conditions spanning three orders of magnitude to flow rate/gram used in this study. These conditions will be referred to as high, medium and low coverage cases. Table. 1 : Summary of the reaction conditions employed in this study to access wide ranging surface coverage

The species that form at each coverage after an exposure of HCHO and 10%O ₂ for 2hr is shown in FIG. 6A and FIG. 7. FIG. 7 shows the evolution of reaction intermediates after exposure for 2hrs at the low (left) and medium (right) conditions. The reaction parameters are as specified in Table 1. At low coverage, the surface is dominated by the presence of monodentate carbonate and a smaller amount of bi-dentate formate. As the exposure increases, the formate peaks grows at a faster rate than the carbonate and eventually at medium coverage the surface has formate and carbonate. On further increasing the HCHO exposure, the surface is dominated by the formate intermediate while peaks corresponding to DOM are also observed. These results indicate the oxidation kinetics sensitivity of the three major intermediates (carbonate, formate, DOM) towards surface coverage.

The reaction mechanism can be elaborated based on the information gathered in these experiments. For example, catalyst deactivation can be described. After 2hr of HCHO exposure, the decomposition of the intermediates was tracked using the integrated areas of the peaks. It is clearly seen that under dry conditions, the formed intermediates are extremely stable and do not decompose easily. Referring to FIG. 8A, the graph shows the evolution of stable formate species after treating the catalyst with HCHO for 120min followed by 0 ₂ treatment. Referring to FIG. 9, the integrated intensity of formate peak is shown after exposure to 20ppm/10%O ₂ for 20, 40, 80 min followed by purging with 10%O ₂ . The intensity does not drop. Surface poisoning due to either formate or carbonate species limits the durability of catalysts at room temperature. It has been remarked previously that the hydroxyl poisoning is responsible for the deactivation of the surface. However, as found here, the formate does not decompose at low coverage, it can be concluded that the activation barrier for the formate is large at room temperature. This is the limiting step rather than the loss of hydroxyls. This was further investigated using Nudged Elastic Band (NEB) theory where the activation barrier was estimated using a set of 7 images from the formate to a C0 ₂ adduct. The large barrier of ~leV barrier obtained in this case is consistent with barrier obtained in literature. FIG. 8B represents a reaction barrier computed using DFT for formate oxidation to C0 ₂ . The large activation barrier arises out of the perpendicular geometric nature of the H atom that lies perpendicular to the surface and the need to break the strong M-Cv bond to from C0 ₂ .

A cycling measurement has been carried out to verify the buildup of formate species at high coverage. While the DOM species decompose to form formate, the oxidation of formate to C0 ₂ does not occur. With each cycle, it can be seen that the reaction kinetics slow, eventually leading to deactivation.

Temperature-assisted catalyst regeneration was investigated. After the catalyst has been deactivated at room temperature (see, FIG. 4), heating up the catalyst in an oxygen rich atmosphere (10% 0 ₂ ) oxidizes all the reaction intermediates to C0 ₂ . Referring to FIG. 10A, the oxidation of surface intermediates to C0 ₂ is shown as a function of temperature using the reactor setup. The corresponding DRIFTS spectra are shown in FIG. 10B.

The XPS spectrum of the saturated surface also showed a C=0 (FIG. 11 A) peak which disappears upon heating to 120 degrees C. (FIG. 1 IB) Referring to FIGS. 11 A- 1 IB, the graphs show C Is spectra of the saturated surface (left) and the surface after heating under a continuous gas flow mixture of 20ppm HCHO-10%O ₂ (right). However, heating up of a completely saturated surface at 50 degrees C also leads to desorption of weakly adsorbed HCHO. (0 min, FIG. 12 A). Interestingly, after purging the saturated samples with 10%O ₂ at room temperature for 60 and 600 min prior to heating leads to a decrease in the amount of HCHO desorbed as seen in FIG. 12A. In a corresponding DRIFTS experiments, the decrease in the peaks in the DOM region indicate that a part of the DOM converts to formate while the rest of it desorbs as HCHO. (FIG. 12B) Hence, the high temperature cleaning treatment should include a purging step to avoid HCHO desorption. Referring to FIGS. 12A-12B, graphs show desorption of HCHO at 50°C after the sample has been saturated at room temperature with 20ppmHCHO/10%O ₂ and purged with 10%O ₂ for 0, 60, and 600min (FIG. 12 A) and DRIFTS spectra of the samples that was heated to 50°C after Omin of 10%O ₂ purging (FIG. 12B). Additionally, room temperature purging with 10%O ₂ after saturation with HCHO also leads to the recovery of surface sites. Referring to FIG. 13, recovery of surface sites can occur after the sample has been purged for 6 hours in 10%O ₂ post saturation.

Water-assisted catalyst regeneration was investigated. Another methodology to regenerate the surface once the catalyst surface has been poisoned by the presence of formate and carbonate species, water vapor is via the introduction of water vapor. In order to probe the role of dosed water in aiding the oxidation process, the role of water was investigated under the three coverage conditions. Referring to FIG. 14, after the catalyst has been exposed to

HCHO/0 ₂ at room temperature, the catalyst surface can be regenerated by a H ₂ 0/0 ₂ treatment. In each case, the sample was exposed to 2hr of HCHO/0 ₂ followed by repeated treatment of H ₂ 0 and 0 ₂ . It can be seen that at all coverages, the surface can be renewed. Elimination of formaldehyde via formate formation and subsequent water aided oxidation can be used as practical solution to catalyst deactivation.

Universality of role of water in formate oxidation was investigated. It has been previously remarked that water plays three distinct roles during heterogeneous catalysis. Firstly, water can displace carbonate species via competitive adsorption. Secondly, molecular water can stabilize reaction intermediates via hydrogen bonding. Thirdly, water can dissociate and provide active hydroxyl species.

In the case of formaldehyde oxidation, the role of water is intricate. The role of water in eliminating any carbonate species that formed was probed by exposing the catalyst to C0 ₂ . The carbonate peaks thus formed ceased to exist upon further water exposure. Referring to FIG. 15, decomposition pathway for carbonates species via competitive adsorption with C0 ₂ , water and nitrogen are shown.

The role of water in oxidizing formate intermediate is non-trivial. Formate is a common poisoning species observed during oxidation of methanol, formic acid and formaldehyde oxidation. Through DFT, Miller et al. (K. L. Miller, J. L. Falconer, and J. W. Medlin, Effect of water on the adsorbed structure of formic acid on Ti02 anatase(lOl), J. Catal. 278 321-28 (2011), which is incorporated by reference in its entirety) have shown that molecularly adsorbed water is required for the oxidation of bi-dentate formate through a mono-dentate intermediate during formic acid oxidation. Recently, Bokhoven and co-workers (M. Sridhar, D. Ferri, J. A. von Bokhoven, and O. Krocher, Water-assisted oxygen activation during gold-catalyzed formic acid decomposition under SCR-relevant conditions, J. Catal. 349 197-207 (2017), which is incorporated by reference in its entirety) have argued that a OOH intermediates is necessary to oxidize formate. In this case, formate formation could not be avoided in the case of co-adsorbed water (different from dosing water on a saturated surface in or a surface that been pre-covered with H ₂ 0. See, for example, FIG. 15. DRIFT spectra obtained in the case of pre-dosed water (see FIG. 16 A) and co-adsorbed water (see FIG. 16B) show the formation of formate and DOM species. This indicates that a critical concentration of water and oxygen species is required at a specific stage of the reaction in order to oxidize the formate intermediate.

In conclusion, the results from coverage studies indicate that dehydrogenation kinetics are sensitive to the surface coverage. While exploring different surface coverage, it has been observed that DOM forms and gradually dehydrogenates to formate at high coverage. This is why different studies across literature have reported different reaction mechanisms. Further, the role of oxygen/water and temperature in regeneration of the surface has been discussed. The implications of the role of water are far reaching as such a technique can be used for methanol, formic acid and formaldehyde oxidation reactions. For example, a similar methodology has been applied for acetaldehyde oxidation, where the intermediates formed during acetaldehyde oxidation can be cleaned away by a water treatment. Referring to FIG. 17, a decrease in the acetate and formate intermediates can occur after exposure to water and oxygen at room temperature.

Other embodiments are within the scope of the following claims.