Method for evaluating performance of petroleum naphtha reforming catalysts

Abstract

A method of evaluating catalyst performance of a catalyst used in a refinery reforming process for converting naphtha to gasoline. More specifically, the catalyst under study is analyzed in parallel at a temperature for a time period in at least two different reaction times to obtain a desired RON in yield, for example, a desired RON in C5+ yield, as a function of time. The reaction is terminated and subsequently the coke is determined and a relationship between the coke and the yield is established, both as a function of the reaction temperature of the catalyst used, thereby enabling a comparison of the catalyst performance determined from the yield versus temperature relationship at a given coke level.

Description

[0001] introduction

[0002] This invention relates to a method for evaluating the catalytic performance of catalysts used in refinery reforming processes that convert naphtha into gasoline. More specifically, the catalyst under study is analyzed in parallel as a fixed bed at a given temperature over time to achieve the desired RON at C5+ yield, taking into account coking and its effects. Background of the Invention

[0003] A significant part of petrochemical refining involves refining naphtha into gasoline. This refining process employs multiphase catalysts. A typical characteristic of this refining process is the formation of coke (or carbon) on the catalyst particles, resulting in partial deactivation of the catalyst. Therefore, this refining process is carried out on an industrial scale using a moving bed reactor, in which particles move through the reactor over time, then exit the reactor and migrate to a regeneration unit, where the coke is removed and the particles are returned to the reactor. This is also known as continuous catalytic regeneration (CCR) reforming. A CCR reforming unit typically has four adiabatic reactors of different volumes and a dedicated regeneration tower, which allows for efficient carbon combustion of the catalyst, redispersibility of active metals, chloride addition, and catalyst drying before reintroducing it into the lead-reactor. As described in George J. Antos and Abdullah M. Aitani's 2nd edition of Catalytic Naphtha Reforming (2004), to ensure that the catalyst activity in the reactor is close to that of fresh catalyst, the catalyst circulation rate is kept high, and the degraded catalyst coking is typically kept below 6% by weight. The catalyst (most commonly a Pt-Sn / Al2O3 variant) is continuously circulated from the inlet of the pre-reactor to the outlet of the lag-reactor. At the outlet, the catalyst is severely deactivated due to coking. High catalyst selectivity (C5+ yield) is always required to maintain product severity.

[0004] This allows for the investigation and comparison of the performance of various catalysts. Catalyst studies are typically conducted in small reactors with volumes of only a few cubic centimeters. These small reactors cannot practically operate like moving bed reactors, so these reactions are carried out in fixed bed reactors. However, the formation of coke is not as effectively removed as in moving bed reactors, making it difficult to translate results from small-scale laboratory trials using fixed bed reactors to plant-scale moving bed reactors. More importantly, over time, partial deactivation not only leads to lower yields (as indicated by C5+) but also a decrease in the RON value of the product stream, which is typically offset by increasing the reactor temperature, as the goal is to achieve a high and stable RON value in the product stream over time. So-called hetero-RON allows for this through an automatic feedback loop between GC analysis of the effluent and the reactor temperature, thus enabling continuous adjustment of the reactor temperature.

[0005] Therefore, there is a need for a method to evaluate the performance of small-scale (e.g., reactors with a volume of a few cubic centimeters) catalysts used in refinery reforming processes for converting naphtha to gasoline, taking into account the coke or carbon deposits formed on the catalyst particles and their impact on the yields of RON and target products (e.g., C5+ or aromatics). Preferably, this process should allow for parallel joint evaluations of individual catalysts to compare their performance. Preferably, the performance of multiple catalysts with the same amount of coke can be compared. Summary of the Invention

[0006] It has now been found that the above-mentioned objective can be achieved, or at least partially achieved, by a method for evaluating the performance of catalysts used in refinery reforming processes that convert naphtha to gasoline, hereinafter referred to as the evaluation method, and the refinery reforming process for converting naphtha to gasoline hereinafter referred to as the naphtha reforming process, wherein the naphtha reforming process forms coke deposits on catalyst particles, and the evaluation method includes:

[0007] a. At least two catalytic reactions are carried out in parallel in at least two parallel reactors using the same amount of the same catalyst, said catalyst being present as particulate matter in a fixed bed within the at least two reactors, said at least two reactors operating under identical LHSV, pressure, and feed composition conditions, the feed consisting of naphtha and hydrogen, said reactions being carried out in the at least two reactors for different reaction times.

[0008] b. Conduct the reaction under condition a at the temperature to achieve the target RON, and adjust the temperature during the reaction time to maintain the target RON during the reaction time under condition a. Adjust the reaction temperature to the same extent using the same catalyst, and record the required reaction temperature at at least 10 measurement points during the longest reaction time to provide a temperature-time relationship.

[0009] c. Measure catalyst performance to provide a yield-time relationship, wherein the catalyst performance is expressed as the target product yield of each catalytic reaction at at least 10 measurement points under condition a within the reaction time of each reaction.

[0010] d. Terminate the catalytic reaction under condition a, wherein at least two reactions terminate after different reaction times, and measure the coke deposits on the catalyst particles to establish a functional relationship between coke deposits and the reaction time of the catalyst used, in order to provide a coke deposit-time relationship.

[0011] e. Use the temperature-time relationship in b, the yield-time relationship in c, and the coke-time relationship in d to calculate the yield-temperature relationship for a given percentage of coke.

[0012] Studies have found that by using the evaluation method described above, various naphtha reforming catalysts can be compared in fixed-bed reactors, despite the formation of coke over time (which varies for each catalyst). The evaluation method of this invention provides an assessment of catalyst performance in moving-bed CCR reactors, even when the reactor is operating at a steady-state coke percentage. Furthermore, this method allows for small-scale testing. Attached Figure Description

[0013] Figure 1 This displays key catalyst performance indicators and operating time.

[0014] Figure 2 This shows the cumulative coke content in each catalyst.

[0015] Figure 3 The important performance metrics, including experimental and fitted data, are displayed.

[0016] Figure 4 A graph showing selectivity versus activity. Detailed Implementation

[0017] In the method according to the invention, in step d, it should be understood that, for a given catalyst, the reaction should proceed for different durations to obtain multiple data points over a period of time. Since coking can only be determined after the reaction has terminated, the number of data points used for the coking-time relationship is largely limited by the number of reactions performed. Therefore, in the method according to the invention, case a specifies that at least two reactions of different durations should exist. To collect more coking data points over time, "at least two" in case a is preferably "at least three," more preferably "at least four."

[0018] In the evaluation method according to the invention, stage b of the method is described as follows: "The temperature is adjusted during the reaction time to maintain the target RON for the reaction time under case a of each reaction, adjusting the temperature of the reaction with the same catalyst to the same extent." The background is that coking or carbon deposition on the catalyst affects activity or selectivity, and this generally means that coking leads to a decrease in the RON of the produced product over time. This RON reduction is undesirable for simulation because large-scale production has regeneration towers, and this situation does not occur on a production scale. To counteract this RON reduction, the temperature can be increased. Since the rate of coking or deposition in the reaction (and therefore the decrease in RON over time) is the same when the catalyst is the same (parallel reactions differing only in duration), the need to increase the temperature to counteract the decrease in RON over time is the same for the same catalyst in parallel tests.

[0019] For the temperature-time relationship in case b of the method of the present invention and the yield-time relationship in case c of the method of the present invention, data points can be collected online during the reaction. Therefore, for both, at least 5, preferably at least 10, and more preferably at least 20 data points should be collected. It is assumed that during the longest reaction time, the reaction temperature at at least 10 measurement points should be recorded (to establish the temperature-time relationship). For a given catalyst used for at least two parallel reactions carried out for different durations, but under the same conditions of the catalyst (e.g., reactants, pressure, temperature), there exists a reaction lasting t1 minutes and at least one reaction lasting longer than t1 minutes. Then, for the reaction with the longest duration, at least 5, preferably at least 10, and more preferably at least 20 data points should be collected to show the reaction time-temperature relationship. When reactions are carried out in parallel with the same catalyst, parallel reactions are carried out at the same temperature with the same catalyst, even if the reaction lasts for a shorter duration (e.g., t1 minutes), although the same time-temperature relationship exists for the shorter duration. In a more specific example: For a catalyst X, the reaction is carried out in three parallel reactors for 50 minutes (reactor 1), 100 minutes (reactor 2), and 150 minutes (reactor 3). The temperature in all three reactors is increased in the same manner to maintain the target RON (for each reactor, the catalyst activity will simultaneously decrease to the same extent due to coking, because under the same conditions, if the temperature were not increased, the RON would decrease over time). In other words, the three reactions in this example proceed in the same manner, but the reactions in reactors 1 and 2 terminate earlier. For reactor 3, the temperature is recorded, for example, every 10 minutes, providing 15 tT data points for reactor 3. For reactors 2 and 1, there are fewer data points (10 and 5 respectively), but for this more finite time, they follow the same tT relationship.

[0020] In the small-scale evaluation method according to the invention, coke formation and deposition occur on the catalyst particles, similar to the situation in large-scale naphtha reforming plants. However, by conducting at least two catalytic naphtha reforming reactions in parallel under the same conditions (aiming for a constant RON, so the temperature typically increases over time), but with different reaction durations, catalyst particles and varying amounts of coke are obtained after the reactions cease. Catalyst particles from the reactor with the longest operating time generally contain more coke than those used in reactions that ceased earlier.

[0021] For example, during the reaction, this method directly measures the temperature required to maintain the target RON yield of the target product (e.g., C5+ hydrocarbons, or certain aromatics, or other products) (the temperature needs to be increased due to coking). The yield can be measured online, for example, by GC. Changes in temperature and yield over time can be recorded. Therefore, temperature-time data points and yield-time data points can be recorded.

[0022] Coking is not measured online (which is impossible while the reaction is in progress), but rather after the final reaction (i.e., the one with the longest duration) has ended. Coking can be measured in various ways, such as by thermogravimetric analysis (TGA), after the particles have been removed from the reactor or while they are still in the reactor. For each sample analyzed, a percentage of coking is obtained, which is related to the time the reactor has been running and active. This provides multiple data points for each catalyst type regarding coking (e.g., expressed as a percentage) and the time the catalyst remains active in the reaction that leads to coking. Thus, this gives a coking (quantity or percentage) - reaction time relationship.

[0023] By comparing various effects or properties over time (target yield, temperature required to maintain constant RON, coking), other relationships can be derived (e.g., using interpolation), such as yield-coking, temperature-coking, or performance under constant coking conditions. This can be ideal because large refineries typically operate their CCR refineries under set coking conditions, for example, because the plant or process is suited to this. Therefore, according to the evaluation method of the invention, this provides a good understanding of catalyst performance under the desired conditions for a particular plant.

[0024] As previously stated, at least two reactions should be carried out under the same catalyst and conditions, differing only in duration. Clearly, the more data points available, the more reliable the evaluation will be. Therefore, the number of reactions carried out in parallel under the conditions in step a should be at least three, preferably at least four.

[0025] This small-scale (small reactor volume) parallel reaction can be readily performed using an apparatus sold by Avantium under the trade name Flowerence, as described in EP 1256377, EP 1888224, and EP 3074121. This apparatus may have four, eight, or even sixteen reactors operating in parallel. Therefore, when using an apparatus with four parallel reactors, a catalyst can typically be tested in a single run with four different reaction times. If the apparatus allows temperature control of one set of four reactors to be different from that of another set of four reactors, an apparatus with, for example, eight reactors operating in parallel can allow eight different reaction times, or can be used to test two different catalysts side-by-side. With this setup, a variety of different CCR catalysts can be tested rapidly and efficiently over a short timeframe. Therefore, in this invention, steps a, b, and c are preferably performed in parallel or sequentially (preferably in parallel) for at least two, preferably at least three, and more preferably at least four different catalysts. Furthermore, in the method according to the invention, steps a, b, and c can preferably be carried out in parallel under different operating conditions of a catalyst, for example, when different target products are of interest or the performance of different RON targets is compared. For example, the evaluation method is carried out in parallel in four reactors, wherein two reactors use catalyst A at a target RON of 102, and two other reactors use catalyst A at a target RON of 100.

[0026] As previously mentioned, reactors using the same catalyst are run for different reaction times (to obtain multiple data points, unless the reaction is intentionally doubled). Different reaction durations can be achieved through various means, but a suitable approach is to have parallel reactions start simultaneously (e.g., by simultaneously feeding liquid feed into all reactors) and terminate them sequentially. Several methods exist for terminating a reactor while allowing other reactors to continue operating, such as by cooling the reactor, stopping the inflow of liquid reactants, or a combination of both. Therefore, in this invention, preferably, by terminating the liquid reactant feed to the reactors at different times, the reaction in step a in at least two reactors proceeds for different durations.

[0027] Similar to commercial CCR reforming units, the evaluation methods of this invention are run to achieve the desired target of producing a component (or group of components) in a preferred high yield. Typically, CCR reforming processes aim to produce relatively high amounts of hydrocarbons having five or more carbon atoms (referred to herein as C5+), or relatively high amounts of aromatics or other target products. Therefore, in this invention, preferably, the target product is C5+ (the sum of hydrocarbon molecules having at least five carbon atoms) or the sum of aromatic components, or a combination thereof. This yield can be the yield over a given time period or it can be the cumulative yield. For some CCR runs, the latter may be preferred.

[0028] In the evaluation method according to the invention, the yield is typically stable for a period after initiation, then decreases over time due to the formation of coke. In commercial CCR operation, the catalyst is continuously circulated, and therefore, the yield of the target component and the average coke content can generally be maintained quite well. This is the difference between the currently protected evaluation method and commercial CCR operation. By obtaining various relationships that change over time (e.g., yield, temperature for maintaining a set RON, amount of coke), the performance or behavior of the catalyst can be evaluated by the method of the invention, particularly regarding the yield at a specified amount of coke. Information processing can be performed by computer and / or graphically, and typically involves interpolation between measurement points. Therefore, it is preferable to obtain relationships based on one or more of b, c, d, and e through interpolation.

[0029] The evaluation method according to the invention is preferably run for a considerable period of time to provide reliable results. Since the reaction is used for CCR operation and the evaluation method is designed to provide a predetermined RON value, it is preferred that the reaction be carried out for a duration of at least 40 hours, preferably at least 60 hours, and more preferably at least 80 hours under evaluation methods a, b, and c.

[0030] Clearly, the more data points for RON, yield, and temperature, the higher the reliability of the interpolation and results. Therefore, in this invention, it is preferable that the measurement of RON, yield, and temperature during the longest reaction time includes at least 20, preferably at least 40 data points for each of RON, yield, and temperature over a period of time.

[0031] The reaction in this invention typically begins with 0% coke and can continue to about 10% (coke weight, determined by TGA based on catalyst weight). Commercial CCR runs are typically carried out with, for example, 4-6% coke. Therefore, in the method now protected, it is preferable to select the longest possible reaction time so that coke (starting at 0%) accumulates to an amount of 2% to 10%, more preferably 3% to 7%, at the end of that reaction time. Obviously, for parallel runs under the same catalyst and conditions but for a shorter duration, this coke (when the reaction terminates) will be less.

[0032] In addition to providing yield performance in the CCR unit at a given coking and associated reactor temperature, the results can be further processed to provide temperature-coking relationships and / or target product yield-coking relationships. Therefore, it is preferable to determine the reactor temperature-coking relationship for each catalyst in the method of the present invention. And similarly, it is preferable to determine the target product-related yield-coking relationship for each catalyst.

[0033] Example

[0034] Three commercially available naphtha reforming catalysts, labeled A, B, and C, were evaluated in a high-throughput parallel test setup featuring 16 reactors. For each catalyst, 0.6 ml of material was loaded into three identical reactors. In a single experiment, a total of 12 reactors were operated simultaneously. At any given time point, all 12 reactors had the same pressure and hydrogen flow rate. The operating conditions for this embodiment are listed in Table 1.

[0035] Table 1

[0036]

[0037] The eluent products were comprehensively analyzed by gas chromatography. Product composition (including C5+ yield, total aromatics yield, and hydrogen yield) and properties (including the Research Octane Number) were determined by the combination of individual components quantified by gas chromatography. In particular, the RON value was determined from the product composition based on the empirical relationship between previously obtained GC-determined composition of catalytic reformed naphtha samples and actual engine knock test results.

[0038] Between measurements of the same reactor, each reactor was sequentially analyzed using GC at a frequency of approximately 9 hours. After each measurement, an algorithm based on PID control automatically adjusted the temperature of the corresponding reactor to ensure that the generated RON value remained close to the RON target.

[0039] During the test, each of the three reactors for each catalyst was successively stopped at three specific time points along the test length. The reactors were individually shut down by interrupting the flow of the liquid (naphtha) and reducing the temperature to approximately 360°C. All reactors continued to operate to observe the same reactor pressure and hydrogen flow rate.

[0040] Gas chromatographic analysis of the now-discontinued reactors confirmed that they no longer produce naphtha for catalytic reforming.

[0041] The experiment ended when the last reactor stopped running. At this point, the spent catalyst was removed from the reactor to analyze the coke content.

[0042] Coke analysis was performed by breaking all beads in each reactor into a fine, uniform powder. Two samples of approximately 15 to 20 mg of powder were collected and analyzed by thermogravimetric analysis (TGA) by applying a constant gas flow to the sample while simultaneously raising the crucible temperature at a rate of 10 °C per minute within a range from room temperature to 800 °C.

[0043] The final weight loss recorded during the analysis corresponds to the amount of coke burned off the catalyst surface after correcting for the water loss and annealing that typically occur in these materials. Weight loss was determined to be approximately 260–550 °C, since below 260 °C only water loss occurs. We corrected for annealing losses by subtracting the curves from those of fresh samples of the same catalyst. Figure 1 Results-runtime relationships for all three catalysts are presented (each pipe represents a reactor, corresponding to the coke deposit value).

[0044] As long as catalyst deactivation does not exceed what is generally considered a reasonable range, coking on the catalyst typically follows a power function with operating time. The equations in Table 2 were obtained by fitting the three coking points of each catalyst over operating time with time using post-mortem TGA analysis. The results are as follows: Figure 2 The figure shows the experimental coking points, with minimum and maximum error bars from repeated coking measurements, fitted curves, and a horizontal line representing the interpolated coking target. The run time corresponding to the coking target (3.75 wt%) was calculated based on these equations. None of the catalysts simultaneously reached the target coking content.

[0045] Table 2

[0046] catalyst The equation for carbon deposits is f(t). Interpolation time (hours) for carbon deposit target A <![CDATA[carbon deposit content (wt%) = 0.208*ToS(h) 0.681 > 69.8 B <![CDATA[Coke content (wt%) = 0.208*ToS (h) 0.480 > 72.9 C <![CDATA[Coke content (wt%) = 0.208 * ToS (h) 0.604 > 143.2

[0047] A polynomial function for the run-time is fitted to the catalyst performance of each catalyst (including reactor temperature, C5+ yield, aromatics yield, and produced hydrogen) and the changes at all data points measured simultaneously in all three reactors. For each catalyst, the activity (reactor temperature) and yield are interpolated based on these fits to determine the run-time required to reach the previously determined target coke content. Figure 3 The experimental data is displayed, overlapping with the fitted curve. Interpolation points are shown on the curve, with the 95% prediction interval derived from the fit.

[0048] Finally, the catalyst performance at a constant coke content was compared by expressing the interpolation yield and interpolation activity (reactor temperature), such as... Figure 4 As shown (vertical and horizontal error bars = 95% prediction interval; label = time to reach the carbon deposit target in hours, the same as the values ​​in Table 2).

Claims

1. A method for evaluating the catalyst performance of a catalyst used in a refinery reforming process for converting naphtha to gasoline, the method being referred to below as an evaluation method, the refinery reforming process for converting naphtha to gasoline being referred to below as a naphtha reforming process, wherein, Naphtha reforming processes can lead to the formation of coke deposits on catalyst particles, and the evaluation methods include: a. At least two catalytic reactions are carried out in parallel in at least two parallel reactors using the same catalyst in equal amounts, said catalyst being present as particulate matter in the fixed beds of the at least two reactors, the at least two reactors operating under identical LHSV, pressure, and feed composition conditions, the feed consisting of naphtha and hydrogen, with the reactions in the at least two reactors proceeding for different reaction times. b. Conduct the reaction under condition a at the temperature to achieve the target RON, and adjust the temperature during the reaction time to maintain the target RON during the reaction time under condition a. Adjust the temperature of reactions using the same catalyst to the same extent, and record the required reaction temperature at at least 10 measurement points during the longest reaction time to provide a temperature versus time relationship. c. Measure catalyst performance to provide a yield versus time relationship, said catalyst performance being the target product yield of each catalytic reaction at at least 10 measurement points under condition a within the reaction time of each reaction. d. Terminate the catalytic reaction under condition a, wherein at least two reactions terminate after different reaction times, and measure the coke deposits on the catalyst particles to establish a functional relationship between coke deposits and catalyst reaction time, which will be used to provide the relationship between coke deposits and time. e. Use the temperature-time relationship in b, the yield-time relationship in c, and the coking-time relationship in d to calculate the yield-temperature relationship for a given percentage of coking.

2. The method as described in claim 1, wherein, The target product is C5+, or the sum of aromatic components, or a combination thereof, where C5+ is the sum of hydrocarbon molecules having at least 5 carbon atoms.

3. The method as described in claim 1 or 2, wherein, The number of reactions carried out in parallel under the conditions in step a is at least three.

4. The method of claim 3, wherein, The number of reactions carried out in parallel under the conditions in step a is at least four.

5. The method of claim 1, wherein, By terminating the feeding of liquid reactants into the reactors at different times, the reaction in step a in at least two reactors proceeds for different durations.

6. The method of claim 1, wherein, For at least two different catalysts, steps a, b, and c are performed in parallel or sequentially.

7. The method of claim 6, wherein, For at least three different catalysts, steps a, b, and c are performed in parallel or sequentially.

8. The method of claim 6, wherein, For at least four different catalysts, steps a, b, and c are performed in parallel or sequentially.

9. The method of claim 1, wherein, The response under assessment methods a, b, and c shall be evaluated for a duration of at least 40 hours.

10. The method of claim 1, wherein, The response under assessment methods a, b, and c shall be evaluated for a duration of at least 60 hours.

11. The method of claim 1, wherein, The response under assessment methods a, b, and c shall be evaluated for a duration of at least 80 hours.

12. The method of claim 1, wherein, Productivity is cumulative productivity.

13. The method of claim 1, wherein, Determine the relationship between reactor temperature and coke deposition for each catalyst.

14. The method of claim 1, wherein, Determine the relationship between the target product yield and coke deposition for each catalyst.

15. The method of claim 1, wherein, Interpolation is used to obtain relationships based on one or more of b, c, d, and e.

Images