Patent Application
20240124379
Isobutane is an important commodity chemical precursor, conventionally produced in refineries via cracking and isomerization. Natural gas liquids, including those extracted along with shale gas, provide a major alternate source for isobutane. Isobutane is primarily used as a feedstock to produce tent-butyl alcohol (TBA) and tent-butyl hydroperoxide (TBHP) by oxidation with O2. Isobutylene, produced via either isobutane dehydrogenation or TBA dehydration, is an important alkylating reagent for production of chemical and pharmaceutical precursors. Isobutylene is also a monomer for the production of synthetic resins and butyl rubber.
Industrial isobutane oxidation uses O2 as oxidant and is performed at 120-140° C. and 2.5-3.5 MPa, with ˜3-10% O2 in the gas phase. At these conditions, isobutane exists as a liquid and O2 is bubbled through the liquid phase. The vapor phase is typically maintained above the upper flammability limit for isobutane. Free radical initiators such as TBHP and di-tent-butyl peroxide (DTBP) are usually co-fed to reduce the induction period. At these conditions, the isobutane conversion is ˜4-8% per hour, producing nearly equimolar amounts of TBA and TBHP at approximately 96% selectivity. The TBHP is used as an oxidant for propylene oxidation to propylene oxide, producing TBA as coproduct.
For isobutane oxidation with O2, the initiation mechanisms in the liquid phase and gas phase are similar. However, the product selectivities strongly depend on the relative concentrations of the substrate and oxidant. Compared to liquid phase oxidation, isobutane oxidation in a gas phase requires longer induction periods even at 260-360° C. and produces a wider spectrum of products. This is caused by side reactions, including over-oxidation of terminal carbons and C—C bond cleavage.
Provided herein are processes for oxidizing alkanes in the liquid phase using ozone. To illustrate the present processes, the Examples, below, demonstrate isobutane ozonation to tent-butyl alcohol (TBA) with 85%-90% selectivity in a biphasic gas-liquid process at ambient temperatures (16-18.5° C.) and mild pressures (˜0.6 MPa). The observed TBA selectivity is much higher than the value reported for either gas-phase ozonation or air oxidation in liquid isobutane. In addition, it is found that O3 utilization (defined as moles of TBA produced per mole of ozone consumed) to form TBA is maximized (˜1.5 moles TBA/mole O3 consumed) by adding certain amounts of protic compounds (e.g., water or alcohols) in the reaction mixture, and by maintaining a certain O3 partial pressure in the gas phase that limits the liquid phase O3 concentration to avoid O3 decomposition. The relatively low CO2 formation (<6% selectivity at optimal conditions) suggests that TBA formation occurs predominantly via hydrotrioxide formation with high atom economy.
In embodiments, a process for oxidizing an alkane comprises combining an alkane and ozone in a liquid phase medium comprising a protic additive under conditions sufficient to oxidize the alkane to products comprising a hydroxylate.
In embodiments, a process for oxidizing an alkane comprises combining an alkane and ozone in a liquid phase medium comprising a protic additive under conditions sufficient to oxidize the alkane to products comprising a hydroxylate, wherein the liquid phase medium is free of added CO2, and adding the protic additive to the liquid phase medium prior to oxidizing the alkane.
Other principal features and advantages of the disclosure will become apparent to those skilled in the art upon review of the following drawings, the detailed description, and the appended claims.
Illustrative embodiments of the disclosure will hereafter be described with reference to the accompanying drawings.
The present disclosure provides processes for oxidizing alkanes in the liquid phase using ozone. By “liquid phase” it is meant that the relevant reactions take place in the liquid phase, as opposed to the gas phase. Ozone is a useful oxidant for hydrocarbons, but significant challenges exist related to its effective utilization. For example, ozone decomposes quickly in the liquid phase. Moreover, intermediate compounds are often much more reactive towards ozone than the starting alkane. Thus, a key challenge in liquid phase ozonation of alkanes is the optimization of ozone utilization towards the desired reaction rather than substrate over-oxidation and ozone decomposition.
In embodiments, a process for oxidizing an alkane includes combining an alkane with ozone in a liquid phase medium. The ozone may be provided as a feed gas mixture, e.g., O3 in O2 or O3 in air. An inert gas, e.g., N2, Ar, etc., may be included in the feed gas mixture. This is useful to keep a vapor phase present above the liquid phase medium outside of its flammability envelope at the temperature and pressure being used in the process. The liquid phase medium generally also comprises a protic additive. This feature is derived from the serendipitous finding that the use of the protic additive can lead to unexpectedly high selectivity for a hydroxylate product as well as unexpectedly high ozone utilization. By “hydroxylate product,” it is meant the product corresponding to the alkane undergoing the oxidative conversion with a single C—H bond of the alkane replaced with a single C—OH bond. These findings are illustrated in the Examples, below, for the oxidation of isobutane to the hydroxylate product, tent-butyl alcohol, with 85% to 90% selectivity at mild temperatures and pressures.
The present processes may be used with a variety of alkanes. The alkane may be linear, branched, or cyclic. In embodiments, the alkane is linear or branched, and not cyclic. The alkane may include various numbers of carbons, e.g., one, two, three, four, five, six, seven, eight, etc. In embodiments, the alkane is a C1-C20 alkane, a C1-C15 alkane, a C4-C20 alkane, a C4-C18 alkane, a C4-C16 alkane, a C4-C15 alkane, or a C4-C12 alkane. Specific illustrative alkanes include methane, propane, n-butane, and isobutane. However, in embodiments, the alkane is not methane. In embodiments, the liquid phase medium does not comprise methane. Combinations of different types of alkanes may be used. In the present processes, the alkane is generally in its liquid form (i.e., a liquid at the temperature and pressure used during the oxidative conversion).
Similarly, the present processes may be used with a variety of protic additives. The protic additive is a compound that is capable of hydrogen bonding by comprising at least one hydrogen atom bound to an electronegative atom. Without wishing to be bound to any particular theory, it is believed that the protic additive stabilizes a hydrotrioxide intermediate generated during the oxidative conversion reaction. (See
The protic additive generally refers to a compound which is added to the liquid phase medium prior to carrying out the oxidative conversion. This is by contrast to the protic additive being a product of the oxidative conversion reaction (e.g., H2O or a hydroxylate product). In embodiments, the protic additive is not a product of the oxidative conversion reaction. In embodiments, the protic additive is not a hydroxylate product, e.g., tent-butyl alcohol, of the oxidative conversion reaction. In embodiments, the present process comprises adding any of the disclosed protic additives to the liquid phase medium prior to the oxidative conversion to produce a product from the alkane.
Although no catalyst is required for the present processes, in other embodiments, a catalyst may be used. If a catalyst is used, the catalyst may contain transition metals or combinations, such as a first-row transition metal (e.g., Fe, Co, Ni) or a platinum group metal, e.g., Pd, Pt. The catalyst may be supported transition metal catalyst in which the selected transition metal is incorporated into or deposited on a surface of a metal oxide substrate, e.g., an oxide of aluminum, silicon, titanium, magnesium, cerium, zirconium, etc. Illustrative metal oxide substrates include silicate and zeolite. Mesoporous silicates such as KIT-5, KIT-6, SBA-16, TUD-1 may be used.
As described above, the oxidative conversion reaction is carried out in the liquid phase medium. The liquid phase medium comprises the selected alkane and the ozone. As noted above, the alkane itself is generally a liquid at the temperature and pressure used during the present processes. Generally, an excess amount of the alkane is used. The liquid alkane solubilizes at least some of the ozone therein. The liquid phase medium generally further comprises any of the protic additives as described above. The catalyst, if present, may either be dissolved or suspended in the liquid phase medium. In embodiments, the liquid phase medium comprises one or more alkanes, ozone, one or more protic additives, and optionally, a catalyst. However, such embodiments also encompass the presence of the resulting products of the oxidative conversion in the liquid phase medium.
By contrast to some existing oxidative conversion processes involving hydrocarbons and oxidants such as ozone, the present processes do not require the use of carbon dioxide (CO2). Thus, at least in embodiments, the liquid phase medium is free of, i.e., does not comprise, CO2 (whether the CO2 is in its gaseous or liquid state). These embodiments refer to the absence of any CO2 (whether the CO2 is in its gaseous or liquid state) separately added to the liquid phase medium. These embodiments, however, do not preclude the formation of an amount of CO2 as a product of the oxidative conversion. However, if any CO2 is present in the liquid phase medium as a product of the oxidative conversion, it is generally present at an amount of less than 1 mmol. This includes less than 0.8 mmol, less than 0.6 mmol, less than 0.4 mmol, less than 0.2 mmol, and less than 0.1 mmol. This includes from 0.001 mmol to 0.8 mmol.
The present processes may be carried out in a variety of reactor systems, including batch reactor systems, semi-continuous flow reactor systems, and continuous flow reactor systems.
The conditions used in the present processes to induce oxidative conversion refer to parameters such as the temperature, the total pressure, as well as others further described below. The temperatures and total pressures used are generally mild. In embodiments, the temperature is at least 15° C. Illustrative temperatures include those in a range of from 15° C. to 60° C., 15° C. to 50° C., 15° C. to 25° C., 15° C. to 20° C., and 16° C. to 19° C. The total pressure refers to the total pressure of the vapor phase present above, and in contact with, the liquid phase medium at the selected temperature. This vapor phase comprises ozone, as well as any other gases present in the gas mixture used to deliver the ozone, e.g., O2, air, an inert gas such as N2, or combinations thereof. The vapor phase also generally comprises some amount of the vapor form of the selected alkane. In embodiments, the total pressure is less than 5 MPa, less than 3 MPa, or less than 1 MPa. Illustrative total pressures include those in a range of from 0.45 MPa to 5 MPa, from 0.45 MPa to 3 MPa, and from 0.45 MPa to 1 MPa. These pressures are significantly lower than those used in ozonolysis reactions using liquid CO2 or CO2 expanded liquids.
The conditions also refer to the amount of a protic additive in the liquid phase medium. The amount is generally selected to maximize O3 utilization, which may be quantified as the number of moles of a desired product (e.g., hydroxylate product)/moles of ozone fed. Illustrative amounts of protic additive include at least 0.5 mmol, at least 1 mmol, at least 10 mmol, or at least 20 mmol. In embodiments, the amount is in a range of from 0.5 mmol to 100 mmol or 1 mmol to 100 mmol. The amount of the protic additive in the liquid phase medium may also be referred to as a volume % of the total volume of the liquid phase medium. In embodiments, the amount is in a range of from 0.2 volume % to 5 volume %, from 0.2 volume % to 2 volume %, from 0.2 volume % to 1.5 volume %, or from 0.2 volume % to 1 volume %. The balance of the liquid phase medium may be composed of the liquid alkane(s) (with ozone and any catalyst, if present, dissolved therein). As demonstrated in the Examples, below, O3 utilization in isobutane oxidation towards tent-butyl alcohol formation increased by 1.5-fold as the amount of water increased from 1.38 to 27.65 mmol. (See also,
The conditions also refer to the partial pressure of non-condensable gases in the vapor phase present above, and in contact with, the liquid phase medium at the selected temperature and total pressure. For example, if the feed gas mixture comprises O3, O2, and N2, the partial pressure of non-condensable gases refers to the pressure of these three gases. The partial pressure is generally selected to maximize O3 utilization. Illustrative partial pressures include from 0.02 MPa to 0.24 MPa, from 0.1 MPa to 0.6 MPa, from 0.2 MPa to 0.5 MPa, and from 0.2 MPa to 0.4 MPa. (See also,
The conditions also refer to the mole fraction of ozone in the feed gas mixture. For example, if the feed gas mixture comprises O3 in O2 and N2, the mole fraction refers to the O3/(O3+O2+N2) mole fraction. The mole fraction is generally selected to maximize O3 utilization. Illustrative mole fractions of ozone include from 1% to 5%, 1.5% to 4.5%, from 2% to 4%. (See also,
The products obtained from the oxidative conversion depend upon the selected alkane as well as the conditions used. By way of example, however, isobutane may be oxidized to one or more of the following products: CO2 (a combustion product), formic acid, acetic acid, acetone, hydroxyacetone, tent-butyl alcohol, and tent-butyl hydroperoxide. The present processes may further comprise collecting one or more of the products produced. Other steps include using the collected products, e.g., in chemical reactions to form other products. By way of illustration, tent-butyl alcohol may be dehydrated to form isobutylene.
The present processes may be characterized by a selectivity for a particular product. By selectivity, it is meant ((moles of particular product)/(total moles of products obtained))*100. In embodiments, the selectivity of the hydroxylate product is at least 80%, at least 85%, or at least 90%. For the conversion of isobutane, the hydroxylate product and these selectivities may refer to tent-butyl alcohol. Moreover, the present processes involving the conversion of isobutane may be characterized by the absence of certain products such as formic acid, acetic acid, and tent-butyl hydroperoxide (i.e., selectivity of less than 2%, less than 1%, less than 0.1%, or less than 0.05%). Finally, the present processes involving the conversion of isobutane may be characterized by a CO2 selectivity of no more than 10%, no more than 8%, no more than 6%, or in a range of from 5% to 10%.
The present processes may be characterized by an O3 utilization value which may be quantified as the number of moles of a desired product (e.g., hydroxylate product)/moles of ozone fed. In embodiments, the O3 utilization value is at least 100%, at least 125%, at least 150%, at least 200%, or in a range of from 100% to 300%.
The selectivities and O3 utilization values described above may be reported with reference to a particular set of conditions used in the process.
In this Example, isobutane ozonation in a liquid phase was investigated using a Teflon-lined Parr reactor to avoid metal surface-induced O3 decomposition. The Example is based, at least in part, on the insight that liquid isobutane, which possesses some compressibility at near-ambient temperatures, may be able to accommodate O3 at higher concentrations as compared to traditional organic solvents. Another insight is the potential to leverage the pressure-tunability of O3 concentration in liquid isobutane. Another advantage of liquid phase operation is the ability to easily maintain the isobutane-rich vapor phase above the upper flammability envelope. It is demonstrated herein that isobutane undergoes facile ozonation in its liquid phase at near ambient temperature, directly producing TBA at ˜90% selectivity. The experimental results strongly suggest that isobutane hydroxylation to TBA proceeds via the formation of the isobutane hydrotrioxide intermediate that, unlike ozonides that form during alkene ozonolysis, decomposes easily to form TBA without major release of energy.
Reactions with ozone should be handled in a well-ventilated area. Condensation of ozone (at ˜112° C.) must be avoided when trapping the volatile organics using a coolant bath, to prevent potential detonation vaporization or ignition.
Isobutane most likely undergoes hydroxylation via hydrotrioxide intermediate with much lower decomposition energy compared to ozonides formed during alkene ozonolysis. However, during initial tests, the products must be handled with care because of the potential formation of explosive and shock-sensitive compounds, such as acetone peroxide from acetone and H2O2.
As for the O2+O3 gas feed in the reaction, the mixture composition should be above the upper flammable limit of 48% at low pressures. The O2/O3 gas feed was diluted by N2 to shrink the flammable envelope for parametric studies involving higher pressures than a few bar. Dilution by 50% N2 was estimated to shift the upper flammable limit from 48% to 31%.
Peng-Robinson model with Wong-Sandler mixing rules as implemented in Aspen Plus was also used to estimate the worst-case scenario with the gas phase as a closed system. While forming CO2 and H2O may release the most chemical energy, forming CO and H2 will generate the largest number of molecules. The upper bound for gas phase pressure was estimated to be 7.5 MPa (forming H2 and CO2), which is substantially less than the Parr reactor's rating of 20 MPa.
Prior to any run, the reactor was evacuated at ˜80° C. under vacuum for ˜45min, in order to remove any residual water, acetone or tent-butyl alcohol from the reactor head. The reactor body was isolated from the liquid phase by a Teflon sleeve, to minimize ozone decomposition on the metal's (dense oxide) surface. The water used in these runs was distilled to remove compounds exhibiting acidity (pH=4-6) from the commercial supplies.
The saturation vapor pressures of liquid pure components involved in this Example at various temperatures were reviewed. Temperatures as low as −60-50° C. could condense the majority of isobutane vapor. In the −20-−10° C. range, the vapor pressure of isobutane is at least 10-times greater than those of other compounds, which allows vaporizing isobutane without losing any more than 1% of either solvents (except CH2Cl2) or products. In the non-ideal mixtures, the interactions between polar solvents and oxygenates further decrease the volatility of the products. These considerations determined the temperatures in the procedures for sample collection and analysis in the following sections.
Batch runs were carried out at the ambient temperature in the laboratory, 17-18.5° C., (isobutane saturation pressure=0.27-0.28 MPa). The external reservoir was filled with an O3/O2 mixture containing 4% mol O3/mol (O2+O3). The desired amount of water was optionally loaded into the sample loop and flushed into the reactor along with 20-40 mL liquid isobutane (Matheson, Ultra High Purity), which was metered by a Teledyne ISCO pump at 17° C. and 0.86 MPa (density=7.576 mol/L). The reactor content was stirred at 1000 rpm.
Seven cycles were carried out as follows: The O3/O2 mixture was charged to raise the reactor pressure to 0.42 MPa (introducing ˜0.08 mmol O3), so that the gas phase contains ≥65% isobutane, above the upper flammable limit of 48%. After ˜30 min, the reactor was depressurized until the reactor pressure was close to isobutane saturation pressure. The reactor temperature was allowed to warm back to ambient temperature for ˜10 min.
After reaction, the reactor and its contents were cooled to −17° C. to reduce the vapor pressures of the products. The cold reactor was then opened to add one cold solvent among methanol (Fisher, 99.9%), dichloromethane (Acros Organics, 99.9%), and n-propyl acetate (Acros Organics, 99%). The isobutane was slowly vaporized leaving behind liquid products dissolved in the solvent.
The reaction was carried out similarly to usual batch runs. At the end of the desired batch time, the reactor was quickly depressurized to isobutane saturation pressure at the reaction temperature, releasing the gas into a gas bag. Then the reactor was immediately moved into the freezer, using liquid nitrogen to accelerate the quench.
The preparation was similar to that for the batch run. The O3/O2 mixture was diluted with N2 to a molar ratio of (O3+O2)/N2=1, to shrink the flammability envelope for gas-phase mixtures in the reactor.
The reaction was initiated by pressurizing the reactor with the O3/O2/N2 mixture, followed by immediately opening the outlet valve and tuning a downstream rotameter to the desired set-point. A separate setup with mass flow controllers was used to calibrate the set-point for the desired flow rate with each effluent composition. The effluent passed through a condenser immersed in ethylene glycol/water mixture cooled to −60-−50° C. by liquid nitrogen, in order to trap isobutane and concentrate the non-condensables (CO2, O2, N2) to be collected in a gas bag. The isobutane lost in the reactor effluent was periodically replenished.
Following the reaction, the reactor was opened to recover the liquid product. The reactor was then rinsed with a solvent (mixed with a predetermined mole fraction of internal standard, hexane). The quantities of products sticking to the Teflon sleeve were calculated from the unrecovered quantities of solvent and internal standard used for rinsing.
The products identified in the product mixtures from this Example include carbon dioxide, formic acid, acetic acid, acetone, hydroxyacetone, tent-butyl alcohol and tent-butyl hydroperoxide.
Gas chromatography analyses of liquid samples were done on an HP 5890 GC with its FID connected to an RXI guard column (5 m×0.53 mm) and an HP-PLOT/Q column (30 m×0.53 mm×40 μm), with injector heated to 200° C. and detector heated to 300° C. A constant ramping oven program from 30° C. to 240° C. at 10° C./min was adopted for (a) resolution between isobutane and methanol peaks, (b) resolution between hexane, propyl acetate and acetic acid, and (c) aiding product identification by retention indices. The gas and liquid samples were also injected on a Shimadzu GC equipped with a 2 m ShinCarbon packed column connected to its TCD, and a HP-PLOT/Q column (30 m×0.53 mm×40 μm) connected to its FID, mainly for analyzing CO, CO2 (including their potential formation in the GC from any formaldehyde and formic acid in the liquid samples) and isobutane.
As a destructive analysis method, gas chromatography could decompose small quantities of formic acid, formaldehyde, and tent-butyl hydroperoxide, if any are present in the product mixtures.
HCOOH→CO2+H2 and HCHO→CO+H2
Tert-butyl hydroperoxide may undergo complex reactions in the GC instrument forming products (including methanol, acetic acid, and numerous unidentified products) distinguishable from chromatograms of samples containing negligible tent-butyl hydroperoxide. The peroxide/alcohol ratios of several product mixtures analyzed by GC and NMR methods were compared. GC analyses of samples with peroxide/alcohol <10−1 should be avoided, as suggested by the deviation of the samples with low peroxide/alcohol ratios from those with high peroxide/alcohol ratios. On the other hand, when peroxide/alcohol is greater than 102, the alcohol formation could also interfere with the analysis.
Formic acid, acetic acid, and tent-butyl hydroperoxide were likely minor byproducts existing in all product mixtures. However, only a few runs produced quantities that were detectable with confidence (selectivities above 0.01% level). For runs that produced tent-butyl hydroperoxide or formic acid substantially above their detection limits, NMR spectra were used to determine their quantities in the product mixture.
In addition to GC results, 1H NMR spectra of samples in deuterated dichloromethane (CD2Cl2, 99.8% D, Acros Organics) confirmed undetectable methanol. Thus, deuterated methanol (CD3OD, 99.8% D, Sigma-Aldrich) was used for better resolution between signals, with dichloromethane (CH2Cl2) as internal standard. Neither formaldehyde nor its hemi-formal with methanol could be detected by proton NMR. Another sample using deuterated acetonitrile (CD3CN, 99.8% D, Sigma-Aldrich) was used to confirm negligible tent-butyl hydroperoxide formation.
1H NMR experiments showed that the tent-butyl hydroperoxide in the product mixture decomposed over extended periods of time. Therefore, all spectra for observing hydroperoxide were collected within 30 min after the end of reactions.
For the ozonation of isobutane only, the formation of formic acid, acetic acid, and tent-butyl hydroperoxide was negligible. Therefore, the molar selectivity for product i is defined as follows:
where ni [mmol] is the quantity of product i detected in the product mixture.
For the supplementary experiments with isobutane/tent-butyl alcohol mixtures, the selectivity is defined as the following.
The carbon balance is defined as follows.
where NC,i is the number of carbon atoms in product i; ni [mol] is the quantity of product i; nC,unreacted is the total carbon quantity in unreacted isobutane (iC4) in the cold trap condensed from reactor and in gas bags; nC,0 is the total quantity of carbon in isobutane fed into the system.
As an example, quantities listed in Table 1 for the base experiment with 5.53 mmol water (
Preliminary experiments of isobutane ozonation at −30° C. in an open vessel suggested unexpected but beneficial effects of condensed ambient moisture on the reaction. As a result, a setup described above and shown in
Conditions: T=17° C., iC4 40 mL (as liquid at 17° C. and 0.86 MPa in Teledyne ISCO pump), O3 mole fraction=4% in O3+O2, iC4 saturation P˜0.25 MPa, total P˜0.42 MPa (˜0.08 mmol O3 each cycle), gas phase iC4 concentration ≥65% (UFL=48%14), reaction time=30 min each cycle, stirring rate 1000 rpm.
To gain insights into the kinetics of TBA formation, the effect of batch time on product yield was studied via multiple batch runs of various durations under otherwise similar operating conditions as listed in Table 2. The molar product quantities were normalized to the moles of ozone fed in each batch. The end-of-run TBA profile was regressed with a kinetic model assuming a first-order dependence on O3 concentration with isobutane in a large excess. It is implicitly assumed that the O3 concentration in the liquid phase is linearly proportional to the O3 partial pressure in the gas phase. This assumption is justified based on a similar trend observed in liquid CO2. Such an analysis yielded a pseudo-first-order rate constant of ˜0.56 min−1 for O3 consumption. The minor co-products acetone and CO2 show overlapping profiles, suggesting that they resulted from the same pathway. It is proposed that these products occurred via the β-scission of the tert-butoxy radical, which would form a methyl radical and acetone. The methyl radical was most likely converted to CO2. Clearly, the rate of the β-scission pathway is much lower than the parallel pathway to produce TBA.
The remarkable TBA selectivity observed in the batch runs motivated the inventors to investigate semi-continuous ozonation in which ozone was continuously fed into the reactor containing an initial charge of isobutane (˜220 mmol). The experimental procedures are discussed above. Specifically, the effects of water, O3 partial pressure and TBA were evaluated during semi-batch runs at 18.5° C. The total gas phase pressure was maintained at 0.75 MPa, consisting of 0.27 MPa isobutane and up to 0.48 MPa of other gases (O3, O2 and N2). N2 was added as an inert component to maintain the gas phase outside the flammability envelope. The molar (O3+O2):N2 ratio was 1, and the O3 mole fraction in the (O3+O2+N2) mixture was 5%. The partial pressure of the non-condensable gases (O3+O2+N2) was chosen as 0.34 MPa for the base case. The liquid products were analyzed by both GC/FID and 1H NMR methods. The carbon balances were closed within experimental error (a sample calculation is provided in Table 1).
Effect of water: As shown in
Effect of O3 partial pressure: Due to the higher O3 decomposition rates in the liquid phase, O3 partitioning between the gas and liquid phases dictates ozone utilization towards TBA formation. Liquid-phase O3 solubility depends on gas phase O3 partial pressure as well as the properties (density, polarity, etc.) of the liquid phase. The O3 partial pressure may be varied by changing either the total pressure of the gas phase of a fixed composition or the O3 fraction in the gas phase at a fixed total pressure. During these investigations, the initial water added was constant at 5.53 mol.
As shown in
When the partial pressure of the non-condensable gases was held at 0.34 MPa and the O3 concentration was varied, a substantial increase in O3 utilization was observed even at 1% O3 (
While the increased gas phase O3 partitioning increased over-oxidation of gas phase isobutane, the slower O3 decomposition in the gas phase also enhanced overall O3 utilization towards TBA formation. Collectively, these data demonstrate that there is a pressure-tunable optimum at which the O3 utilization is maximized.
In summary, even though gas-phase ozonation is less selective, the gas phase may serve as a desirable reservoir at low temperatures, where O3-based combustion and decomposition are quite slow. On the other hand, increased O3 concentrations in the liquid may promote facile decomposition in the liquid phase, which would compete with the activation of substrate C—H bonds with high bond dissociation energies (320 -450 kJ/mol). With the appropriate choice of temperature and gas-phase partial pressure, the vapor-liquid equilibrium of O3 may be harnessed to optimize its liquid phase concentration, and thereby to minimize both O3 decomposition and substrate over-oxidation in the liquid phase.
Effect of TBA in the initial mixture: The highest quantity of TBA produced in previous runs was approximately 5% of the isobutane fed to the system. It must be recognized that the isobutane is intentionally held as an excess reagent and is therefore limited by O3 availability. In order to assess the effect of TBA accumulation on the product distribution, ozonation of mixtures of TBA+isobutane was evaluated. The net product formation (with initially added TBA subtracted) is shown in
In general, the presence of TBA in the initial mixture promoted O3 utilization, which is similar to the effect of added water. However, the net TBA formation in such reactions decreased. Most likely, TBA underwent oxidation to form acetic acid and formic acid with a combined selectivity of >2%. For comparison, the combined selectivity of acetic acid and formic acid was <0.02% in the isobutane ozonation runs without added TBA. Peroxyacetic acid is a possible intermediate that decomposes to acetic acid. The TBHP detected during the runs with initial TBA added was at least 102 times greater compared to previous runs with initially added water. Two factors could contribute to the observed results: (a) the solvent polarity (and hence O3 solubility) is significantly affected as TBA is more soluble in isobutane than water; and (b) compared to water, the bulkier tent-butyl group could pose hindrance to the formation of hydrogen-bonded cyclic structures from TBA and other molecules containing hydroxyl, hydroperoxyl, or hydrotrioxyl groups.
As noted above, when isobutane ozonation was performed in the presence of water or TBA, the TBA formation was substantially more selective than TBHP formation. These observations suggest a role for these protic components in aiding the decomposition of the alkyl hydrotrioxide intermediate (
Due to the competing O3 decomposition, the TBA formation per mole of consumed O3 cannot surpass unity based on considering only the two pathways for unimolecular hydrotrioxide decomposition (
Byproducts including acetone, acetic acid, formic acid, and CO2 formed during all runs, while hydroxyacetone and methanol were detected only under certain reaction conditions. At low ozone exposure (i.e., at short contact time and low O3 partial pressures), nearly equimolar amounts of acetone and CO2 (
The foregoing results suggest that in order to maximize TBA yield and O3 utilization, reactor operation in which the gas and liquid phases are continuously admitted and withdrawn is needed to control the O3 partial pressure in the gas phase as well the TBA/water accumulation in the liquid phase.
This Example demonstrates facile isobutane ozonation with 85%-90% TBA selectivity in a biphasic gas-liquid process at room temperature (16-18.5° C.) and mild pressures (˜0.6 MPa). The process could be operated safely with a non-flammable vapor phase that is well above the upper flammability limit. In contrast to oxidation with O2 (the conventional industrial process), hydroperoxide formation was drastically reduced. It was suggested that the reduced formation of such products results from an alternate reaction mechanism when protic compounds and O3 co-exist. With O3, conventional free radical initiators like TBHP and DTBP are not required to initiate the reaction. In fact, such free radical initiators promote O3 decomposition. Isobutane most likely undergoes hydroxylation via an alkyl hydrotrioxide intermediate that decomposes releasing much less energy compared to ozonides formed during alkene ozonolysis.
The O3 utilization to form TBA is maximized by (a) an optimum concentration of protic compounds, such as water and alcohols in the liquid phase; and (b) an optimum O3 partial pressure in the gas phase. Beyond the optimum concentration of TBA and water, undesired free-radical pathways adversely affect product selectivity. Similarly, beyond the optimum gas-phase O3 partial pressure, the increased liquid phase O3 concentrations promote O3 decomposition as well as over-oxidation of TBA. Even though TBA selectivity is 85-90% during the first 2-3 h, TBA is susceptible to further oxidation upon prolonged contact with ozone. Hence, to enhance overall atom economy, continuous reactor operation is useful to operate the gas phase at the optimum O3 partial pressure and to optimize the O3 contact time with the liquid phase in order to simultaneously maximize O3 utilization and TBA (desired product) yield, avoiding over-oxidation involving C—C bond cleavage. The proposed method may be extended to activating higher alkanes with tertiary carbons where the C—H bonds are more susceptible to ozonation than those on the primary carbons.
The commercially available tert-butyl alcohol (Sigma-Aldrich, 99.7%, b.p. 23.2° C.) is solid at a typical reaction temperature of 17° C. However, as suggested by the reaction products, a mixture of tert-butyl alcohol/acetone 17 is liquid around 17° C.
In order to evaluate the over-oxidation of tert-butyl alcohol/acetone mixture by ozone, semi-batch runs with such initial mixtures were carried out with the same average residence time of 5.2 min as the isobutane ozonation semi-batch runs. Likely due to lower ozone solubility in these liquid mixtures than in those containing isobutane, the ozone consumption was significantly less during these runs. Therefore, a catalyst bed of Pd/KIT-6 was added at the reactor outlet to decompose the unreacted ozone and to avoid its reaction with the gas bag material.
Listed in Table 3 are results from the ozonation of tert-butyl alcohol/acetone mixture analyzed by NMR experiments. In particular, 1H NMR results confirmed no detectable oxidation of tert-butyl alcohol to tert-butyl hydroperoxide. In contrast to all other results, with less ozone exposure, the nearly zero net-formation of acetone suggests that while acetone may form from ozonation reactions, it may also be ozonized. Also, the detection of methanol suggests that it could be an intermediate that was fast oxidized to formic acid and CO2 over longer exposure to ozone. While tert-butyl alcohol and acetone should be more prone to oxidation than isobutane, the carbon increments in the last column (estimated to be the carbons from tent-butyl alcohol) suggest ozone utilization is less than that during isobutane ozonation. It could be inferred that the reaction rate in such isobutane-absent mixtures was likely around 10−1 of that in isobutane-rich mixtures. Due to the low vapor pressure of the mixture (˜3.4 kPa), over-oxidation in gas-phase contributed dramatically less than isobutane (P=0.25-0.27 MPa) ozonation. Therefore, the results should represent the oxygenate over-oxidation in the liquid phase.
Investigation into Tert-Butyl Hydroperoxide Decomposition
For the majority of the semi-batch runs, the final selectivities towards tert-butyl hydroperoxide were less than 0.02%. However, those runs could not provide evidence whether (a) the OH-containing compounds (water and tert-butyl alcohol) promoted the peroxide decomposition, or (b) the formation of peroxide was inhibited. Therefore, tert-butyl hydroperoxide decomposition was evaluated in various environments to elucidate the mechanisms for the observed low peroxide production.
While tert-butyl hydroperoxide decomposition occurs in product mixtures, exactly overlapping spectra were collected from a sample of tert-butyl hydroperoxide/nonane within 12 h after adding a drop of acetic acid (glacial, Fisher). By 44 h after adding acetic acid of 1.5 orders of magnitude higher concentration than that detected in the product mixture, only 4.4% tent-butyl hydroperoxide was converted. Such slow process indicates that acetic acid cannot contribute to either the low peroxide production or the post-reaction hydroperoxide decomposition during the 2-3 h runs. Also, as confirmed experimentally, acetic acid alone does not significantly affect tent-butyl hydroperoxide reactions in the GC instrument.
The co-existence of water and acetic acid may potentially promote tent-butyl hydroperoxide conversion with water, aiding the dissociation of acetic acid. However, mixtures of tent-butyl hydroperoxide/nonane+acetic acid+water in CD3CN likely contained protons that were excessively active for observation with proton NMR.
The proton NMR spectra were compared from samples before and after treating a mixture of tent-butyl alcohol+acetone+tert-butyl hydroperoxide+water with ozone for 66 min. While the —OOH signal decreased by 13.7% and broadened after ozonation, no tent-butyl hydroperoxide decomposition in the no ozonation sample could be detected 24 h after spectrum collection. The contrast confirmed that tent-butyl hydroperoxide was decomposed by ozonation, but the process was rather slow in the polar solution, possibly due to the lower ozone solubility in absence of isobutane.
The ozonation of tent-butyl hydroperoxide/water/isobutane mixtures with various initial peroxide content (TBHPin column) is shown in Table 4. The last column (TBAoz) estimates the tent-butyl alcohol formed from isobutane ozonation, by subtracting the quantity of converted hydroperoxide from the total quantity of alcohol. Even though the product selectivity from hydroperoxide decomposition is unknown, the low acetone formation also confirms the detrimental effect of hydroperoxide addition to ozone utilization, even at a hydroperoxide/isobutane ratio as low as 0.04%. The results suggest that the hydroperoxide and ozone promote the decomposition of each other. Therefore, the hydroperoxide formation at the initial stages of isobutane ozonation should be low to achieve the high ozone utilization observed in
In summary, two mechanisms together appear to contribute to the observed low tent-butyl hydroperoxide during isobutane ozonation: (a) in the presence of OH-containing compounds (such as water, acetic acid, and tent-butyl alcohol), tent-butyl hydroperoxide formation is less competitive than alternative pathways, which consumed the hydrotrioxide intermediate faster, and (b) the decomposition of the drastically less peroxide is also promoted by ozonation. As the product accumulates, the solvent properties (including polarity, coordination, and ozone solubility) change with the solution composition, and alter the selectivities among the decomposition pathways of the hydrotrioxide intermediates.
At low ozone exposure (short contact time and low O3 partial pressures), β-scission (
Prolonged contact of oxygenates and O3, as well as higher O3 partial pressures, leads to the over-oxidation in the liquid-phase (suggested by
The detection of hydroxyacetone indicates oxidation of terminal methyl groups, which was likely responsible for the over-oxidation forming C1-C3 products (for example, via the further hydroxyacetone oxidation in
Methanol could possibly form (see
In this Example, experiments similar to those in Example 1 were carried out but using a higher temperature (50° C.).
The dioxygen (UHP Plus), argon (HP), and isobutane (UHP) were all purchased from Matheson.
The procedure and experimental apparatus of these experiments have been described in detail in Example 1. The dioxygen stream was used to generate ozone by an Atlas Ozone Generator, and then mixed with argon in a reservoir. Briefly, a Teflon-lined Parr vessel was vacuumed prior to charging the desired amounts of the liquid alkane from an ISCO syringe pump cooled to 10° C. An option to direct the liquid alkane stream through a sample loop containing distilled water was provided to meter in controlled amounts of water. The reactor stirrer was set at 1000 rpm. For each of the sequential batch runs, the reactor was pressurized with an O2/O3/Ar gas mixture of known composition. After reaching a predetermined pressure, the gas inlet valve was shut off, and the reaction was allowed to proceed for 15 min in the closed batch reactor. Then the gas outlet was opened. The alkane that escapes with the gas phase was partially condensed in a cold trap held around −60 to −50° C. and ambient pressure to concentrate the CO2. The gas from the condenser was collected in Tedlar sample bags. After the reactor pressure reaches the vapor pressure of the alkane mixture, and equilibrated to the room temperature, the procedure was repeated. At the end of all the sequential batch runs, the reactor was placed in an ice bath kept in a walk-in freezer at −18° C. At this temperature, the vapor pressure of all compounds remaining in the reactor is much reduced. Then a weighed amount of cold methanol was added into the reactor, and the volatile alkane was allowed to vaporize at 0-4° C. and condensed in the cold trap around −60 to −50° C. After adding 2-pentanone as an internal standard, the methanolic liquid sample was injected into a GC equipped with a HP-PLOT/Q column to resolve ≥C2 products and a flame ionization detector (FID). The methanolic liquid sample was also added to D2O with maleic acid as internal standard to quantify formic acid by 1H NMR. The gas samples collected in tedlar bags were injected into another GC equipped with a thermal conductivity detector to analyze the CO2, and a FID to analyze the hydrocarbons.
As isobutane oxidation is limited by the quantity of ozone, conversions are estimated with detected products from isobutane as follows, treating the CO2 as a byproduct of acetone formation via beta-scission.
wherein niC
The selectivity for product i is defined as follows:
wherein the denominator represents the total moles of all products. Two measures were adopted for ozone utilization. One is ratio of (moles of TBA formed)/(initial moles of O3). Since dioxygen alone cannot oxidize isobutane to detected products at 50° C., the following definition was used to characterize the average quantity of electron transfer induced by each mole of ozone.
During the experiment, the cold trap and its contents were weighed to estimate the amount of condensed isobutane. The carbon balance is defined as follows.
The formation of C—C cleavage products suggests that alkoxyl free radicals may be involved in the reactions. Some secondary radicals may react with dioxygen to yield peroxyl radicals. This process could potentially be promoted at elevated temperature. However, the ozone decomposition rate will also be enhanced at higher temperatures and may pose a limitation. Interestingly, as shown in 5 (Row #1), significant increases in ozone utilization are observed for isobutane hydroxylation at 50° C. compared to the highest value of 80% (TBA/O3=151%) around 17° C. (see Example 1) while maintaining a high TBA selectivity >87%. On average, the 5.9-11 mol of oxidizing equivalents induced by each mol of O3 indicates that O2 is involved in the reaction. On Row #2, the cumulative product formation of 7 sequential batch runs without removing the products from the reactor does not scale compared to the result of a single batch run on Row #1. The result suggests the ozone utilization may be inhibited by progressive alcohol formation, which is consistent with the conclusion from the empirical model.
a Definition as in Eq. (1);
b TBA/O3 molar ratio as used in Example 1.
Other information and data may be found in United States Patent Application 63/158,522, which is hereby incorporated by reference in its entirety.
The word “illustrative” is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as “illustrative” is not necessarily to be construed as preferred or advantageous over other aspects or designs. Further, for the purposes of this disclosure and unless otherwise specified, “a” or “an” means “one or more”.
If not already included, all numeric values of parameters in the present disclosure are proceeded by the term “about” which means approximately. This encompasses those variations inherent to the measurement of the relevant parameter as understood by those of ordinary skill in the art. This also encompasses the exact value of the disclosed numeric value and values that round to the disclosed numeric value.
The foregoing description of illustrative embodiments of the disclosure has been presented for purposes of illustration and of description. It is not intended to be exhaustive or to limit the disclosure to the precise form disclosed, and modifications and variations are possible in light of the above teachings or may be acquired from practice of the disclosure. The embodiments were chosen and described in order to explain the principles of the disclosure and as practical applications of the disclosure to enable one skilled in the art to utilize the disclosure in various embodiments and with various modifications as suited to the particular use contemplated. It is intended that the scope of the disclosure be defined by the claims appended hereto and their equivalents.
The present application claims priority to U.S. provisional patent application number 63/158,522 that was filed Mar. 9, 2021, the entire contents of which are incorporated herein by reference.
This invention was made with government support under 1922649 awarded by the National Science Foundation. The government has certain rights in the invention.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US22/71022 | 3/8/2022 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 63158522 | Mar 2021 | US |
