Evaluation of H-BEA Zeolite Dealumination and Adsorption Study of Alkylaromatics in Commercial Y-Type Zeolites

Evaluation of H-BEA Zeolite Dealumination and Adsorption Study of Alkylaromatics in Commercial Y-Type Zeolites Felipe Mart´ınez Barreto Undergraduate Thesis in Partial Fulfillment of the Requirements for the Degree of Bachelor in: Chemical Engineering Thesis Advisor: Julio C´esar Vargas Saenz UNIVERSIDAD NACIONAL DE COLOMBIA FACULTY OF ENGINEERING DEPARTMENT OF CHEMICAL ENGINEERING BOGOT´ A, D.C. July 25th, 2025 Solo ustedes, mis pilares, mis hero´ınas, saben c´omo ha sido este camino. Los tres amores de mi vida, sin ustedes simplemente no estar´ıa d´onde estoy. Gracias por impulsarme a llegar hasta aqu´ı, por levantarme en los momentos dif´ıciles, por regalarme sonrisas en los m´as bellos, y por motivarme siempre a seguir so˜nando con metas a´un m´as grandes, no solo en lo acad´emico, sino en lo emocional especialmente. Para mis estrellitas en el firmamento que me cuidan siempre, est´a Tesis va dedicada a ustedes. Las amo... Mam´a, Dani y Sarita. ACKNOWLEDGMENTS I would like to thank my thesis advisor, Professor Julio C´esar Vargas Saenz. This semester, under his mentorship, I learned so much—especially the deep love for knowledge and the commitment to give back to our country through teaching and academia. I hope to one day follow the same path. Thank you for your patience during the reconstruction of the reactor, and for the countless hours you dedicated to helping me understand and carry out each of the experiments. I would also like to especially thank Professor Raj Gounder, who gave me the opportunity to begin this journey at Purdue University in West Lafayette. Thank you for trusting me, for allowing me to work in your laboratories, and for your always timely advice, which constantly motivated me to keep pursuing research. I hope to continue down that path and that our paths will cross again soon. To my family: Rolis, Johana, and the girls. To Mam´a Lucy, thank you for every time you sent me a plate of ”Arrocito con Pollo”. Throughout the thesis, and during my entire degree, I felt your unconditional love and support. Words are not enough to thank you. To my university friends: Sofi, Carlos, Nakay, Brayan, and Cami. Thank you for the good times, for being there when I needed you the most, and for making “la Nacho” that magical place I always wanted to return to. Thanks to you, I’ll carry with me the memory of a university filled with laughter, affection, and unforgettable moments. I wish you all the best in the new journeys ahead. And finally, to the National University of Colombia, my alma mater, a treasure I will carry with me for the rest of my life. You taught me to love what I do, to grow as a person, and to become the human being I am today. Thank you, deeply. Contents ACKNOWLEDGMENTS ....................... 2 1. CHAPTER 1: Evaluation of H-BEA Zeolite Dealumination 10 1.1 Abstract.............................. 10 1.2 Introduction............................ 10 1.3 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . 13 1.3.1 H-BEA Zeolite Physicochemical Properties . . . . . . . . 13 1.3.2 H-BEA Dealumination Experimental Design . . . . . . . 14 1.3.3 Characterization of H-BEA Dealuminated Samples . . . 16 1.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . 20 1.5 Conclusions ............................ 34 2. CHAPTER 2: Adsorption Study of Alkylaromatics in Commercial Y-Type Zeolites ................... 36 2.1 Abstract.............................. 36 2.2 Introduction............................ 37 2.3 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . 38 2.3.1 Physicochemical Properties of Zeolite HY Powder (ZeolystCBV100) ....................... 38 2.3.2 Equipment Description . . . . . . . . . . . . . . . . . . . 39 2.3.3 Calibration of Mass Flow Controllers . . . . . . . . . . . 44 2.3.4 TPD Experimentation Description . . . . . . . . . . . . 51 2.3.5 Mass Spectrometer Calibration . . . . . . . . . . . . . . 53 2.4 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . 54 2.4.1 Mass Spectrometer Calibration Results . . . . . . . . . . 54 2.4.2 Adsorption/Desorption Results . . . . . . . . . . . . . . 57 2.5 Conclusions ............................ 67 5. REFERENCES ............................ 69 6. APPENDICES ............................ 72 A. Standard Operating Procedure’s (SOP’s) ........... 72 A.1 Standard Operating Procedure (SOP) - Differential Reactor Operation ............................. 72 A.2 Standard Operating Procedure (SOP) - Adsorption/Desorption 74 3 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia B. Mass Spectrometer Calibration Results ............ 75 B.1 Smoothed Results for Mass Spectrometer Calibration . . . . . 75 B.2 Area under smoothed peaks for Mass Spectrometer Calibration 82 B.3 Results of Probe Injection Tests in the Differential Reactor: Empty and with a Bed of Ground Quartz and SiC . . . . . . . 89 C. Saturation peaks and TPD Results ............... 93 4 List of Figures 1 Dealumination Experimental Design for H-BEA (15) using EDTAandHCl.......................... 15 2 Rigaku SmartLab diffractometer (XRD) at Chemical Engineering Purdue Laboratories . . . . . . . . . . . . . . . . . . . . . 16 3 Micromeritics 3-Flex instrument (N2adsorption) at Chemical Engineering Purdue (Gounder Group Laboratories) . . . . . . 17 4 Inductively Coupled Plasma Optical Emission Spectroscopy (ICP-OES) at Chemical Engineering Purdue (Gounder Group Laboratories) ........................... 18 5 Temperature Programmed Desorption (TPD) at Purdue Chemical Engineering (Gounder Group Laboratories) . . . . . 19 6a) Triangles - HCl, Crosses - Na2H2EDTA 0.13M, Diamonds -Na2H2EDTA 0.24M, Squares - Na2H2EDTA pH = 8. b) H4EDTA ............................. 20 7 H-BEA 15 XRD Pattern (Top/Down)-([1] H-BEA 15 the lowest one/Experimental Obtained) . . . . . . . . . . . . . . . . . . . 21 8 HClXRDPattern......................... 22 9Na2H2EDTA 0.13 and 0.24 M XRD Pattern . . . . . . . . . . 22 10 Crystallinity Results a) Na2H2EDTA 0.13 and 0.24 M b) HCl 23 11 H4EDTA XRD Pattern and Relative Crystallinity . . . . . . 24 12 [H+]/[Al] and [Si]/[Al] for a) HCl, b) Na2H2EDTA at pH = 8, c) Na2H2EDTA 0.13, d) Na2H2EDTA 0.24 M . . . . . . 25 13 HCl, Na2H2EDTA pH=8, Na2H2EDTA 0.13 and 0.24 FALn/FAL0curves ....................... 27 14 Stacked EFAL and FAL dealumination a) HCl, b) Na2H2EDTA 0.13M, c) Na2H2EDTA 0.24M . . . . . . . . . 28 15 N2Adsorption Isotherms a) HCl, b) Na2H2EDTA at 0.13 M c) Na2H2EDTA at0.24M.................... 29 16 SurfaceAreaHCl......................... 30 17 Surface Area Na2H2EDTA a) 0.13 M, b) 0.24M . . . . . . . . 31 18 PoreVolumenHCl ........................ 32 19 Pore Volumen Na2H2EDTA a) 0.13 M, b) 0.24M . . . . . . . 33 20 Gas Phase Differential Reactor Diagram . . . . . . . . . . . . 39 21 Possible arrangement using two four-way valves . . . . . . . . 41 5 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia 22 Differential Reactor and Mass Spectrometer Analysis. (a) Gas Mixer, (b) Furnace and Quartz U-tube Reactor, (c) Gas PreHeating Controllers, (d) Mass Flow Controllers, (e) Temperature Display of the Streams Entering the Valve Arrangement, (f) Furnace Temperature Controller, (g) QGA Mass Spectrometer. (h) Four Ways Valves Arrangement . . . . . . . . . . . . 43 23 a) Furnace-Reactor System and Packed Bed Reactor with Zeolyst CBV 712 Diluted at 4.5% in SiC (see Figure 33) . . . . . 43 24 (a) Gas Mixer, (e) Temperature Display of the Streams Entering the Valve Arrangement, (f) Furnace Temperature Controller, (g) QGA Mass Spectrometer. . . . . . . . . . . . . . . 44 25 (c) Gas Pre-Heating Controllers, (d) Mass Flow Controllers, (h) Four Ways Valves Arrangement . . . . . . . . . . . . . . . 44 26 Pure He Mass Flow Controller Calibration . . . . . . . . . . . 45 27 Pure He from Gas Mixer Mass Flow Controller Calibration . . 46 28 Pure CO2from Gas Mixer Mass Flow Controller Calibration . 46 29 Pure CO2andHeDensity .................... 48 30 CO2He Mixture Density using Peng-Robinson at 18°C . . . 48 31 CO2He Mixture Relative Error for lineal 5th order regression 49 32 CO2He Mixture Valve opening calibration to obtain 50 mL/minat40PSIG ....................... 51 33 U Reactor Arrangement . . . . . . . . . . . . . . . . . . . . . 52 34 Area under peaks (Blank sample) for a) Toluene and b) Benzene 55 35 Area under peaks (SiC sample) for a) Toluene and b) Benzene 55 36 Benzene concentration and Temperature Ramp Results a) 25°C, b) 75°C and c) 150°C ................... 57 37 Benzene TPD for Zeolite H-Y diluted in SiC . . . . . . . . . . 58 38 Toluene concentration and Temperature Ramp Results a) 25°C, b) 75°C and c) 150°C ................... 60 39 Benzene TPD for Zeolite H-Y diluted in SiC . . . . . . . . . . 61 40 Benzene TPD Gaussian Convolution 25°Cn=5......... 63 41 Benzene TPD Gaussian Convolution 75°Cn=4......... 63 42 Benzene TPD Gaussian Convolution 150°C n=3 . . . . . . . . 64 43 Toluene TPD Gaussian Convolution 25°Cn=4......... 64 44 Toluene TPD Gaussian Convolution 75°Cn=5......... 65 45 Toluene TPD Gaussian Convolution 150°C n=3 . . . . . . . . 65 46 Toluene and Benzene TPD reported by [2] at 25°C as AdsorptionTemperature......................... 67 6 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia 47 Smoothed Toluene peaks at 25°C (Blank sample) . . . . . . . 75 48 Smoothed Toluene peaks at 75°C (Blank sample) . . . . . . . 76 49 Smoothed Toluene peaks at 150°C (Blank sample) . . . . . . . 76 50 Smoothed Benzene peaks at 25°C (Blank sample) . . . . . . . 77 51 Smoothed Benzene peaks at 75°C (Blank sample) . . . . . . . 77 52 Smoothed Benzene peaks at 150°C (Blank sample) . . . . . . . 78 53 Smoothed Toluene peaks at 25°C (SiC sample) . . . . . . . . . 78 54 Smoothed Toluene peaks at 75°C (SiC sample) . . . . . . . . . 79 55 Smoothed Toluene peaks at 150°C (SiC sample) . . . . . . . . 79 56 Smoothed Benzene peaks at 25°C (SiC sample) . . . . . . . . 80 57 Smoothed Benzene peaks at 75°C (SiC sample) . . . . . . . . 80 58 Smoothed Benzene peaks at 150°C (SiC sample) . . . . . . . . 81 59 Area under Toluene peaks at 25°C (Blank sample) . . . . . . . 82 60 Area under Toluene peaks at 75°C (Blank sample) . . . . . . . 83 61 Area under Toluene peaks at 150°C (Blank sample) . . . . . . 83 62 Area under Benzene peaks at 25°C (Blank sample) . . . . . . 84 63 Area under Benzene peaks at 75°C (Blank sample) . . . . . . 85 64 Area under Benzene peaks at 150°C (Blank sample) . . . . . . 85 65 Area under Toluene peaks at 25°C (SiC sample) . . . . . . . . 86 66 Area under Toluene peaks at 75°C (SiC sample) . . . . . . . . 86 67 Area under Toluene peaks at 150°C (SiC sample) . . . . . . . 87 68 Area under Benzene peaks at 25°C (SiC sample) . . . . . . . . 87 69 Area under Benzene peaks at 75°C (SiC sample) . . . . . . . . 88 70 Area under Benzene peaks at 150°C (SiC sample) . . . . . . . 88 71 Area under peaks as function of Temperature in (Blank sample) a)Toluene b) Benzene ...................... 89 72 Area under peaks as function of Temperature in (SiC sample) a)Toluene b) Benzene ...................... 90 73 Area under peaks as function of Toluene 25°C pulse volume, comparison between Blanck and SiC . . . . . . . . . . . . . . 90 74 Area under peaks as function of Toluene 75°C pulse volume, comparison between Blanck and SiC . . . . . . . . . . . . . . 91 75 Area under peaks as function of Toluene 150°C pulse volume, comparison between Blanck and SiC . . . . . . . . . . . . . . 91 76 Area under peaks as function of Benzene 25°C pulse volume, comparison between Blanck and SiC . . . . . . . . . . . . . . 92 77 Area under peaks as function of Benzene 75°C pulse volume, comparison between Blanck and SiC . . . . . . . . . . . . . . 92 7 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia 78 Area under peaks as function of Benzene 150°C pulse volume, comparison between Blanck and SiC . . . . . . . . . . . . . . 93 79 Benzene injection in 3µL pulses at 25°C followed by TPD analysis. ................................ 94 80 Benzene injection in 3µL pulses at 75°C followed by TPD analysis. ................................ 94 81 Benzene 25°C Adsorption TPD . . . . . . . . . . . . . . . . . 95 82 Benzene 75°C Adsorption TPD . . . . . . . . . . . . . . . . . 95 83 Benzene 150°C Adsorption TPD . . . . . . . . . . . . . . . . 96 84 Toluene injection in 3µL pulses at 25°C followed by TPD analysis. 96 85 Toluene injection in 3µL pulses at 75°C followed by TPD analysis. 97 86 Toluene injection in 3µL pulses at 150°C followed by TPD analysis. ................................ 97 87 Toluene 25°C Adsorption TPD . . . . . . . . . . . . . . . . . . 98 88 Toluene 75°C Adsorption TPD . . . . . . . . . . . . . . . . . . 98 89 Toluene 150°C Adsorption TPD . . . . . . . . . . . . . . . . . 99 8 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia of 1 and 24 hours at 100°C. Finally, aqueous solutions of disodium EDTA were prepared at concentrations of 0.13 and 0.24 M and applied for 1, 2, 4, and 8 hours at 100°C. In general terms, the experimental design follows the scheme shown in Figure 1, starting with H-BEA (15) zeolite supplied by Clariant (HCZB-30). Approximately 2.0000 ±0.0001 grams of sample were taken for dealumination and 20 ml of the corresponding acid solution was added to a Teflon vessel. The acid treatment was then carried out at 100°C for the durations mentioned previously for each acid. After the dealumination step, the samples were centrifuged at 5000 rpm and washed five times with DI water to remove the supernatant. Finally, the solids were left in a static oven at 373 K overnight. Figure 1: Dealumination Experimental Design for H-BEA (15) using EDTA and HCl With the sample dried and placed in the vials, morphological analyzes of the dealuminated structures were carried out using XRD, N2adsorption, ICP, and TPD techniques to determine surface and volumetric properties, such as specific surface area and total, external, and micropore volume, as well as the BJH pore size distribution. In addition, XRD patterns were 15 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia analyzed to assess whether any destruction of the crystalline structure had occurred after treatment. Furthermore, the amount of acid sites [FAL] and the Si/Al ratio of the structure were measured to enable a deeper analysis of the studied phenomenon. The following sections describe each of the techniques in more detail. 1.3.3 Characterization of H-BEA Dealuminated Samples •XRD The crystal structures were analyzed by obtaining powder X-ray diffraction data using a Rigaku SmartLab diffractometer equipped with a Cu Kαradiation source (λ= 1.54 ˚ A, operating conditions: 40 kV and 44 mA). Diffraction measurements were performed over a 2θrange of 4°to 50°with a step of 0.01°and a scanning rate of 0.0167°per second. The equipment can be seen in Figure 2: Figure 2: Rigaku SmartLab diffractometer (XRD) at Chemical Engineering Purdue Laboratories •N2 Adsorption 16 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Nitrogen adsorption isotherms were acquired at 77 K using a Micromeritics 3-Flex instrument. Approximately 0.05 g of sample was first degassed under a dynamic vacuum (pressure below 0.67 kPa), then heated at 393 K for 2 hours followed by further heating at 623 K for 9 hours with a controlled rate of 0.167 K per second before conducting the adsorption experiments. 3-Flex equipment can be seen in Figure 3: Figure 3: Micromeritics 3-Flex instrument (N2adsorption) at Chemical Engineering Purdue (Gounder Group Laboratories) •ICP-OES For elemental analysis, inductively coupled plasma optical emission spectroscopy (ICP-OES) was used, employing a Thermo Scientific iCAP 7000 Plus Series instrument as it may be seen in Figure 4. For 17 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia this purpose, about 0.02 g of the solid sample was dissolved in 2.5 g of hydrofluoric acid (48 wt %, Alfa Aesar). After allowing the reaction to proceed for over 24 hours, 1 g of nitric acid (70 wt %, Sigma-Aldrich) was added, and the solution was further diluted with 50 g of deionized water. Figure 4: Inductively Coupled Plasma Optical Emission Spectroscopy (ICPOES) at Chemical Engineering Purdue (Gounder Group Laboratories) •TPD Ammonia temperature-programmed desorption (TPD) was carried out to determine the quantity of protonic (H+) sites present on BEA zeolite samples. Prior to the TPD analysis, the samples underwent an ionexchange procedure using ammonium nitrate to replace sodium or other 18 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia exchangeable cations with ammonium ions. For each run, between 0.02 and 0.06 grams of sample were packed into a quartz U-tube reactor and held in place with quartz wool on both ends. The TPD measurements were conducted using a Micromeritics AutoChem II 2920 Chemisorption Analyzer seen in Figure 5. After ion exchange, the samples were subjected to a temperature ramp under helium flow to thermally decompose the ammonium ions, leading to the desorption of ammonia and allowing quantification of the Brønsted acid sites present in the material. Figure 5: Temperature Programmed Desorption (TPD) at Purdue Chemical Engineering (Gounder Group Laboratories) 19 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia 1.4 Results and Discussion The first analysis obtained, which provides the clearest overview of the dealumination process for each of the species, was the elemental analysis by ICP. Based on the methodology described previously, it was possible to calculate the amount of Si and Al in %wt, and from this estimate the [Si/Al] ratio for each of the aforementioned species as a function of time, as shown in the following Figure 6: Figure 6: a) Triangles - HCl, Crosses - Na2H2EDTA 0.13M, Diamonds - Na2H2EDTA 0.24M, Squares - Na2H2EDTA pH = 8. b) H4EDTA Furthermore, it was found that there is no significant change in the Si wt% of the dealuminated samples, as also reported by [3] in their results. This observation will become even more relevant in a later analysis. However, in the case of Al wt%, a noticeable decrease was observed, which is reflected in a progressive increase in the [Si/Al] ratio for each of the species. In the case of HCl, it is evident that the dealumination process is stronger than that observed with the three different EDTA forms evaluated. Notably, the [Si/Al] ratio does not converge to a stable value, which may indicate that this particular treatment removes both (FAL) and (EFAL) throughout the dealumination process. In contrast, for all EDTA-based treatments, the removal of aluminum appears to stabilize, suggesting a preference for extracting the more easily accessible structures, EFAL, until it reaches a point where further removal is no longer possible. This behavior implies 20 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia that, unlike HCl, EDTA is not capable of extracting structural aluminum from the zeolite framework. Later, it will be confirmed or not by analyzing the TPD results. Finally, in terms of the ICP results, it is worth noting that compared to H4EDTA and Na2H2EDTA at pH = 8 treatments, Na2H2EDTA solutions at concentrations of 0.13 and 0.24 M are more effective at removing aluminum, reaching [Si/Al] values of up to 28. This suggests that treatment with H4EDTA and Na2H2EDTA at pH = 8 does not possess enough acidity to remove any form of EFAL. For this reason, they will not be discussed in detail in the future, unlike HCl and Na2H2EDTA at 0.13 and 0.24 M, which will remain the focus. On the basis of the previous analysis, the next technique to further complement the understanding of the dealumination process was the study of crystallinity through XRD. Initially, special attention was given to the patterns of HCl and Na2H2EDTA at 0.13 and 0.24 M, as these are the conditions that, up to this point, appear to have induced significant structural changes. First, the XRD pattern was compared with the literature reported in [1], as it may be seen in Figure 7: Figure 7: H-BEA 15 XRD Pattern (Top/Down)-([1] H-BEA 15 the lowest one/Experimental Obtained) 21 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia As can be observed, the reference zeolite exhibits the same characteristic peaks reported by [1] for H-BEA (15) zeolites, specifically within the ranges of 7-8°and 22-23°. Taking this into account, the following Figure 8 and Figure 9 compare this reference pattern as the baseline in all diagrams with the dealuminated samples, initially for HCl and Na2H2EDTA at 0.13 and 0.24 M for each treatment time: Figure 8: HCl XRD Pattern Figure 9: Na2H2EDTA 0.13 and 0.24 M XRD Pattern At first glance, no significant changes are observed in the patterns, meaning there is no evidence of a total degradation of the structure. For this reason, the calculation of Relative Crystallinity with respect to the 22 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia reference pattern is carried out in order to perform a more detailed analysis of the changes in the crystalline structure as a function of the acid used and the treatment time. The following presents the expression used for the Crystallinity calculation and the corresponding results: CRel =PISample PIP attern (1) The summation was performed for the characteristic peaks located at 7 to 8 °and 22 to 23 °, resulting in the following: Figure 10: Crystallinity Results a) Na2H2EDTA 0.13 and 0.24 M b) HCl In Figure 10, a similar behavior can be observed in all treatments, showing an initial increase in crystallinity followed by a decline. This can be complemented by the observations in [4], where an initial removal of EFAL located in the micropores leads to pore opening. However, as time progresses, the interaction with FAL becomes more frequent. Up to this point, based on the results obtained, this reasoning appears to be consistent with the findings. Nevertheless, data from TPD analysis presented later will demonstrate a novel effect that supports this preferential removal behavior throughout the dealumination process. Finally, XRD and crystallinity results were obtained for H4EDTA, confirming—along with the ICP results in Figure 6—that no significant dealumination occurred. The same conclusion applies to Na2H2EDTA at 23 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia pH = 8. Figure 11: H4EDTA XRD Pattern and Relative Crystallinity Now, the complementary analysis to ICP for this experimentation was TPD, which allows determining the number of Brnsted acid sites associated with an FAL, through ion exchange with NH4NO3followed by ammonia desorption using a programmed temperature ramp and an online chromatograph in AutoChem. This means that, from the total aluminum content estimated by ICP, it is now possible to distinguish between FAL and EFAL, allowing the estimation of the [H+] Al ratio. With this, the following graphs were constructed: 24 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Table 3: Surface Area HCl Sample Time (h) Surface Area (m2/g) Total Micropore External Parent-HBEA (15) HCZ30 0 618 468 151 H-BEA (15) HCl 0.25 643 449 194 0.5 592 417 175 1 549 379 169 2 566 400 156 For its part, the surface treated with Na2H2EDTA does not seem to show a significant change in the surface area, both for micropores and external sites of the crystalline structure, as can be seen in Figure 17: Figure 17: Surface Area Na2H2EDTA a) 0.13 M, b) 0.24M 31 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Table 4: Surface Area for H-BEA (15) under Na2H2EDTA Treatment Sample Time (h) Surface Area (m2/g) Total Micropore External Parent-HBEA (15) HCZ30 0 618 468 151 H-BEA(15) Na2H2EDTA 0.13M 1 614 444 169 2 620 444 175 8 605 435 170 H-BEA(15) Na2H2EDTA 0.13M 1 603 432 172 2 614 438 176 8 591 420 171 Just as with the surface area, an initial increase in pore volume can be observed, associated with the effect of aluminum removal that was blocking the pores. However, as in the previous case, when the structure degrades, the volume of each type of pore also decreases, as shown in Figure 18: Figure 18: Pore Volumen HCl 32 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Table 5: Pore Volume for H-BEA (15) under HCl Treatment Sample Time (h) Pore Volume (m3/g) Total Micropore External Parent-HBEA (15) HCZ30 0 0.369 0.244 0.125 H-BEA (15) HCl 0.25 0.395 0.236 0.159 0.5 0.363 0.231 0.132 1 0.337 0.196 0.141 2 0.335 0.213 0.122 For the pore volume of the Na2H2EDTA-treated sample, there was also no significant change, which is desirable for the dealumination process, as shown in Figure 19: Figure 19: Pore Volumen Na2H2EDTA a) 0.13 M, b) 0.24M 33 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Table 6: Pore Volume for H-BEA (15) under Na2H2EDTA Treatments Sample Time (h) Pore Volume (m3/g) Total Micropore External Parent-HBEA (15) HCZ30 0 0.369 0.244 0.125 H-BEA(15) Na2H2EDTA 0.13M 1 0.378 0.222 0.157 2 0.383 0.219 0.164 8 0.375 0.218 0.157 H-BEA(15) Na2H2EDTA 0.24M 1 0.371 0.225 0.146 2 0.375 0.217 0.158 8 0.367 0.209 0.158 1.5 Conclusions Based on the results obtained using the various previously described techniques, it was found that the dealuminating agents with the most significant effects were HCl and Na2H2EDTA at concentrations of 0.13 and 0.24 M. In contrast, the basic and anhydrous forms of EDTA produced no observable structural or physicochemical changes on the surface. Notably, the most effective dealumination was achieved with Na2H2EDTA 0.24 M after 8 hours, reaching a FALn/FAL0ratio of 0.92, which corresponds to an 8% loss of framework aluminum (FAL) compared to the reference material. Additionally, the [H+/Al] ratio increased from 0.62 to 0.93, indicating a decrease in the extra-framework aluminum (EFAL) population from 38% to 7%, i.e., a 31% reduction relative to the initial value. The second-best performance was observed with Na2H2EDTA 0.13 M, showing a FALn/FAL0ratio of 0.87 after 8 hours, equivalent to a 13% loss of FAL, and an increase in the [H+/Al] ratio from 0.62 to 0.89, corresponding to a 27% reduction in EFAL content. In contrast, the HCl treatment resulted in an undesired effect, with a FALn/FAL0ratio dropping to 0.39 after just 2 hours, indicating excessive removal of FAL. Moreover, the [H+/Al] ratio increased only slightly, from 0.62 to 0.65, suggesting that no clear selectivity was achieved between EFAL 34 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia and FAL removal. Figure 14 clearly illustrates this behavior. Furthermore, combined ICP and TPD analysis revealed partial reincorporation of EFAL into the framework as FAL in the samples treated with Na2H2EDTA 0.13 and 0.24 M. This phenomenon is observed in Figure 14 as an increase in the slope of framework aluminum content and also in Figure 13 through the evolution of the FALn/FAL0ratio. However, based on XRD results, this reincorporation does not appear to preserve the crystallinity of the original structure. A time-dependent trend in the removal rates of EFAL and FAL was also identified, as seen in Figure 12. Initially, HCl treatment removes primarily EFAL, which is reflected in the 3-Flex analysis as an increase in external pore volume (micropore opening). However, after a certain point, the trend shifts toward predominant removal of FAL, resulting in a gradual decrease in both surface area and pore volume. In contrast, Na2H2EDTA treatments at 0.13 and 0.24 M selectively remove EFAL throughout the entire time range studied, with negligible changes in surface and porosity properties. In conclusion, treatment of H-BEA zeolite (Si/Al = 15) with Na2H2EDTA at 0.13 and 0.24 M allows for selective removal of EFAL species, achieving [H+/Al] ratios of 0.89 and 0.93, respectively, with controlled losses of FAL (13% and 8%). These modifications were accomplished without significant changes in key properties such as surface area, pore volume, or relative crystallinity compared to the reference material. For future research, it would be valuable to understand, at a molecular level, the interactions between Na2H2EDTA and the different forms of EFAL and FAL, to investigate how EFAL reincorporation may be influenced by diffusion limitations, and to explore how the reduced acidity of EDTA—due to sodium counterions—affects dealumination efficiency. Further studies on zeolites with narrower pore systems are also encouraged. 35 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia 2 CHAPTER 2: Adsorption Study of Alkylaromatics in Commercial Y-Type Zeolites 2.1 Abstract The adsorption and desorption behavior of alkyl aromatic compounds on H-Y CBV 100 zeolite was investigated using temperature-programmed desorption (TPD) experiments with toluene and benzene as probe molecules. The study aimed to evaluate the influence of adsorption temperature and bed configuration on desorption profiles, focusing on diffusion limitations within the porous zeolite network. A diluted bed using SiC was employed to minimize transport artifacts, allowing a clearer distinction of adsorption site distributions and energetic profiles. Results showed distinct differences between the desorption behaviors of benzene and toluene. Benzene exhibited sharper, well-defined peaks without shoulders, indicating less hindered desorption, while toluene showed broader peaks with pseudo-peaks (shoulders), particularly at lower adsorption temperatures (25°C and 75°C), reflecting stronger diffusional constraints. Increasing the adsorption temperature led to a shift of desorption peaks toward lower temperatures and a notable reduction in peak intensity, attributed to reduced molecular loading and weaker cooperative interactions. In the case of benzene, additional high-temperature peaks were observed at elevated adsorption temperatures, likely due to enhanced access to micropores by more mobile, high-energy molecules. A qualitative analysis of the number and distribution of adsorption sites was performed by fitting the TPD data using a Gaussian convolution approach with the lsqcurvefit optimization function in MATLAB. While this method provided insight into the dominant desorption sites, it remains a phenomenological approximation. Comparative analysis with a previous study using undiluted beds [2] confirmed the positive impact of SiC dilution in minimizing diffusional resistance and enhancing the interpretability of desorption behavior in zeolite systems. 36 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia 2.2 Introduction The study of adsorption and desorption phenomena in porous materials has been widely investigated due to its importance in separation processes, storage systems, and heterogeneous catalysis. In particular, Y-type zeolites, owing to their well-defined crystalline structure and the presence of Brønsted acid sites, have been extensively used in studies focusing on the interaction of organic and inorganic molecules with microporous surfaces. However, due to the complexity of their pore network, mass transport within these structures is often limited by intracrystalline or interparticle diffusion phenomena, which can significantly affect the experimentally observed adsorption and desorption profiles. Several studies have highlighted how diffusive transport can distort the data obtained through techniques such as temperature-programmed desorption (TPD). For instance, the introduction of mesoand macroporosity in hierarchical zeolites has been shown to enhance molecular mobility and reduce the distortion of TPD profiles [12]. Likewise, diffusion inside microporous frameworks can lead to broadened peaks or multiple desorption signals in TPD curves, which are often associated with a distribution of adsorption energies and internal transport barriers [13, 14]. Additional work has emphasized how bed compaction, porosity, and crystal size affect the desorption kinetics of aromatic compounds, introducing thermal hysteresis and diffusional delay effects [15]. In line with this research, a previous study focused on the desorption behavior of toluene and benzene in H-Y CBV 100 zeolite using TPD analysis [2]. The resulting desorption profiles exhibited broad peaks, high-temperature shifts, and the presence of shoulder-like features, all of which suggest internal diffusion limitations and a heterogeneous distribution of active sites. The use of a compact bed in that study revealed the influence of transport phenomena superimposed on the energetic behavior of the adsorbate–adsorbent system. Other investigations have explored strategies to reduce transport effects by modifying bed architecture. Dilution of the zeolite phase with inert materials such as SiC has been shown to improve thermal uniformity and facilitate desorption by reducing the resistance to bulk flow and internal diffusion [16]. Moreover, the development of mesostructured zeolites with 37 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia enhanced pore accessibility and shorter diffusion paths has further enabled the generation of sharper and more interpretable TPD peaks [17]. Within this framework, the present chapter presents a comparative analysis of the TPD results for benzene and toluene adsorbed on H-Y CBV 100 zeolite using a bed diluted with SiC, as opposed to conventional configurations such as that employed in [2]. The objective is to assess how adsorption temperature and bed structure influence the emergence of diffusive phenomena, and how these, in turn, affect the qualitative interpretation of adsorption energetics and the distribution of active sites within the zeolite. 2.3 Materials and Methods 2.3.1 Physicochemical Properties of Zeolite HY Powder (Zeolyst CBV100) Zeolyst CBV100 is a commercially available HY-type zeolite (NaY) synthesized as a fine white powder, odorless and insoluble, intended for adsorption and catalytic applications [18]. It possesses a faujasite (FAU) framework, characterized by a three-dimensional network of supercages ( 13˚ A) interconnected via 12-membered ring windows ( 7.4˚ A), which enables access for relatively large molecules [19, 20]. The material typically exhibits a Si/Al molar ratio around 5–5.1, providing strong Brønsted acidity associated with framework aluminum sites, making HY zeolites highly active in acid-catalyzed reactions [21, 22]. The surface area is high ( 900m²/g) with significant micropore and total pore volume, although external surface area remains low [21, 23]. CBV100 is utilized extensively in catalytic cracking processes due to its strong acidity and thermal stability. It is also effective in polymer degradation studies, where its strong acid sites accelerate the onset of cracking reactions [22]. Additionally, its structure allows for applications in ionic separations and molecular sieving [24]. Upon dealumination (e.g., via steaming or acid treatments), CBV100 can be converted to HY or H-USY forms, enhancing hydrothermal stability and generating mesoporosity that improves catalytic lifetime and access to active sites [20, 17]. In NOxreduction, copper-loaded CBV100 has demonstrated stable catalytic performance above 300°C in flue-gas conditions [25]. 38 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia 2.3.2 Equipment Description Based on the Master’s thesis by Cindy Ly Tavera [2], the equipment used for her adsorption tests of alkyl aromatics on commercial zeolites was reassembled, as it had been out of operation since 2022. The setup of this system is shown in Figure 20. Figure 20: Gas Phase Differential Reactor Diagram The system can be better understood by first providing an overview of the equipment. At its core, the setup centers around a Differential Reactor, consisting of a quartz U-tube enclosed within a furnace. The furnace supplies the necessary energy to maintain a constant temperature or to follow a programmed temperature ramp, regulated precisely by a temperature controller. To achieve better control over the reaction or adsorption rate, a slight dilution of the feed was essential. For this purpose, two gas inlets were connected: one for a mixed gas stream such like Helium (He), Argon (Ar), Hidrogen (H2), Oxygen (O2), Methane (CH4), Carbon Dioxide (CO2) or Nitrogen (N2) and the other for a pure gas flow. The mixed gas line is fed by 39 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia a gas mixer composed of several Mass Flow Controllers (MFCs), while the second line supplies pure helium or argon directly. The specific use of each gas stream during the experimental procedure will be discussed later. Each of the gas inlet lines is equipped with a temperature controller, a heating tape, and a thermocouple. This configuration allows for precise temperature control during the gas preheating stage, ensuring that the desired temperature is reached inside the reactor under steady-state conditions. Both lines will independently enter an arrangement of two four-way valves (diverter valves), which will allow for different configurations depending on the specific case, as shown below in Figure 21: 40 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia The following mathematical development was considered for the calibration of the CO2-He mixture with a defined outlet composition and a total flow of 50 mL/min: Mass Balance and Composition Definition in Volumetric Fraction ˙mmix = ˙mHe + ˙mCO2(4) yCO2=QCO2 QHe +QCO2 (5) Since the CO2–He system is considered approximately ideal under atmospheric pressure conditions according to [26], the following approximation can be made. However, if the system behaves non-ideally, an excess model should be used to predict VEand calculate the respective Zvalue for each species: ρmixQmix =ρHeQHe +ρCO2QCO2(6) Where Qiis the volumetric flow rate of component iin (mL/min), and ρiis the density in (kg/mL). Assuming ideal gas behavior, the density is approximately linear with respect to temperature and pressure. However, to validate this assumption, the density of both components and the mixture was modeled using Peng-Robinson in Aspen Plus. It was found that He fits a linear trend more closely than CO2, which is consistent with the intermolecular interactions of each gas. The mixing temperature remained at ambient conditions, i.e., 18◦C in Bogot´a, and a pressure of 40 PSIG. 47 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 29: Pure CO2and He Density The mixture density was then estimated as a function of CO2composition at the ambient temperature of Bogot´a, resulting in the following plot: Figure 30: CO2He Mixture Density using Peng-Robinson at 18°C It was found that a fifth-order polynomial regression resulted in errors 48 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia below 0.2%, while linear regression produced errors between 0.9% and 2.4%: Figure 31: CO2He Mixture Relative Error for lineal 5th order regression With the densities of the pure components and the mixture calculated using the Peng-Robinson equation, and knowing that the desired total flow is 50 mL/min with a defined volumetric composition, the system of Equations 5and 6allows determining the necessary flow rates of He and CO2to be introduced into the gas mixer. This results in a direct calibration curve for the percentage openings of both mass flow controllers: yCO2=QCO2 QHe +QCO2 QCO2=yCO2(QCO2+QHe) QHe =QCO21 yCO2 −1(7) Substituting into Equation 6, we obtain: 49 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia QHe =ρmixQmix 1 yCO2 −1ρHe +ρCO2 1 yCO2 −1(8) Using the value of QHe from Equation 7, QCO2can be determined. With these values and the calibration curves, the corresponding opening percentages for each flow controller were obtained. The following results were observed: Table 7: Comparison between theoretical and experimental flow rates for different CO2/He mixtures yCO2(vol/vol) QCO2QHe %Area CO2%Area He Flow Rate Flow Rate Relative (mL/min) (mL/min) Theoretical Experimental Error (%) (cm3/min) (cm3/min) 0.05 2.5 47.1 12.4 82.7 50.00 50.34 0.67 0.10 4.9 44.5 24.7 78.0 50.00 49.93 0.13 0.15 7.4 41.9 37.0 73.5 50.00 50.13 0.25 0.20 9.8 39.4 49.2 69.1 50.00 50.19 0.37 This confirms that the linear adjustment of densities using the PengRobinson model consistently represents the CO2–He mixture behavior. The final calibration curve to obtain a 50 mL/min gas mixture at 40 PSIG is shown below: 50 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 32: CO2He Mixture Valve opening calibration to obtain 50 mL/min at 40 PSIG 2.3.4 TPD Experimentation Description For this section, ammonium-form zeolite HY-type (Zeolyst CBV 100) will be used. This material will first undergo calcination in the U-shaped reactor, followed by either a TPD experiment or a study of diffusional phenomena through the injection of toluene and benzene probes into the zeolite. In both cases, the reactor is configured as follows: 51 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 33: U Reactor Arrangement The reactor has a diameter of 1.5 cm and an approximate length of 30 cm. Inside, a zeolite layer with a thickness of 0.10 cm will be placed in its pure form, or 0.17 cm when diluted with silicon carbide at 4.5 wt%. This zeolite layer is supported by quartz wool, which provides mechanical stability and ensures that the bed remains in place. In addition, a 0.10 cm layer of ground quartz is included to improve temperature distribution within the reactor and to prevent the formation of hot spots, particularly during the calcination step. Lastly, a second layer of ground quartz is added to promote uniform gas flow through the bed and to avoid stagnation zones. This packed bed distribution, will have significant implications in Transport Phenomena along the TPD compared to the only Zeolite packed bed reactor. A fixed bed was loaded with 60 ±5 mg of sieved zeolite mixed with 1273 ±5 mg of SiC, within the Mesh 140–200 range, corresponding to particle sizes of approximately 74 to 105 µm. The bed length was 1.7±0.2 cm. Adsorption of the probe molecule was carried out at three temperatures: 52 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia 25°C, 75°C, and 150°C, based on the previous TPD results reported by [2]. Liquid pulses of 3, 4, or 6 µL of toluene and benzene were injected, rapidly evaporating into a 50 mL/min helium stream (inert carrier gas). Each pulse contained a known amount of the probe molecule, which was transported in vapor phase to the bed. Saturation of the solid was considered to be achieved when the signals of successive pulses showed no significant differences in height or area, indicating that the solid could no longer adsorb additional amounts of the injected species. This method enabled controlled stepwise saturation and was preferred over direct saturation to improve resolution and control of the adsorbed amount, as mentioned in [2]. Once the bed was saturated, thermal desorption of the adsorbed species was performed stating in 298 K. The bed was heated up from there to 623 K at a linear rate of 10 K/min while maintaining a constant flow of helium. The desorption profile was recorded as a function of temperature, allowing analysis of the interaction strength between the probe molecule and the zeolite surface. 2.3.5 Mass Spectrometer Calibration For the calibration of the Mass Spectrometer, the influence of the packed bed on the area under the peak curve at different temperatures was evaluated. To this end, 3 pulse injections of 3, 4, and 6 µL each were performed at three temperatures: 25, 75, and 150°C. The experiments were conducted using both an empty quartz reactor (Blanck) and reactors packed with ground quartz and silicon carbide (SiC). The corresponding peak areas obtained from each experiment are presented in Appendice B. These areas were determined after smoothing the response curves using the Savitzky–Golay method, applying a stronger regression window in regions without peaks and a lighter one around the peak areas. 53 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia 2.4 Results and Discussion 2.4.1 Mass Spectrometer Calibration Results From the data obtained and presented in Appendix B, the following was found: Table 8: Areas under the curve for Toluene pulses at different temperatures and packing materials (% Alkyl Aromatic ·second) T (°C) Pulse (µL) 3 4 6 Sample 25 338 466 795 Blank 25 456 659 943 SiC 75 231 329 497 Blank 75 369 474 723 SiC 150 327 433 713 Blank 150 395 510 766 SiC Table 9: Areas under the curve for Benzene pulses at different temperatures and packing materials (% Alkyl Aromatic ·second) T (°C) Pulse (µL) 3 4 6 Sample 25 681 849 1225 Blank 25 1033 1265 1920 SiC 75 637 861 1208 Blank 75 768 920 1370 SiC 150 815 1002 1293 Blank 150 792 902 1254 SiC When plotting the peak area as a function of the pulse size of the probe at the three temperatures, for both the blank tests and those with SiC (see Appendice B.3), a decreasing trend in peak area with increasing temperature is observed up to a certain point. Beyond that point, the peak area measured by the mass spectrometer increases again in the case of toluene. However, for benzene, the increase in peak area is only observed after 75°C, as shown in Figure 34 and Figure 38: 54 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 34: Area under peaks (Blank sample) for a) Toluene and b) Benzene Figure 35: Area under peaks (SiC sample) for a) Toluene and b) Benzene In the case of Toluene, it is important to consider two key concepts: axial dispersion and transport restriction. The former is directly related to the gas density, while the latter depends on the medium through which the 55 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia gas flows. In this experiment, when the quartz differential reactor is empty, the observed peak area is lower compared to the case where the reactor is filled with a bed of ground quartz mixed with SiC, across all temperature ranges. This is attributed to the reduction of axial dispersion caused by flow restriction, which allows higher concentrations of Toluene to accumulate within the gas phase. Additionally, a temperature-dependent trend is observed: at 25°C, the peak area is highest, both in the blank and with SiC ; it then decreases at 75°C and increases again at 150°C. To understand this behavior, two specific physical aspects of the system must be considered. First, since the injection is done by pulsing, Toluene enters as discrete droplets, which evaporate more or less rapidly depending on the temperature. Second, temperature also plays a direct role in the axial dispersion of the alkylaromatic compound. These two opposing factors produce the observed trend: at lower temperatures, the effect of lower gas density (which increases concentration) dominates; as temperature increases to 75°C, the higher density reduces concentration and thus peak area. However, once the boiling point of toluene (110.6°C) is surpassed, the effect of immediate evaporation becomes dominant. Therefore, at 150°C, although the gas density is higher, the rapid vaporization of the droplets results in significantly higher concentration and consequently, a larger peak area. This affirmation may be seen in Appendice B.3. For the case of Benzene, the same relationship is observed: the peak area is greater with the SiC-packed bed compared to the empty reactor. On the other hand, in the empty reactor, there appears to be no significant difference between 25°C and 75°C due to density effects, likely because the boiling point of Benzene (80°C) is close to both temperatures. However, once the system reaches 150°C, Benzene undergoes rapid vaporization, causing a sudden increase in concentration, which in turn leads to a larger peak area. In the case of the packed bed, the structure appears to disperse the probe more effectively at 25°C, overcoming the effect of density and resulting in an increased peak area. However, at 75°C, the influence of density begins to play a more significant role. 56 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 40: Benzene TPD Gaussian Convolution 25°C n=5 Figure 41: Benzene TPD Gaussian Convolution 75°C n=4 63 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 42: Benzene TPD Gaussian Convolution 150°C n=3 And for Toluene: Figure 43: Toluene TPD Gaussian Convolution 25°C n=4 64 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 44: Toluene TPD Gaussian Convolution 75°C n=5 Figure 45: Toluene TPD Gaussian Convolution 150°C n=3 From there, the following Tables were tabulated: Furthermore, as a final comment in this section, there is a significant difference compared to the data obtained in the thesis by [2], in which a greater number of diffusional limitations can be observed for the H-Y Zeolite 65 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia T (°C) n A µ σ 25 1 0.29 46.45 9.15 2 1.91 131.13 31.20 3 4.22 234.73 19.15 4 3.98 202.47 26.78 5 0.45 330.40 14.12 75 1 2.13 138.59 30.70 2 3.89 193.68 21.26 3 2.68 218.56 15.72 4 0.40 306.20 14.57 150 1 0.46 99.03 27.05 2 2.28 181.58 27.26 3 2.77 213.47 19.72 4 0.13 310.47 12.73 Table 10: Fitted parameters from Gaussian convolution of Benzene TPD curves at different adsorption temperatures. T (°C) n A µ σ 25 1 0.70 101.25 20.40 2 4.33 223.87 49.65 3 6.32 301.49 34.60 4 0.19 405.08 10.47 75 1 0.26 102.17 17.15 2 3.81 275.44 28.79 3 2.69 209.40 45.06 4 0.08 373.07 10.99 5 0.08 425.14 6.75 150 1 0.0001 99.0421 0.0343 2 1.1727 253.8591 19.8697 3 2.1445 211.2541 29.8077 Table 11: Fitted parameters from Gaussian convolution of Toluene TPD curves at different adsorption temperatures. CBV 100, as shown below in Figure 46: 66 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 46: Toluene and Benzene TPD reported by [2] at 25°C as Adsorption Temperature The above comparison with the results obtained in the present thesis is consistent, considering that in Cindy Ly’s experiments, the zeolite bed was not diluted with SiC. This hinders the transport of alkyl aromatics through the bed, resulting in broader TPD profiles that reach significantly higher temperatures than those observed in the present work. In contrast, in this thesis, the cavities formed due to the grain size of the SiC promote enhanced mobility, facilitating transport compared to the undiluted bed experiments. 2.5 Conclusions Based on the results obtained in this section, clear differences were observed in the TPD desorption profiles between benzene and toluene, which are mainly attributed to molecular size and diffusional limitations within the porous structure of the H-Y zeolite CBV 100. Benzene exhibited more defined curves, with well-separated peaks and no shoulders, indicating a desorption process less limited by diffusional phenomena. In contrast, toluene showed peaks with pseudo-peaks (shoulders), especially at adsorption temperatures of 25°C and 75°C, suggesting greater difficulty in the adsorption–desorption 67 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia processes due to internal diffusion barriers and a greater diversity of adsorption sites. When analyzing the effects of adsorption temperature, it was found that the desorption peaks tend to shift toward lower temperatures as the adsorption temperature increases. This behavior can be explained by a lower population of adsorbed molecules, which reduces cooperative interactions and favors earlier desorption. Moreover, the intensity of the peaks was found to decrease significantly as the adsorption temperature increased, indicating that the system tends to remain in the adsorbed state for a shorter time due to the higher kinetic energy of the alkyl aromatic molecules. In the case of benzene, new higher-energy peaks appeared when the adsorption temperatures were higher, which may be understood as the result of weaker interactions with neighboring adsorbed molecules and increased kinetic energy that allows access to micropores that were previously inaccessible. The application of the Gaussian convolution model allowed only for a qualitative (not phenomenological) characterization and identification of the predominant adsorption sites. However, it remains necessary to apply a phenomenological model capable of adequately predicting the observed adsorption–desorption behaviors. Finally, when comparing the results of this thesis with those obtained in Cindy Ly’s thesis, it was found that the absence of SiC dilution in her experimental setup led to greater restrictions in molecular mobility within the system, resulting in broader TPD profiles and significantly higher maximum desorption temperatures. In contrast, the use of a SiC-diluted bed in this work enhanced the diffusion of aromatic compounds through the bed, reducing transport limitations and clarifying more precisely the restrictive effects of adsorption–desorption in the H-Y zeolite CBV 100, particularly with respect to the influence of bed configuration (dilution). 68 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia References [1] N. V. Vlasenko, Y. N. Kochkin, G. M. Telbiz, O. V. Shvets, and P. E. Strizhak, “Insight into the active site nature of zeolite h-bea for liquid phase etherification of isobutylene with ethanol,” Reaction Chemistry & Engineering, vol. 8, no. 6, pp. 1356–1366, 2023. [2] C. L. T. M´endez, “Evaluaci´on de las propiedades texturales y de transporte difusivo de zeolitas y mesoestructuradas sobre el comportamiento catal´ıtico en una reacci´on de ruptura,” Tesis de Maestr´ıa, Universidad Nacional de Colombia, Bogot´a, Colombia, 2020, facultad de Ciencias, Departamento de Qu´ımica. [3] X. Guo, L. Guo, Y. Zeng, R. Kosol, X. Gao, Y. Yoneyama, G. Yang, and N. T. , Catalytic oligomerization of isobutyl alcohol to jet fuels over dealuminated zeolite Beta. Elsevier Ltd, 2021. [4] T. Yoshioka, R. Ohnishi, T. Hirose, Y. Kamimura, H. Okubo, Y. Kuroda, Y. Kuroda, Y. K. Takahashi, R. A. Sasaki, M. Takata, Y. Wakihara, and M. Takata, “Dealumination of small-pore zeolites through pore-opening migration process with the aid of pore-filler stabilization,” Science Advances, vol. 8, no. 25, p. eabo3093, 2022. [Online]. Available: https://www.science.org/doi/10.1126/sciadv.abo3093 [5] A. Corma, J. P´erez-Pariente, and C. Mart´ınez, “Structure and catalytic behavior of beta zeolite,” Journal of Catalysis, vol. 148, no. 2, pp. 569– 574, 1994. [6] A. Corma, “State of the art and future challenges of zeolites as catalysts,” Journal of Catalysis, vol. 216, no. 1–2, pp. 298–312, 2003. [7] S. I. Zones, “Conversion of faujasite to high silica zeolite beta via seeding,” Journal of the Chemical Society, Faraday Transactions, vol. 87, no. 24, pp. 3709–3716, 1991. [8] T. F. D. Jr., “The implications of the discovery of zeolite beta,” Microporous and Mesoporous Materials, vol. 35–36, pp. 245–252, 2000. [9] R. M. Barrer, Hydrothermal Chemistry of Zeolites. Academic Press, 1982. 69 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia [10] J. Weitkamp, “Zeolites and catalysis,” Solid State Ionics, vol. 131, no. 1–2, pp. 175–188, 2000. [11] Y. Fan, X. Bao, X. Lin, G. Shi, and H. Liu, “Acidity adjustment of hzsm-5 zeolites by dealumination and realumination with steaming and citric acid treatments,” The Journal of Physical Chemistry B, vol. 110, no. 31, pp. 15 411–15 416, 2006. [12] J. P´erez-Ram´ırez, C. Christensen, K. Johannsen, and I. Schmidt, “Hierarchical zeolites: Enhanced molecular diffusion and catalysis of alkylaromatics,” Catalysis Today, vol. 82, pp. 145–154, 2003. [13] M. Haouas, F. Taulelle, and J. Sommer, “Recent advances in understanding the structure and dynamics of zeolite catalysts by solid-state nmr,” Chemical Society Reviews, vol. 37, pp. 1116–1130, 2008. [14] Y. Li and J. Yu, “New stories of zeolite structures: their descriptions, determinations, predictions, and evaluations,” Chemical Reviews, vol. 114, no. 14, pp. 7268–7316, 2016. [15] T. Nguyen and E. Iglesia, “Transport phenomena in packed beds of zeolite crystals: effects of porosity and crystal size on the tpd profiles of alkanes and aromatics,” Journal of Catalysis, vol. 323, pp. 34–46, 2015. [16] E. Hern´andez-Maldonado and R. Yang, “Desorption kinetics and diffusional limitations in temperature-programmed desorption of volatile organic compounds in zeolites,” Industrial Engineering Chemistry Research, vol. 42, no. 13, pp. 3103–3112, 2003. [17] T. Yokoi, K. Mochizuki, and T. Tatsumi, “Design of mesostructured zeolites with improved accessibility and catalytic performance,” Microporous and Mesoporous Materials, vol. 280, pp. 203–211, 2019. [18] Zeolyst International, “Safety data sheet: Zeolite cbv100 powder (nay),” 2023, available at: https://www.zeolyst.com/products/ hy-zeolite-cbv100. [19] D. W. Breck, Zeolite Molecular Sieves: Structure, Chemistry, and Use. New York: Wiley-Interscience, 1974. 70 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia [20] M. Nichterwitz, M. Borchardt, and T. Thielmann, “Solvent-free hierarchization of zeolites by carbochlorination,” Journal of Materials Chemistry A, vol. 5, pp. 221–229, 2017. [21] Zeolyst International, “Product information: Zeolite hy cbv100,” 2022, available at: https://www.knowde.com/stores/zeolyst-international/ products/cbv-100. [22] C. Blazek, A. Klamt, and R. Gl¨aser, “Catalytic cracking of polymers over zeolites: Effect of acidity and pore size,” Microporous and Mesoporous Materials, vol. 94, no. 1-3, pp. 129–138, 2006. [23] A. Rodriguez and L. Pe˜na, “Textural properties and structural behavior of zeolyst nay cbv100,” SSRN Electronic Journal, 2023, available at: https://ssrn.com/abstract=4134264. [24] S. Turner, M. J. Hudson, and J. Klinowski, “Characterization of zeolite y as a reference material for industrial applications,” Microporous and Mesoporous Materials, vol. 103, no. 1–3, pp. 219–226, 2007. [25] D. Garcia, E. Martinez, and J. Lopez, “Ammonia-scr performance of cu/hy zeolites derived from cbv100 under flue-gas conditions,” Industrial & Engineering Chemistry Research, vol. 63, no. 2, pp. 401–409, 2024. [26] J. D´avila, Z. E. Heinemann, and M. Kriebernegg, “Accurate calculations of compressibility factor for pure gases and gas mixtures,” in Proceedings of the 5th European Conference on the Mathematics of Oil Recovery, Leoben, Austria, Sep. 1996, pp. –, experimental values of Zfor CO2and He at 295 K. 71 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia APPENDICES A Standard Operating Procedure’s (SOP’s) This section includes the Standard Operating Procedures (SOPs) for the operation of the Mass Spectrometer and the execution of the adsorption and desorption runs in the quartz differential reactor, developed in this thesis. A.1 Standard Operating Procedure (SOP) - Differential Reactor Operation 1. Check that all plugs are connected to their respective controllers (Box, He/Ar, and Valve 1) and to their power sources. 2. Turn on the heating tapes with the controllers and set the set point to preheat the gas. 3. Turn on the furnace and program the corresponding temperature ramps according to the following protocol: a. Press the up and down arrows simultaneously. Use Mode to go to Global and check that RATE is in Ptyp. Press DSPY. b. Program the ramp: Press Mode, then use the up arrow to go to Prog, select File 1 with Mode, then step 1 with Mode,Mode (enter set point), Mode (input desired set point value), Mode (set the heating rate in °C/min), Mode to go to File 1, step 2, and set SOAH to maintain temperature for xhours, yminutes, zseconds. c. Once the heating ramp is programmed, turn on the furnace using Hold/Run (hold and select the file you want to activate). d. Ensure the furnace controller switch is pointing upwards. Note: It is important to heat slowly, with a maximum rate of 1 °C/min, as overshooting the set point is possible due to the furnace’s electric power. 4. Once the furnace is on, open the valves of the gases to be used to 40 psi. 72 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 54: Smoothed Toluene peaks at 75°C (SiC sample) Figure 55: Smoothed Toluene peaks at 150°C (SiC sample) 79 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 56: Smoothed Benzene peaks at 25°C (SiC sample) Figure 57: Smoothed Benzene peaks at 75°C (SiC sample) 80 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 58: Smoothed Benzene peaks at 150°C (SiC sample) 81 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia B.2 Area under smoothed peaks for Mass Spectrometer Calibration Figure 59: Area under Toluene peaks at 25°C (Blank sample) 82 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 60: Area under Toluene peaks at 75°C (Blank sample) Figure 61: Area under Toluene peaks at 150°C (Blank sample) 83 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 62: Area under Benzene peaks at 25°C (Blank sample) 84 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 63: Area under Benzene peaks at 75°C (Blank sample) Figure 64: Area under Benzene peaks at 150°C (Blank sample) 85 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 65: Area under Toluene peaks at 25°C (SiC sample) Figure 66: Area under Toluene peaks at 75°C (SiC sample) 86 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 67: Area under Toluene peaks at 150°C (SiC sample) Figure 68: Area under Benzene peaks at 25°C (SiC sample) 87 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 69: Area under Benzene peaks at 75°C (SiC sample) Figure 70: Area under Benzene peaks at 150°C (SiC sample) 88 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 81: Benzene 25°C Adsorption TPD Figure 82: Benzene 75°C Adsorption TPD 95 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 83: Benzene 150°C Adsorption TPD Figure 84: Toluene injection in 3µL pulses at 25°C followed by TPD analysis. 96 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 85: Toluene injection in 3µL pulses at 75°C followed by TPD analysis. Figure 86: Toluene injection in 3µL pulses at 150°C followed by TPD analysis. 97 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 87: Toluene 25°C Adsorption TPD Figure 88: Toluene 75°C Adsorption TPD 98 Bachelor Thesis - Felipe Mart´ınez Barreto Universidad Nacional de Colombia Figure 89: Toluene 150°C Adsorption TPD 99