The 28th Special CU-af Seminar 2020

The 28th Special CU-af Seminar 2020 October 20, 2020 the presence of TSF within the structures. In GPTMS functionalized TSF, a band at 1100.65 cm-1 and a weak band at 850.67 cm-1 assigned to Si-O-Si and Si-OH respectively were seen (Fig. 2, 3rd line from bottom and top line). This indicates that some of the methoxysilane groups in GPTMS were already hydrolyzed to form silanol groups, and some of these groups further condensed to become Si-O-Si. On the epoxy end of GPTMS, the spectrum of the GPTMS-functionalized TSF showed disappearance of the bands assigned to the bonds in the epoxy ring (Fig. 2, 3rd line from bottom). These results confirmed that the epoxy ring was opened, and used for bonding with amine groups in TSF. Furthermore, the more prominent Si-O-Si and Si-OH bands were observed in TSF-BCSG hybrids. That was the result of the addition of, BCSG to the system. (Fig. 2, top line) Figure 2. The FTIR spectra of GPTMS, SF, GPTMS-functionalized TSF, and TSF-BCSG 10. Morphology and pore structure The morphology of the scaffolds was observed under SEM (Fig. 3). Both TSF-BCSG 0 and TSF-BCSG 10 had pores with interconnection. The crosslinked scaffolds had a larger pore diameter of 177± 74 µm and the pores were more evenly distributed, while the uncrosslinked scaffolds had a smaller pore diameter of 128±36 µm. This suggests that GPTMS affects the morphology of the scaffolds. The greater size of the pores in TSF-BCSG 10 could be explained by the fact that GPTMS is a foaming agent.11 Figure 3. The electron micrographs of TSF-BCSG 0 (A) and TSF-BCSG 10 (B). 39

The 28th Special CU-af Seminar 2020 October 20, 2020 Stability in water The stability of the scaffolds was determined by its insolubility in water (Fig. 4). The stability is hence inversely proportional to weight loss. Apparently, TSF-BCSG 0 lost their weight twice as much did the TSF-BCSG 10. That was because the crosslinking stabilized the scaffolds by forming covalent linkages between the silicate network and the fibroin network. In contrast, the two networks in TSF-BCSG 0 had not been crosslinked, and thus were hold together only by physical entanglement and weak intermolecular interactions such as van der Waals forces. Figure 4. Percentage of weight loss of TSF-BCSG 0 and TSF-BCSG 10 after immersion in de-ionised water for 24 h at 37 °C. In vitro biodegradation Degradation of the scaffolds in protease XIV reaffirmed the success of crosslinking. Both types of the scaffolds lost their weight overtime in the protease XIV solution (Fig. 5). However, TSF-BCSG 0 always lost the weight more than TSF-BCSG 10 over 10 days, reconfirming that the crosslinking by GPTMS stabilized the scaffolds. Interestingly, degradation of the two types of the scaffolds during the first day appeared to be most different, as TSF-BCSG 0 lost more than half of its weight while TSF-BCSG 10 lost less than 20% of its weight. This agrees with our previously described experiment. After that, the rate of weight lost progressively decreased. This experiment provides a starting point for tuning the degradation rate. Figure 5. Percentage of remaining weight of TSF-BCSG 0 and TSF-BCSG 10 after in vitro degradation in protease XIV solution for 10 days at 37 °C. 40

The 28th Special CU-af Seminar 2020 October 20, 2020 Conclusion The TSF-BCSG hybrid scaffolds were successfully obtained using the foaming techniques. The FTIR results confirmed the crosslinking of TSF and BCSG through GPTMS. The resulting GPTMS-crosslinked scaffolds showed the desirable interconnected pores. The crosslinked scaffolds also showed enhanced stability in water, which is necessary for implantation. The work herein provides a foundation for reinforcing brittle bioactive glass with proteinaceous polymers. Examination of bioactivity and biocompatibility is needed in the future. References 1. Khan, Y.; Yaszemski, MJ.; Mikos, AG.; Laurencin, CT. J. Bone Jt.Surg. 2008, 90, 36-42. 2. Guenaelle, B.; David, M.; Magali, C.; Luc, M.; Laurence, V. TissueEng. 2012, 21, 133-156. 3. Li, G.; Liu, H.; Wang, J. Mater. Sci. Eng. C. 2012, 32, 627-636. 4. Kundu, B.; Rajkhowa, R.; Kundu, S.; Wang, X. Adv. Drug Deliv. Rev.2013, 65, 457-470. 5. Fu, Q.; Saiz, E.; Rahaman, M.; Tomsia, A. Mater. Sci. Eng. C. 2011,31, 1245-1256. 6. Aza, PN.; Guitan, F.; Aza, S.Scr. Metall. Mater. 1994, 31, 1001–1005. 7. Pereira, MM.; Jones, JR.; Orefice, RL.; Hench, LL. J. Mater. Sci.2005, 16, 1045-1050. 8. Ren, L.; Tsuru, K.; Hayakava, S.; Osaka, A. J. Sol-gel Sci. Technol.2001, 21, 115-121. 9. Mahony, O.; Tsigkou, O.; Ionescu, C.; Minelli, C.; Hanly, R.; Ling,L.; Smith, M.E.; Stevens, M.M.; Jones, J.R. Adv. Funct. Mater. 2010, 20, 3835-3845. 10. Kim, UJ.; Park, J.; Kim, HJ.; Wada, M., Kaplan, DL. Biomaterials.2005, 26, 2775-2785. 11. Lei, B.; Wang, L.; Chen, X.; Chae, S. J. Mater. Chem. B. 2013, 1,5153-5162. 41



Synthesis and Applications of Citronella Oil Nanoemulsion Assoc. Prof. Dr. Kawee Srikulkit



The 28th Special CU-af Seminar 2020 October 20, 2020 Synthesis and Applications of Citronella Oil Nanoemulsion Assoc. Prof. Dr. Kawee Srikulkit Abstract Citronella oil/PMMA core was prepared by suspension polymerization in PVA protective colloid, MMA monomer and benzoyl peroxide as an initiator. Two ratios of 1 MMA : 2 oil and 1 MMA : 1 oil were investigated at two different stirring speeds of 600 rpm and 1000 rpm, respectively. Optical microscope showed that 1 : 2 ratio and stirring speed of 600 rpm produced larger citronella oil/PMMA core than 1 : 1 ratio due to larger citronella oil/MMA droplet size. To produce a small citronella oil/PMMA core, low ratio or high stirring speed or combination was recommended. Microcapsule of citronella oil/PMMA core was achieved by shell formation of melamine formaldehyde polymer. FTIR analysis showed that spectral fingerprint of encapsulated citronella oil/PMMA core disappeared, confirming the successful sheath/core formation. Department of Materials Science Faculty of Science, Chulalongkorn University Bangkok, Thailand 45

The 28th Special CU-af Seminar 2020 October 20, 2020 Introduction and Objectives The plant based citronella oil is widely used for effective mosquito repellent. Lemongrass growing in tropical area including Thailand is a main source of citronella oil. Apart from insect repellency, citronella oil has strong antifungal and antimicrobial activities. Citronella oil is popular for topical use. To be continually effective, most citronella oil formulations require to be reapplied onto skin every 30-60 minutes due to their volatility. In order to prolong citronella oil release, encapsulation techniques are involved including host-guest complexes (beta-cyclodextrin) and microencapsulations. Cyclodextrins exhibit hydrophobic cavity which can form complex with citronella oil. However, the capacity of hydrophobic cavity is limited. Moreover, cyclodextrin is prone to hydrolysis (ring break down) as well as expensive, deterring an interest in applying for citronella oil formulation for mosquito repellent. Microencapsulation techniques are alternative ways in order to prolong the release of volatile citronella oil. Inside a microcapsule, citronella oil is referred as a core surrounding by a wall called shell, coating or membrane. Polymeric materials such as ethyl cellulose, polyvinyl alcohol, sodium alginate, and gelatin are a typical coatings. Techniques to manufacture microcapsules are classified into three major methods including physical methods, physicochemical methods and chemical methods. Physical methods are as follows: pan coating, air-suspension coating, centrifugal extrusion, vibrational nozzle, and spry-drying. Physicochemical methods include ionotropic gelation and coacervation-phase separation. For chemical methods, they are divided into 4 different methods such as interfacial polycondensation, interfacial cross-linking, in-situ polymerization and matrix polymerization. The application of microcapsules onto textiles requires binders in order to achieve durability property. Therefore, in this work, suspension polymerization combined with encapsulation technique was studied in order to prepare citronella oil microcapsule, aiming at the microcapsule with an ease of textile application without requirement of binders. Experimental Materials and Method Citronella oil was bought from local supplier. Methyl Mathacrylate monomer was kindly provided by Thai MMA co., Ltd. Commercial grade polyvinyl alcohol (DP = 1700 -1800, DD = 85%) was purchased from a local supplier (N.P.S. CHEM (Thailand) Co., Ltd. Benzoyl peroxide was purchased from Fluka. Melamine formaldehyde condensate under the trade name of ACROFIX was purchased from a local supplier. Synthesis of Citronella Oil Microcapsule 500 mL of 2 wt% PVA solution was prepared in a 1000 ml glass beaker and placed on a magnetic stirrer at 600 rpm. Then, a mixture solution of methyl methacrylate : citronella oil 1 : 2 by volume or 25 mL:50 mL) in a presence of benzoyl peroxide (0.5 wt% of MMA) was slowly added under stirring. The reaction temperature was raised to 75 oC and kept at the constant temperature for 3 h. Then, the system was cooled down to room temperature. At the room temperature, 10 ml of ACROFIX was added and pH value was adjusted to 2-4 by adding hydrochloric acid. The encapsulation was continued for 24 h under stirring at room temperature. Effect of ACROFIX amounts of 20 ml and 30 was investigated. The resulting microcapsule was characterized using optical microscope. In separate experiment, MMA : citronella oil ratios of 1 : 2, and 1 :1 and stirring speeds of 600 rpm and 1000 rpm were carried out. 46

The 28th Special CU-af Seminar 2020 October 20, 2020 Results and Discussion Effect of MMA : citronella oil ratios on morphology of citronella oil core Polymerization of MMA in the presence of citronella oil using suspension technique results in citronella oil core. Note that polyvinyl alcohol was required in order to stabilize MMA droplet during polymerization reaction. MMA : citronella oil volume ratio is a factor affecting size and morphology of oil core. In this experiment, two ratios of 1 : 2 and 1 : 1 were investigated at two different stirring speeds. The morphology was observed by the optical microscope as shown in Fig. 1 At 1 : 2 ratio and stirring speed of 600 rpm, citronella oil cores with mixed sizes is spherical. In a similar manner, at 1 : 2 ratio and stirring speed of 1000 rpm, spherical shapes with smaller sizes are observed due to higher stirring speed. It is expected that the higher the stirring speed the smaller the core size. In case of 1: 1 ratio and stirring speed of 600 rpm, its average core size is much smaller when compared to 1 : 2 ratio. Also, core density is much higher. It was assumed that 1 : 1 ratio exhibited small and stable droplet in PVA protective colloid, resulting in small core size with a high number of cores. The reaction mechanism occurring in the droplet involved polymerization of MMA monomer initiated by BPO to become poly (methyl mathacrylate, PMMA). The PMMA then entrapped citronella oil, leading citronella oil core. Fig. 1 OM photographs of citronella oil/PMMA cores Encapsulation of citronella oil core Melamine formaldehyde polymer was employed as a precipitated shell material. Encapsulation of citronella oil core was carried out using melamine formaldehyde condensate resin. The resin was added into stirred citronella oil core dispersion and pH value was adjusted to 2-3 by HCl. At beginning, the resin was soluble in water continuous phase. Then, polycondensation reaction catalyzed by acid underwent, resulting in melamine formaldehyde polymer which became insoluble in water phase. As a result, phase separation occurred and then the melamine formaldehyde polymer precipitated as a coating shell covering the oil core. 47

The 28th Special CU-af Seminar 2020 October 20, 2020 Fig. 2a shows microcapsules of 1 MMA : 2 oil core at 600 rpm. It is clearly seen that melamine formaldehyde polymer shell is present on the outer core. In case of the large core, incomplete encapsulation is found. This problem can be solved when stirring speed of citronella oil core reaction increased from 600 rpm to 1000 rpm, resulting in smaller citronella oil core. As a result, complete encapsulation was obtained. For1MMA:1oilcore,thesmallestmicrocapsuleswithwellsize distribution was achieved as shown in Fig. 2c. Based on the obtained results, the encapsulation mechanism is proposed as shown in Scheme 1. Fig. 2 Microcapsules of citronella oil/PMMA cores by melamine formaldehyde polymer 48

The 28th Special CU-af Seminar 2020 October 20, 2020 Scheme 1. Schematic diagram of citronella oil/PMMA core encapsulated by melamine formaldehyde polymer. FTIR analysis Fig. 3 shows the comparison between FTIR spectrum of citronella oil and citronella oil core. The overlapping peaks appearing around 1700 cm-1 corresponds to oil carbonyl group as well as PMMA ester group. As seen, fingerprints of oil and PMMA are similar due to the similarity of bonds and functional groups. Therefore, they are miscible with each other, resulting in homogenous core. After encapsulation with melamine formaldehyde polymer, the citronella oil/ PMMAcore was encapsulated. ItcanbeconfirmedbyFTIRanalysisthatthefunctionalgroupsofcitronella oil/PMMA disappear from the FTIR spectrum due to the fact that melamine formaldehyde polymer shell is free from carbonyl functional group. It can be concluded that encapsulation of citronella oil/PPM core was successful. Fig.3 FTIR spectra of citronella oil/PMMA core and its microcapsule Conclusion Citronella oil/PMMA core was prepared by suspension polymerization of MMA containing citronella oil using BPO as an initiator. Core size and size distribution were controlled by MMA: citronella oil ratio and stirring speed; the higher the ratio the larger the core size and the higher the stirring speed the smaller the core size. Microcapsule of citronella oil/PMMA core was achieved by shell formation of melamine formaldehyde polymer. FTIR analysis showed that spectral fingerprint of encapsulated citronella oil/PMMA core disappeared, confirming the successful sheath/core formation. References 1. M.M. Miro Specos, J.J. Garcia, P. Marino, and L.G. Hermida, Microencapsulated citronella oil for mosquito repellent finishing of cotton textiles, Transactions of the Royal Society of Tropical Medicine and Hygiene, Vol. 104, issue 10 (2010) 653-658. 49

The 28th Special CU-af Seminar 2020 October 20, 2020 2. J. Azzi, L. Auezova, P.E Danjou, S.Fourmentin, and H. Greige-Gerges First Evaluation of Drug-in-Cycyclodextrin in Liposomes as an Encapsulating System for Nerolidol, Food Chemistry, Vol. 255 (2018) 399-404. 3. M. Kfoury, L. Auezova, H. Greige-Gerges, and S. Foumentin, Promising Applications of Cycyclodextrins in Food: Improvement of Essential Oils Retention, Controlled Release and Antiradical Activity, Carbohydrate Polymers, Vol. 131, (2015) 264-272 4. E.F. de Matos, B.S. Scopel, and A. Dettmer, Citronella Essential Oil Microencapsulation by Complex Coacervation with Leather Waste Gelatin and Sodium Alginate, Journal of Environmental Chemical Engineering, Vol. 6 issue 2 (2018) 1989-1994. 5. A. D. Ribeiro, J. Marques, M. Forte, F.C. Correia, and C.J. Tavares, Microencapsulati on of Citronella Oil for Solar Activated ControlledRelease as an Insect Repellent., Applied Materials Today, Vol. 5 (2016) 90-97. 6. A. Gharsallaoui, G. Roudaut, O. Chambin, A. Voilley, and R. Saurel,Applications of Spry-drying in Microencapsulation of Food Ingredients: An Overview, Vol. 9 (2007) 1107-1121. 7. J.H. Xin, and X.W. Wang, Insect Repellent Textiles (Book Chapter),Engineering of High Performance Textiles, (2018) 335-348 8. P.K. Inguva, S.M. Ooi. P.M. Desai, and P.W. Heng, Encapsulation of Volatiles by Homogenized Partially-Crosslinked Alginates, Vol. 496, (2015) 709-716. 9. M. Tavares, M.R.M da Silva, L.B. de Siqueira, R. Rodrigues, and E. Ricci-Júnior, Trends in Insect Formulations : A review, International Journal of Pharmaceutics, Vol. 539, (2018), 190-209. 10. A.P.B. Balaji, A. Ashu,S. Manigandan, T.P. Sastry, and N. Chandrasekaran., Polymeric Nanoencapsulation of Insect Repellent:Evaluation of Bioefficacy on Culex Quinquefasciatus Mosquito Pupulation and Effective Impregnation onto Cotton Fabrics for Insect Repellent Clothing, Vol. 29, (2017) 517-527. 11. M. Christofoli, E.C.C. Costa, K.U. Bicalho, V. Cássia Domingues, and C. Melo Cazal, Insecticidal Effect of Nanoencapsulated Essential Oils from Zanthoxylum Rhoifolium (Rutaceae) in BemisiaTabaci Populations, Industrial Crops and Products, Vol 70., (2015) 301-308. 12. J. Rodríguez, M.J. Martín, M.A. Ruiz, and B. Clares., Current Encapsulation Strategies for Bioactive Oils : From Alimentary to Pharmaceutical Perspectives, Food Research International, Vol 83, (2016) 41-59. 13. M. Montazer, and T. Harifi, Nanofinishes for Protective Textiles (book chapter), Nanofinishing of Textile Materials, (2018) 265-294. 50

Sequencing and Characterization of a hypovirulence dsRNA virus from an oomycete Dr. Thanyanuch Kriangkripipat



The 28th Special CU-af Seminar 2020 October 20, 2020 Sequencing and Characterization of a hypovirulence dsRNA virus from an oomycete Dr. Thanyanuch Kriangkripipat Abstract A pythiaceous fungus, isolate R84, contained three dsRNA segments with approximately 8.0, 3.7 and 2.3 kilobases was isolated from rubber tree with symptoms of abnormal leaf fall from eastern Thailand. Phytophthora spp. were the only known agent of leaf fall disease in Thailand. The isolate R84 failed to sporulate on media commonly used for sporulation in Phytophthora. This suggested that the virus might confer the trait. However, phylogenetic analysis based on sequences of the internal transcribed spacer rDNA region showed that the isolate R84 was Phytopythium cucurbitacearum. Virus curing using single zoospore isolation method showed that strain R84 had at least 2 viruses. The growth rate, sexual and asexual reproduction, hymexazol resistance and leaf infection of the isolate R84 suggested that the dsRNA infection was in a symptomless manner. This was the first report of dsRNA elements in Phy. cucurbitacearum and of Phy. cucurbitacearum on para rubber trees. Department of Microbiology Faculty of Science, Chulalongkorn University Bangkok, Thailand 55

The 28th Special CU-af Seminar 2020 October 20, 2020 Introduction and Objectives Thailand is the world’s largest rubber exporter, amounting to 36.2% of world exports (1). Para rubber tree requires warm temperature around 25-28°C, well-distributed rainfall of 100-150 days throughout the year with 2000–4000 mm of annual precipitation, and high humidity ranging between 67-82% for high rubber yield (2, 3). These optimal conditions for latex production are similar to the climate in the south and the east region of Thailand. The most planted rubber clone in Thailand is RRIM 600, which has been recommended by Rubber Research Institute of Thailand (RRIT) for high latex yield (4, 5). However, the rubber clone is susceptible to various diseases, especially Phytophthora spp. causing abnormal leaf fall (ALF) and black stripe disease (6). Pythiaceous fungi is the common name of the fungus-like organisms in class Oomycetes, order Peronosporales, including Phytophthora (P.), Pythium (Py.) and Phytopythium (Phy.) (7, 8). Physiology and habitat of oomycetes are similar to those of filamentous fungi, but they are more related to brown-algae and diatoms. Despite the fact, oomycetes are actively studied by mycologists. Pythiaceous fungi are recognized among plant pathologists as devastating plant pathogens. P. infestans is the most notorious one (9, 10). Species reported causing diseases in para rubber tree are members of family Peronosporaceae, including P. capsici, P. citrophthora, P. meadii, P. nicotianae, P. palmivora and family Pythiaceae, including Phy. vexans (previously named Py. vexans), all 6 species of the genus Phytophthora have been reported worldwide, while Phy. vexans was mentioned only in China (11-13). The most common species infecting para rubber tree in Thailand are P. palmivora and P. botryosa (14). Species in the Peronosporales produce different types of spores to complete their life cycle. In the asexual stage, sporangia or zoosporangia are produced and germinate through the plant’s stomata, releasing motile zoospores, which are responsible for primary infection and disease outbreak in the rainy season. In addition, the chlamydospore, an asexual spore, and the oospore, an sexual spore, are responsible for long term survival under stress conditions (15). ALF and black stripe diseases caused by Phytophthora may reduce the field latex yield by up to 30–50% (16). Mycovirus is the virus that infects fungi and has been commonly found in all major taxonomic groupsoffungi(17).Mostmycoviruseshavethegenomesofdouble-strandedRNA(dsRNA) or positive single-stranded RNA (ssRNA) (18). Almost all mycoviruses live quietly without interfering with their hosts but some shows hypovirulence or hypervirulence (18). Introduction of Cryphonectria hypoviruses in field condition showed a successful control of chestnut disease in Europe that initiated the interest in mycovirus research (19). RNA viruses have been reported in few of pythiaceous fungi. Phytophthora endornavirus 1 (PEV1) was the first virus found infecting Phytophthora sp. from Douglas fir in USA which later seen in P. ramorum, causing the sudden oak death (20-22). Cai and colleagues detected 4 patterns of dsRNA segments in 9 P. infestans isolates from USA and Mexico and named PiRVs 1-4 (23). Double-stranded RNA element was reported in 4 species of the genus Pythium, Py. butleri, Py. irregulare, Py. nunn, and Py. polare. No virus-like particle (VLP) was found in Py. butleri but the dsRNA was detected using antisera against Polyinosinic: poly-cytidylic acid which is similar to dsRNA structure (24, 25). Py. irregulare causing disease in cucumber was detected containing dsRNA both with or without VLP (26), while all Australian isolates with different dsRNA patterns have VLP (27). Pythium nunn virus 1, isolated from Py. nunn, is a virus with a bipartite dsRNA genome classified in the genus Ganmmapartitivirus in the family Partitiviridae (28). Py. polare strain OPU1176 has 3 monopartite dsRNAviruses, Pythium polare RNAvirus 1 (PpRV1) related to the unclassified arthropod toti-like viruses, Pythium polare RNA virus 2 (PpRV2) related to Beihai barnacle virus 15, and Pythium 56

The 28th Special CU-af Seminar 2020 October 20, 2020 polare bunya-like RNA virus 1 (PpBRV1) (29). However, virus of pythiaceous fungi associated with para rubber tree has never been reported. An oomycete strain R84 was isolated from para rubber tree in Chanthaburi province, Thailand. The strain R84 has tree dsRNA segments of 2.3, 3.7 and 6.6 kilobase pairs in length. The strain could not produce sporangium on many media recommended for Phytophthora sporulation (30, 31). The virus may confer failure to produce sporangia of the strain R84. This study aimed to characterize dsRNA elements, expected to be a virus of the strain R84, to identify the host, and to study the effects of viral infection on the host. Materials and Methods Pythiaceae storage For short-term storage, agar discs from the edge of a colony grown on 5%V8 agar (5%V8A; The Campbell Soup Co.; USA) (30) were added into sterilized distilled water and kept at 25°C in the dark. The virus-containing isolate, isolate R84, was deposited at the culture collection of the Department of Microbiology, Faculty of Science, Chulalongkorn University, Bangkok, Thailand in the Thailand Bioresource Research Center (TBRC) as MSCU 1025. Double-stranded RNA extraction The methods for isolation of dsRNA were adapted from Das and colleague (32). using cellulose fiber (C6288; Sigma; USA). Double-stranded RNA banding patterns were determined by electrophoresis on 1% agarose gel in TAE buffer, stained with ethidium bromide, and visualized under UV illumination. Nuclease treatment of dsRNA The RNA elements were treated with DNase I, RNase A or S1 nuclease. Each sample of 200 ng of dsRNA was treated with 1 µg/mL of RNase A (Bio Basic; Canada) in 0.1× SSC (15 mM NaCl, 1.5 mM sodium citrate) and 2× SSC (300 mM NaCl, 30 mM sodium citrate) for treatments as low salt and high salt conditions, respectively, and then incubated at room temperature for 10 min. S1 nuclease 0.1 U (Promega; USA) or DNase I 0.1 U (Bio Basic; Canada) was added in the 20 µL reaction comprising 500 ng of dsRNA and buffers and then incubated at 37ºC for 30 min. The nuclease-treated reactions were cleaned up as described by Das and colleague (32) and determined using electrophoresis on 1% agarose gel in TAE buffer, stained with ethidium bromide and visualized under UV illumination. Partial sequence of dsRNA For cDNA synthesis, 1 µg of S1 nuclease-treated dsRNA was used as a template for first-strand cDNA synthesis using SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen, USA) with random hexamer primers or primers designed from the RNA-dependent RNA polymerase (RdRp) gene of Phytophthora viruses and Pythium viruses. Second-strand cDNA was constructed using protocol adapted from D’Alessio and Gerard (33). The cDNA was ligated into pMiniT plasmid from NEB PCR cloning kit (New England Biolabs, USA). The ligation reaction was transformed into E. coli DH10B competent cells (New England Biolabs, USA). Plasmids with an insert fragment were sent for sequencing by Macrogen Inc., Korea. The sequencing results were blast with GenBank in NCBI (National Center for Biotechnology Information). Protein sequence was predicted using ExPASy-Translate tool (http://web.expasy. org/translate/) and compared the similar sequence in database using BLASTP (NCBI). Protein 57

The 28th Special CU-af Seminar 2020 October 20, 2020 alignment was proceeded using Clustal Omega software (34). Table 1 Primers Identification of oomycetes Colony patterns and asexual and sexual organs produced by isolate R84 were determined on 5%V8A with or without fungicides and antibiotics, potato dextrose agar (PDA) and black bean agar (31) at 25°C. Sporangia production was induced via a bating technique (co-incubation of culture agar plugs with black sesame seeds in sterile natural pond water) and then incubated at 25°C under fluorescent light. Molecular identification was carried out using ITS1/ITS4 primers (36). The amplicons from ITS1/ITS4 reaction were ligated into a TA cloning vector, RBC TA cloning kit (RBC Bioscience; Taiwan) and transformed into Escherichia coli DH5α. Plasmid clones containing the ITS fragment were extracted using a HiYield Plasmid Mini Kit (RBC Bioscience; Taiwan) and sequenced by Macrogen Inc (South Korea). The sequencing results were then subjected to blast analysis with the GenBank database in the National Center for Biotechnology Information (NCBI; USA).Multiple sequences alignment were performed using the Clustal Omega software (34) and a phylogenetic tree was constructed using the Mega X program and the neighbor-joining method with 1,000 bootstrap replications (37). Koch’s postulates To determine if the dsRNAs affected the virulence of the host, zoospore suspensions of the dsRNA-containing isolate and a dsRNA-free isolate were inoculated onto detached para rubber leaves. The dsRNA-free strain, Phy. cucurbitacearum isolate L69, was isolated from a rubber farm in Rayong province, Thailand (12°54’55.0”N 101°28’52.4”E). Leaves of the RRIM600 variety were collected from an ALF-free organic farm (13°23’19.0”N 101°13’07.8”E) in Chonburi province, Thailand. The leaves were surface sterilized by rinsing with tap water, soaking in 0.5% hypochlorite for 45 s, washing twice with sterile deionized water and then blotting dry on sterile gauze. A sample of 20 mL of zoospore suspension (1 × 104 spores per Petri dish) in sterile deionized water were added to each leaf. The plates were incubated at 25°C under fluorescent light and determined daily for 7 d. On day 7, tissue from the edge of an actively growing lesion was cut and transferred onto 5%V8A. Mycelia from the infected leaves were morphologically identified. 58

The 28th Special CU-af Seminar 2020 October 20, 2020 Growth rate and resistance to hymexazol A 6 mm diameter agar plug of each isolate was removed from the edge of a colony aged 3 d grown on 5%V8A at 25°C. The plug was put on 5%V8A to determine its growth rate or on 5%V8A supplemented with 0 mg/mL, 50 mg/mL, 100 mg/mL, 200 mg/mL, 300 mg/mL or 400 mg/mL hymexazol (3-hydroxy-5-methylisoxazole) to determine fungicide resistance. The plates were incubated at 25°C in the dark and the colony diameter was measured daily for 3 d. The growth rates on 5%V8A and on 5%V8A supplemented with hymexazol of each isolate were calculated from three replicates. Statistical analysis Data were analyzed using one-way analysis of variance facilitated by IBM SPSS 22.0 software (IBM; USA). Means were compared using Duncan’s multiple range test with multiple comparison corrections and critical value at p < 0.05. Results and Discussion Characterization of the dsRNA To determine type of nucleic acid in isolate R84, nucleic acid purified by cellulose column chromatography was treated with DNase I (data not shown), RNase A, and S1 nuclease. Figure 1 showed that the nucleic acid resisted S1 nuclease and RNase A in high salt buffer digestion which confirmed dsRNA nature of the viral genome. In figure 1A, dsRNA band of the viral genome of 2,320 base pairs (bp) in length was disappeared and dsRNA band of the viral genome of 3,750 bp in length was very faint. This could be the reason that dsRNA reactions were precipitated to remove salt before separating on agarose gel. Figure 1 RNase A and S1 nuclease treatment of cellulose column chromatography purified nucleic acid of isolate R84, where M = 1Kb DNA Ladder RTU (GeneDireX; Taiwan): (A) nucleic acid treated with RNase A, where Lane 1 = no treatment control; Lane 2 = RNase A treatment under low salt condition; Lane 3 = treatment under high salt condition; (B) nucleic acid treated with S1 nuclease, where Lane 1 = total RNA of Phytopthora botryosa; Lane 2 = total RNA of P. botryosa with S1 nuclease treatment; Lane 3 = cellulose column chromatography purified nucleic acid of isolate R84 without S1 nuclease; Lane 4 = cellulose column chromatography purified nucleic acid of isolate R84 treated with S1 nuclease. 59

The 28th Special CU-af Seminar 2020 October 20, 2020 Genomic sequencing of dsRNA Double-stranded RNA isolated from the cellulose column was treated with S1 nuclease to remove ssRNA from host’s mRNA before cDNA was synthesized. First-stranded synthesis was carried out using random hexamer primers. From sequencing results of cDNA clones, none of the clone contained the RdRp gene. There was a clone named 26 containing a fragment of 233 nt, which could be predicted to encode amino acid (aa) sequence of 77 aa in direction of 3’ to 5’ of Frame 2. Nucleotide sequence of the clone 26 is shown in figure 2. Figure 2 Nucleotide sequence of clone 26 The cDNA sequence from clone 26 showed no significant similarity with any nucleotide sequences in GenBank. Amino acid sequence was predicted and queried to protein databases using BLASTP (NCBI). Amino acid sequences from database showing highly similar to cDNA of clone 26 and statistical significance are listed in Table 2. Table 2 Amino acid sequences producing significant alignments to cDNA of clone 26 Surprisingly, the amino acid sequences from database showing significant similarity to cDNA of clone 26 were all mannosyltransferase protein from plants. Horizontal gene transfer among viruses, fungi and plants are common (Lui et al., 2010 doi:10.1128/JVI.00955-10). The substrate-specific mannosyltransferase protein family has not been reported in virus but there were glycosyltransferases reported in Polyprotein of PEV1 encoding putative UDP glycosyltransferases (UGTs) which showed significant alignments with bacterial and fungi UDP-glucose: sterol glucosyltransferases and the highest similarity score belonged to the protein of the fungus, Ustilago maydis. Protein analogue sequence analysis of other species in endornavirues revealed two viruses in the genus, bacteriopharge and plant virus, also contained UGT motifs (21). The concept of ‘glycovirology’ was reviewed by Markine-Goriaynoff and colleagues (39), one mechanism that viruses use to modify sugar molecules is by expressing their own glycosyltransferase for many advantages. More work needed to be done to determine whether the cDNA from clone 26 is a part of the viral genome, a southern blot of purified dsRNA using 60

The 28th Special CU-af Seminar 2020 October 20, 2020 clone 26 as a probe will be performed. To increase the chance of obtaining the viral RdRp degenerated primers and primers from conserved motif of viruses from pythiaceous fungi were used for cDNA synthesis. A fragment of approximately 600 bp in length was obtained from primers designed from RdRp region of PpRV1 as shown in figure 3. The fragment will be cloned and sequenced. Figure 3 Six hundred base pairs of cDNA synthesized from 8 kb fragment of dsRNA in isolate R84 using primers designed from RdRp region of PpRV1. M=1Kb DNA Ladder (Goldbio; USA). Lane 1 and 2 are from separate RT-PCR reactions. Molecular characterization of isolate R84 Various Phytophthora spp. reported as para rubber tree pathogen usually sporulate on induction media (31). Since the isolate R84 did not sporulate on induction media, molecular identification was performed. DNA of isolate R84 was amplified using conventional PCR with the fungal and oomycete universal primers, ITS1 and ITS4, with the binding site located in the conserved non-coding region between 18S, 5.8S and 28S of ribosomal RNA genes (36). Surprisingly, the sequencing result showed that the fragment had high similarity to rRNA gene of Py. cucurbitacearum isolate 1241Pc (HQ237483.1; identity, 902/904 (99%); query cover, 100%; e-value, 0.0). The partial ITS sequence of strain R84 was deposited at NCBI with Accession No. MH243441. Evolutionary analyses of isolate R84 with Phytopythium (formerly Pythium clade K), Pythium clade A–J and Phytophthora were performed based on the nucleotide sequence of ITS1/ITS4 products. The phylogram in figure 4 showed that isolate R84 was closely related to Py. cucurbitacearum isolate CBS748.96 (AY598667). AY598667 was used in the phylogram because HQ237483.1 was not accredited (40). Unlike Phytophthora, Phytopythium rarely forms spores on solid media (13). Sporulation was induced in sterile natural pond water. Sporangia produced by isolate R84 were varied in shape and papillation. Most zoosporangia were globose without a papillum (Fig. 5A). Other shapes were infrequently found such as ovoid and pyriform, with or without a papillum (Fig. 5B, 5C and 5E–5G). Zoospore discharge via vesicle formation could be seen (Fig. 5E and 5F). Isolate R84 was noted as a homothallic species because sexual structures were found in the pure culture, showing a smooth-walled oogonium with a paragynous antheridium (Fig. 5D). 61

The 28th Special CU-af Seminar 2020 October 20, 2020 Figure 4 Phylogram based on the internal transcribed spacer rDNA region of isolate R84, Pythium clade A-J, Phytopythium spp., Phytophthora spp. and Aspergillus nidulans, as an out group Figure 5 Asexual structures and sexual structures of isolate R84 induced using baiting technique after 4 d incubation at 25°C, showing sporangia of isolate R84: (A) nonpapillate globose; (B) nonpapillate ovoid; (C and G) papillate pyriform; (E) discharged globose intercalary; (F) discharged ovoid terminal; (H) formed on simple sympodial sporangiophore; (D) sexual structures of isolate R84, with oogonia (arrowheads) with paragynous antheridia (arrows) and mature plerotic oogonium with a thick-walled oospore (*) and scale bar = 50 µm 62

The 28th Special CU-af Seminar 2020 October 20, 2020 The isolate R84 has more than one virus R84 has three dsRNA segments. However, no virion particles were observed by transmission electron microscopic observation (results not shown). As a result, dsRNA in isolate R84 could be the true genome or the replicative form and could be composed of more than one type of virus. Single zoospore isolation as a method for virus curing was performed (20). A few isolates from single zoospore had only the largest dsRNA fragment as shown in figure 6. These results suggested that the largest dsRNA is the whole genome of a virus in the isolate R84. Most of the RdRp gene of mycoviruses is approximately 2kb in length (20, 29). The two smaller dsRNA could be a genome of at least one more virus. Figure 6 S1 nuclease treatment of the large dsRNA fragment from a single zoospore of the strain R84. M=1Kb DNA Ladder (Goldbio; USA) Koch’s postulates of isolate R84 on detached para rubber leaf Many isolates with morphological characteristics similar to isolate R84 were consistently isolated along with other Phytophthora spp. from necrotic petioles. Strain L69 isolated from Rayong province, Thailand (12°54’55.0”N 101°28’52.4”E) was molecularly identified as Phytopythium cucurbitacearum using ITS region sequencing and was devoid of dsRNA elements (results not shown). Suspension of 1 × 104 zoospores was added to surface-sterilized para rubber leaf and incubated for 7 d. On young leaves, lesions were observed 36 hr after inoculation, while control leaves remained symptomless (Fig. 7). On mature leaves, lesions could be seen on day 4 from both isolates (results not shown). The pathogens were re-isolated from infected leaves and morphologically identified, thus fulfilling Koch’s postulates. Fig u re 7 Yo u n g p a r a rubber leaves infected with zoospores of isolate R84 and isolate L69 over time 63

The 28th Special CU-af Seminar 2020 October 20, 2020 Growth rate and hymexazol resistance of Phy. cucurbitacearum Growth on 5%V8A and hymexazol resistance were investigated to determine whether the presence of the virus altered the physiology of isolate R84. Since attempts to generate a virus-free strain from isolate R84 were not successful, six isolates of Phy. cucurbitacearum from different para rubber tree plantations were used. The results showed that the growth rate and hymexazol sensitivity of isolate R84 were not significantly different from other Phy. cucurbitacearum isolates (Fig. 8). The fact that isolate R84 and six other virus-free isolates were indistinguishable in colony morphology, growth rate, ability to produce reproductive structures and hymexazol resistance of Phy. cucurbitacearum suggested that the virus of isolate R84 is cryptic. Figure 8 Growth rates of different Phy. cucurbitacearum strains isolated from para rubber leaves with ALF on 5%V8A supplemented with different concentration of hymexazol at 25°C. Statistical significance was determined by one-way analysis of variance with multiple comparison correction by Duncan’s Multiple Range test. Different letters indicate significance groups (p < 0.05). (error bar = ±SD) Conclusion This is the first report of virus-like dsRNA in Phy. cucurbitacearum causing disease in the para rubber tree. The isolate has at least 2 viruses, one with a monopartite genome of approximately 8 kb in length; however, the two smaller fragments of approximately 3.7 and 2.3 kb in length could not be isolated. The viruses caused an asymptomatic infection in its host. Phy. cucurbitacearum has never been reported in Thailand. It was found associated with durian tree-decline in Indonesia (41). In eastern Thailand, where para rubber tree plantations intermingle with durian farms, the presence of Phy. cucurbitacearum in the three studied provinces of eastern Thailand could negatively affect the durian industry in the region. 64

The 28th Special CU-af Seminar 2020 October 20, 2020 References 1) Workman, D., http://www.worldstopexports.com/natural-rubber-exports-country/ 2018 2) Priyadarshan, P.M., In Advances in Agronomy, 2003: 351-400: Academic Press. 3) Yu, H., Hammond, J., Ling, S., Zhou, S., Mortimer, P.E. and Xu J.,Ind. Crops. Prod. 2014(62) : 14-21. 4) Pethin, D., Nakkanong, K. and Nualsri, C., Scientia Agricola. 2015(72) : 306-13. 5) Puwaphut, R., Nakkaew, A. and Phongdara, A., S. J. S. T. 2016(38) : 15-22. 6) Havanapan, P-o., Bourchookarn, A., Ketterman, A.J. and Krittanai, C., J. Proteomics. 2016(131): 82-92. 7) de Cock, A.W., Lodhi, A.M., Rintoul, T.L., Bala, K., Robideau, G.P., et al. Persoonia, 2015(34) : 25-39. 8) van den Berg, A.H., McLaggan, D., Diéguez-Uribeondo, J. and van West, P., Fungal. Biol. Rev. 2013(27): 33-42. 9) Baten, M.A., Asano, T., Motohashi, K., Ishiguro, Y., Rahman, M.Z.,et al., Mycol. Prog. 2014(13): 1145-1156. 10) Kamoun, S., Furzer, O., Jones, J.D.G., Judelson, H.S., Ali, G.S., et al., Mol. Plant Pathol. 2015(16): 413-434. 11) Jayasuriya, K.E., Whesunderaz, R.L.C. and Deraniyagala, S.A., Ann.Appl. Biol. 2003(142) : 63-69. 12) Laohasakul, B., Boonyapipat, P. and Plodpai, P., Plant Disease, 2017(101): 1057-1057. 13) Zeng, H.C., Ho, H.H. and Zheng, F.C., Mycopathologia, 2005(159):601-606. 14) Johnston, A., In Rubber, ed. Webster, C.C. and Baulkwil, W.J.I., Longman Scientific and Technical, New York, USA. 1989:415-458. 15) Erwin, D.C. and Ribeiro, O.K., Phytophthora Diseases Worldwide,St Paul, Minnesota : The American Phytopathological Society. 1996 16) Chee, K., Planters’ Bulletin of the Rubber Research Institute of Malaysia, 1969(104): 190-8. 17) Nuss, D.L., Nat. Rev. Microbiol. 2005(3): 632-642. 18) Pearson, M.N., Beever, R.E., Boine, B. and Arthur, K., Mol. Plant Pathol. 2009(10): 115-28. 19) Dawe, A.L. and Nuss, D.L., Annu. Rev. Genet. 2001(35): 1-29. 20) Cai. G. and Hillman, B.I., Adv. Virus Res. 2013(86): 327-350. 21) Hacker, C.V., Brasier, C.M. and Buck, K.W., J. Gen. Virol. 2005(86): 1561-1570. 22) Kozlakidis, Z., Brown, N.A., Jamal, A., Phoon, X. and Coutts, R.H., Virus. Genes. 2010(40) : 130-134. 23) Cai, G., Myers, K., Hillman, B.I. and Fry, W.E., Virology, 2009(392): 52-61. 24) Dantzer, R., Curr. Top Behav. Neurosci. 2017(31): 117-138. 25) Moffitt, E.M. and Lister, R.M., Phytopathology, 1975(68): 851-859. 26) Klassen, G.R., Kim, W.K., Barr, D.J.S. and Désaulniers, N.L., Mycologia, 1991(83): 657-661. 27) Gillings, M.R., Tesoriero, L.A. and Gunn, L.V., Plant Pathol. 1993(42): 6-15. 28) Shiba, K., Hatta, C., Sasai, S., Tojo, M., Ohki S.T. and Mochizuki, T., Arch. Virol. 2018(163):2561-2563. 29) Sasai. S., Tamura, K., Tojo, M., Herrero, M.L., Hoshino, T., et al.Virology, 2018(522): 234-243. 30) Jeffers, S.N. and Martin, S.B., Plant Disease, 1986(70): 1038-1043. 31) Sopee, J, Sangchote, S. and Stevenson, W.R., Phytoparasitica,2012(40): 269-278. 32) Das, S., Falloon, R.E., Stewart, A. and Pitman, A.R., Fungal. Biol. 2014(118): 924-34. 33) D’Alessio, J.M. and Gerard, G.F., Nucleic Acids Res.1988(16): 1999-2014. 65

The 28th Special CU-af Seminar 2020 October 20, 2020 34) Sievers, F., Wilm, A., Dineen, D., Gibson, T.J., Karplus, K., et al. Mol. Syst. Biol. 2011(7) : 539-539. 35) Zhan, F., Zhu, W. and Zhan, J., Plant Disease, 2016(100): 1253-1253. 36) White, T., Bruns, T., Lee, S. and Taylor, J., In PCR Protocols : A Guide to Methods and Applications, ed. Innis, M., Gelfand, D. Shinsky, J. and White, T., Academic Press, (1990)315-22. 37) Tamura, K., Stecher, G., Peterson, D., Filipski, A. and Kumar, S., Mol. Biol. Evol. 2013(30) : 2725-2729. 38) Liu, H., Fu, Y., Xie, J., Cheng, J., Ghabrial S.A., et al. BMC Evol. Biol. 2012(12): 91-91. 39) Markine-Goriaynoff, N., Gillet, L., Van Etten, J.L., Korres, H., Verma, N. and Vanderplasschen, A., J. Gen. Virol. 2004(85): 2741-2754. 40) Levesque, C.A. and De Cock, A.W.A.M., Mycol. Res. 2004(108): 1363-1383. 41) Santoso, .PJ., Aryantha, I.N.P., Pancoro, A. and Suhandono, S., Asian J. Plant Pathol. 2015(9) : 59-71. 66

Removal of Polycyclic Aromatic Hydrocarbons (PAHs) in Waste Tire Pyrolysis Oil via Catalytic Hydrogenation Assoc. Prof. Dr. Napida Hinchiranan



The 28th Special CU-af Seminar 2020 October 20, 2020 Removal of Polycyclic Aromatic Hydrocarbons (PAHs) in Waste Tire Pyrolysis Oil via Catalytic Hydrogenation Assoc. Prof. Dr. Napida Hinchiranan Abstract Waste tire pyrolysis oil (WTPO) contains polycyclic aromatic hydrocarbon (PAHs) of 18,000 ppm, which are classified as toxic componds to induce cancer in human or animals. To produce the cleaner WTPO, hydrogenation was applied for removing PAHs in WTPO before using in a general combustion process. This research was consisted of 2 steps: hydrogenation of naphthalene, a model compond of PAHs, for catalyst screening and hydrogenation of real WTPO. For the first step, nickel supported on gamma alumina (Ni/γ-Al2O3) promoted by molybdenum (Mo), tungstate (W) or platinum (Pt) was selected as the hydrogenation catalysts. Although NiPt catalyst provided the highest naphthalene conversion to 55.8% under 30 bar initial H2 pressure at 350 ºC for 4 h, it were less tolerant to organosulfurs found in WTPO (1,400 ppm). At similar condition, WTPO hydrogenation over NiMo and NiW catalysts could achieve 68.8% and 54.2% PAHs reduction in the presence of sufur compounds. Keywords Naphthalene, Nickel, Promoter, Waste Tire Pyrolysis Oil, Polycyclic Aromatic Hydrocarbons, Hydrogenation. Department of Chemical Technology Faculty of Science, Chulalongkorn University Bangkok, Thailand 71

The 28th Special CU-af Seminar 2020 October 20, 2020 Introduction and Objectives Energy security is one of the strategies of every countries. Amid the depletion of traditional energy source (fossil fuel, coal, oil and natural gas) with environmental concerns, “Waste to Energy” concept can be applied to convert waste into heat and electricity as well as waste minimization in the community in the same time [1]. Approximately, 2.7 billion automotive tires were produced worldwide in 2017. Among this volume, 1 billion units were disposed [2]. It is estimated that the amount of waste tires will continuously increase due to the expansion of vehicle markets to support various activities according to the growth of economics and population. Thus, disposal of these waste tires via landfill or burn is inappropriate since they are classified as non-biodegradable materials and they can generate pollutants such as sulfur dioxide (SO2) and polycyclic aromatic hydrocarbons (PAHs) during their combustion. Although the recycle usefully convert the waste tires as other products, it cannot consume all of the waste tires generated. The way to significantly decrease the volume of waste tires is to convert as tire-derived fuels due to their high gross calorific value of 41 – 44 MJ/kg [3]. According to the higher energy density of liquid fuels, the pyrolysis conducted under inert atmosphere at high temperatures (300 - 900 °C) can transform waste tire as the liquid fuels called as the waste tire pyrolysis oil (WTPO) [4]. However, the direct utilization of WTPO in general combustion engines is limited due to the large amount of organosulfurous compounds produced from the vulcanizing agents [5] and high concentration of certain polycyclic aromatic hydrocarbons (PAHs) generated via Diels-Alder aromatization of alkenes and dienes during pyrolysis [6]. To eliminate the organosulfurous compounds, the hydrodesulfurization catalyzed by nickel-molybdenum (NiMo) and cobalt-molybdenum (CoMo) supported on gamma-alumina (γ-Al2O3) potentially removed the organosulfurous compounds found in the WTPO from 1.15 wt% (11,500 ppm) to ca. 0.14 wt% (87.8 wt% sulfur removal) [5]. Nevertheless, much less attention has been studied to eliminate the PAHs compounds in the WTPO. PAHs are compounds of two or more benzene rings such as naphthalene, phenanthrene, anthracene and pyrene etc. [7]. Williams and Taylor [8] reported that the WTPO contained a huge PAHs content as 1.4 – 10.1 wt% depended on the pyrolysis temperature. Among the PAHs found in the WTPO, some species are active in the human and bacterial cell test including the mutagenicity bioassays such as phenanthrene, benzo [a] pyrene, benzo [e] pyrene or chrysene [7, 8]. This problem limits the replacement of WTPO for conventional petroleum-based fuels. Moreover, it has been reported that the diesel having high PAHs content has very low cetane number [9]. Thus, it is necessary to remove the PAHs in order to improve the quality of WTPO before applying in the engines. There are several techniques to eliminate the PAHs such as thermal treatment, photodegradation, chemical oxidation and cracking [10]. However, these techniques have slow decomposition rate and high energy consumption. Moreover, some intermediates generated from chemical oxidation have higher toxicity than inherent PAHs due to their higher solubility in water resulting in the high potential to be absorbed in the body of human or animals [10]. Hydrogenation is recommended method since it has high efficiency to convert PAHs as the less toxic compounds such as tetralin, which has high LD50 level with fast biodegradable rate than naphthalene [10]. Nickel (Ni) based-catalysts are generally used for hydrotreating process due to its lower cost which is more practical for mass-scale operation than the noble metal-based ones. However, they can be poisoned by sulfur compounds to decrease its activity [11]. Thus, the Ni catalysts require some promoters with high resistance to sulfurous compounds for hydrogenation of WTPO. The previous literatures showed that the cooperation with molybdenum (Mo) and tungsten (W) induced Ni having higher 72

The 28th Special CU-af Seminar 2020 October 20, 2020 resistance to sulfur-poisoning and the obtained bimetallic catalysts also have high efficiency in hydrodesulfurization (HDS) to remove sulfur compounds during oil refinery [5, 11]. Moreover, the hydrogenation of naphthalene using catalysts consisting of Ni and platinum (Pt) supported on ZSM-5 showed 96-99% naphthalene conversion with high sulfur tolerance in the feed stream [12]. Thus, this research aims to study and compare the efficiency of Ni supported on gamma alumina (Ni/γ-Al2O3) catalysts with and without the addition of Mo-, W- or Pt-promoter in the hydrogenation of naphthalene used as a model of PAHs compound and hydrogenation of real WTPO.In the case of naphthalene hydrogenation, the appropriate promoters for Ni catalyst was selected from the catalyst ability to achieve the high degree of naphthalene conversion with high tolerance to thiophene, which was mostly found in WTPO. For the real WTPO, these appropriate catalysts were used for the removal of PAHs including the elimination of sulfurous compounds in WTPO were simultaneously investigated. Materials and Methods Preparation of WTPO and quantification of PAHs in WTPO The waste tire powder (50 g/batch) was pyrolyzed at 500 ºC for 30 min under nitrogen atmosphere (50 mL/min) in the fixed bed reactor. The pyrolytic vapor was then trapped in the container emerged in an ice bath. The yield of the obtained liquid product or WTPO was ca. 33.2 wt% and it was consisted of the 1.50 wt% sulfur compounds. Then, WTPO was fractionated using liquid chromatography to quantify the level of PAHs in WTPO following the direction described in elsewhere [8]. Briefly, the slurry of silica gel saturated with hexane was carefully poured into the glass column ( = 1 in; L = 20 in) containing a cotton at the column bottom. Then, 1 g WTPO was dropped into the column. The 100 mL hexane was slowly transferred into the column to elute the aliphatic fraction. The rest WTPO in the column was then eluted by 200 mL benzene to separate the PAHs fraction. The PAHs species in the benzene fraction was analyzed by gas chromatography-mass spectroscopy (GC-MS). Preparation and characterization of γ-Al2O3-support and Ni-based catalysts γ-Al2O3 was prepared via hydrothermal method as described in elsewhere [13]. Al(NO3)3 aqueous solution (0.215 M) was mixed with cetyltrimethyl ammonium bromide (CTAB) aqueous solution (0.093 M) with the constant Al2O3/CTAB molar ratios of 2:1. Under vigorous stirring at room temperature, 12.5 wt% ammonia solution was slowly dropped into the above mixture until the pH of the obtained mixture reached to 9.0. The mixture was then transferred into a closed teflon bottle and aged at 90ºC for 60 h. The obtained precipitate collected after centrifugation (rotational speed = 10,000 rpm for 5 min) was washed by distilled water and absolute ethanol. Then, it was dried at 110ºC overnight and finally calcined at 550ºC for 6 h in air. The catalysts used in this research were prepared via incipient wetness impregnation method. The mixtures of nickel(II) nitrate hexahydrate (Ni(NO3)2·6H2O) and ammonium heptamolybdate tetrahydrate ((NH4)6Mo7O24·4H2O), ammonium paratungstate (NH4)10H2(W2O7)6 or tetraammineplatinum (II) nitrate (Pt(NH3)4(NO3)2) in the presence of citric acid (molar ratio of Ni/ citric acid = 1) were impregnated onto the γ-Al2O3 support to obtain NiMo, NiW and NiPt catalysts, respectively. Then, the resulting catalysts were dried at 110 °C for 2 h and calcined at 550 °C for 4 h in air. The total content of metals (Ni and promoter) was kept constant at 20 wt% based on the amount of support (promoter atomic ratio = 0.27). These catalysts were characterized using N2 adsorption- desorption, X-ray diffractometry (XRD), H2-temperature program reduction (H2-TPR) and ammonia- temperature program desorption (NH3-TPD) to detect the textural properties, NiO crystallite 73

The 28th Special CU-af Seminar 2020 October 20, 2020 size, reduction temperature and acidity of these catalysts, respectively. Hydrogenation of naphthalene or PAHs in WTPO The hydrogenation of naphthalene and PAHs in WTPO was performed in 250 mL stainless-steel high pressure reactor. The catalyst (10 wt% based on the reactant) was loaded into the reactor and typically in situ reduced under 30 bar initial hydrogen (H2) pressure at 400 ºC for 2 h. When the temperature and pressure decreased to the atmospheric condition, 100 mL naphthalene or WTPO solution in n-decane (10 wt%) was charged into the reactor using a syringe. The reactor was then pressurized by H2 and heated to the desired reaction temperature under stirring rate of 550 rpm for 4 h. The effects of various parameters on the reduction of naphthalene or PAHs in WTPO were quantified using gas chromatography equipped with flame ionization detector (GC-FID) and gas chromatography–mass spectrometry (GC-MS). The test of sulfur tolerance for all studied catalysts was also examined by using thiophene (1,400 and 10,000 ppm). Results and Discussion Catalyst characterization The textural properties of all calcined catalysts were summarized in Table 1. For the unpromoted Ni catalyst, it had the surface area (SA), pore volume (PV) and pore diameter (PD) as 157 m²/g, 0.39 cm³/g and 8.24 nm, respectively. The addition of Mo- or W-promoter significantly decreased the surface area and total pore volume to 99 m²/g and 0.22 cm³/g for NiMo catalyst and to 145 m²/g, and 0.24 cm³/g for NiW catalyst. In the case of Mo-promoter, it was possible that the Mo species migrated and deposited in the mesopores of the catalyst support [14]. Whereas, the WO3 was possibly blocked inside the pore of catalyst resulting in the reduction of catalyst surface area and pore volume [15]. For the NiPt catalyst, the results in Table 1 showed that the addition of Pt-promoter did not affect the surface area and pore volume of the catalysts. However, it was noticed that the pore diameter of the NiPt catalyst was lowest when it was compared to that of NiMo and NiW catalysts at the similar promoter atomic ratio. This was due to the strong interaction between Ni and Pt atom to decrease the size of NiO crystallite phase [16] and they could be deposited inside the pore of catalyst resulting in the reduction of pore volume and pore diameter. Table 1 Textural properties and NiO crystallite size of all calcined catalysts   The effect of the Mo-, W- and Pt-promoter on the crystallite size of NiO phase in the Ni-based catalysts was also detected by XRD analysis and calculated by using Scherrer’s equation [17] as shown in Fig. 1 and summarized in Table 1. For unpromoted Ni catalyst (Fig. 1a), the signals of NiO phase were found at 2θ of 43.4° 74

The 28th Special CU-af Seminar 2020 October 20, 2020 Fig. 1. XRD patterns of calcined Ni-based catalysts with and without the addition of Mo-, W- or Pt-promoter. and 62.5° [16, 17], while the signals at 2θ of 37.5°, 45.7° and 67.2° were attributed to γ-Al2O3 phase [18]. The average crystallite diameter of NiO in the unpromoted catalyst was 27.7 nm. When the Mo-promoter was loaded onto the Ni catalyst, the diffraction peaks with a high intensity for MoO3 were observed at a 2θ of 12.7°, 25.8°, 27.3° and 33.9° [19], while the diffraction peak of nickel molybdenum oxide (NiMoO4) was present at a 2θ of 14.6° [19]. The sharp peaks detected at 2θ of 23.5° and 29.7° were ascribed to the phase of ammonium nickel molybdate ((NH4)HNi2(OH)2(MoO4)2) [20]. Table 1 shows that the addition of Mo-promoter could decrease the NiO crystallite size in the NiMo catalyst to 17.5 nm reflecting the higher degree of NiO dispersion [21]. For the NiW catalysts, the results showed the diffraction peaks for WO3 phases at a 2θ of 23.7°, 28.9°, 32.3° [22] and the diffraction peaks of nickel tungsten oxide (NiWO4) were exhibited at a 2θ of 31.5° and 41.8° [15] and 54.5° [PDF 01-072-1189]. The average crystallite size of NiO in the NiW catalyst decreased to 16.8 nm. It implies that the addition of appropriate W content could provide the excellent dispersion of both NiO and WO3 [15]. In the case of Pt-promoter, the diffraction peaks of PtO3 phase in the NiPt catalyst were not detected, while the crystallite size of NiO significantly reduced to 9.6 nm indicating the highest dispersion of NiO on the surface of the γ-Al2O3 support [23, 24] Fig. 2 shows the H2-TPR profiles of the Ni-based catalysts with and without the addition of promoters. The unpromoted Ni catalyst had a broad reduction peak in the range of 460-850 °C. Below 650 °C, it was the reduction zone of NiO, while the reduction of the nickel aluminate phase was found at above 700 °C [19]. For the H2-TPR profile of the NiMo catalyst, three major peaks of reduction were observed at 490, 560 and 770 °C attributed to the reduction of Mo6+ to Mo4+, NiO and NiMoO4 on the alumina surface, and nickel aluminate (NiAl2O4) spinal structure, respectively [25, 26]. The addition of Mo-promoter significantly shifted the NiO reduction peak from 650 °C to 560°C. This indicated that the addition of Mo-promoter enhanced the metal dispersion and also provided the lower interaction between NiO particles and the alumina surface [21]. To consider the NiW catalysts, Fig.2 showed the three zones of the reduction at 560 °C, 650 °C and 850 °C, which were regarded as the reduction temperature of NiO, WO3 and NiAl2O4, respectively [27]. It was observed that the addition the W-promoter in the catalysts increased 75

The 28th Special CU-af Seminar 2020 October 20, 2020 the reduction temperature of NiO. It has been reported that Ni-based catalysts with W-promoter shift toward the higher reduction temperature of NiO as compared to unpromoted Ni-based catalysts Fig. 2. H2-TPR profiles of the Ni-based catalysts with and without the addition of Mo-, W- or Pt-promoters. [15]. This was possible that the addition W-promoter increased the metal dispersion via enhancement of metal-support interaction. In the case of NiPt catalysts, three major peaks were also observed. The lowest reduction temperature at 220°C was assigned to the reduction of PtO2 [17], while the reduction temperature at 470 °C and 790 °C were NiO and NiAl2O4 phase, respectively. The results showed that the addition of the Pt-promoter decreased the reduction temperature of NiO phase since he hydrogen of Pt was transferred to Ni by the hydrogen spillover effect [16]. The number and strength of the acid sites in each catalyst were evaluated by NH3-TPD analysis as shown in Table 2. Generally, the acid sites are classified as weak (< 200 °C) medium (200–350 °C) and strong (350–600 °C) acid sites depending on the desorption temperature of NH3 [28]. The unpromoted Ni catalyst had the weak, medium and strong acid sites as 119, 196 and 540 µmol NH3/g, respectively with the total acidity as 855 µmol NH3/g. The addition of Mo-promoter increased the total acidity to 885 µmol NH3/g. This was due to the increase in the Lewis acid sites by Mo4+. For the addition of W-promoter, the number of total acidity of NiW catalyst significantly decreased to 549 µmol NH3/g. This reflected that the well dispersed W covered the acid sites of support resulting in the reduction of the acidity [15]. However, the addition Pt-promoter did not affect to total acidity of the NiPt catalysts. This result was also observed by Ning et al. [29]. Table 2 Surface acidity of all Ni-based catalysts with different types and contents of promoters 76

The 28th Special CU-af Seminar 2020 October 20, 2020 Hydrogenation of naphthalene It has been widely reported that the naphthalene hydrogenation is a typical consecutive reaction as shown in Scheme 1. The mechanism for hydrogenation of naphthalene can be divided as two steps. The first step is the hydrogenation of naphthalene to from tetralin. Then, tetralin was converted to decalin [20]. Scheme 1. Reaction routes of naphthalene hydrogenation [20]. To investigate the effect of the promoter types on the product distribution, the reaction was performed under the central condition under 30 bar initial H2 pressure at 350 oC for 4 h. At this condition, Fig. 3 shows that the system using Ni and NiMo catalysts yielded the liquid product in the range of 90–95 wt%, while the reaction catalyzed by NiW and NiPt catalysts provided the lower content of liquid product (65–85 wt%). This was possible that the W- and Pt-promoter promoted the catalytic cracking to yield the lower content of liquid product with higher gaseous product formation [30, 31]. Fig. 3 Effect of the Ni-based catalysts with and without the addition of Mo-, W- or Pt-promoters on product distribution. 77

The 28th Special CU-af Seminar 2020 October 20, 2020 In the absence of the sulfurous compounds, the results in Table 3 show that the unpromoted Ni catalysts exhibited the lowest efficiency to convert naphthalene to achieve only 29.5% conversion with only 21.1 wt% tetralin and 3.8 wt% decalin yields. Whereas, the addition of Mo-promoter to Ni-catalyst provided the higher hydrogenation efficiency to obtain 44.7% naphthalene conversion with 40.2 wt% tetralin and 4.5 wt% decalin. This was due to the effect of higher acidity that could promote hydrogenation [32, 33]. For the NiW catalyst, this catalyst also increased the naphthalene conversion to 38.8%. However, its catalytic efficiency was lower than that of NiMo catalyst at the same promoter atomic ratio. This was possibly due to the lower acidity resulted in the poorer activity for catalyzing hydrogenation. Moreover, the reducibility of NiW catalyst was lower than that of NiMo catalyst as seen in Fig. 2. In the case of the NiPt catalyst, the results in Table 3 indicated that this catalyst gave the highest hydrogenation efficiency to provide the highest level of naphthalene conversion at 55.8% with the yields of tetralin and decalin as 50.4 and 5.4 wt%, respectively. This was resulted from the smallest crystallite size of NiO reflecting the highest degree of metal dispersion. The high catalyst surface area with high total acidity of NiPt catalyst also promoted hydrogenation activity. In addition, the Pt had high activity and selectivity for aromatic ring hydrogenation [34]. Table 3 The conversion and product yields of the Ni-based catalysts with and without the addition of Mo-, W- or Pt-promoters According to the high content of organosulfurous compounds found in WTPO (14,800 ppm), the sulfur tolerance of each catalyst was required to be examined. In the system of hydrogenation of naphthalene model compound, 1,400 and 10,000 ppm of thiophene were added into thenaphthalenefeedstock(Table4).At1400ppmthiopheneconcentration, it was found that the unpromoted Ni catalyst was deactivated by thiophene resulted in the poor efficiency for naphthalene hydrogenation. Moreover, the activity of unpromoted Ni catalyst was totally inhibited when the thiophene content was increased to 10,000 ppm. This indicated that Ni catalyst had low resistance to sulfurous compounds [11]. To consider the effect of Mo, W and Pt-promoters for sulfur tolerance of the catalysts, the addition of Mo enhanced the level of sulfur tolerance of Ni-based catalyst and it could remain the catalytic efficiency to achieve 48.9 and 45.7% naphthalene conversion in the presence of thiophene at 1,400 and 10,000 ppm, respectively. These results were similar to the system using NiW catalyst that the level of naphthalene hydrogenation was still remained at ca. 38.0% for both thiophene concentration at 1,400 and 10,000 ppm. In the case of Pt promoter, the efficiency of NiPt catalyst for naphthalene hydrogenation was significantly decreased when the reaction system contained thiophene. This means that the NiPt catalysts had low sulfur tolerance. From the above results, it could be implied that the unpromoted Ni catalyst and the NiPt catalyst were not appropriate for applying as the hydrogenation catalyst for PAH removal in WTPO due to the sulfur poisoning effect [35]. 78

The 28th Special CU-af Seminar 2020 October 20, 2020 Thus, NiMo and NiPt catalysts were appropriate and selected to apply for further hydrogenation of real WTPO. Table 4 Conversion and product yields of Ni/γ-Al2O3 with or without promoter in the absence and presence of thiophene Hydrogenation of WTPO After hydrogenation catalyzed by NiMo and NiW catalysts, the color of WTPO was changed from black to light brown as shown in Fig. 4. It could be explained that some color induced substances such as unsaturated carbonyl groups, polycyclic aromatic hydrocarbons, quinones, heterocyclic and nitrous compounds were eliminated during PAHs removal by hydrogenation. Table 5 shows the product distribution obtained from the hydrogenation of WTPO using both catalysts under the central condition of 30 bar initial H2 pressure at 350 oC for 4 h. It was found that NiMo catalyst provided the liquid product at 74.0 wt%, while the use of NiW catalyst gave the slightly lower content of liquid product at 70.5 wt%. Fig. 4 Waste tire pyrolysis oil before and after hydrogenation using NiMo and NiW catalysts. 79

The 28th Special CU-af Seminar 2020 October 20, 2020 Table 5 The product distribution of the hydrogenated Waste Tire Pyrolysis Oil Fig. 5 shows the GC-MS chromatograms indicating the existence of PAHs species in WTPO obtained fractionation by column chromatography before and after hydrogenation and the results were summarized in Table 6. Before hydrogenation, WTPO (Fig. 5a) contained 1.18 wt% PAHs concentration and the species of PAHs found in WTPO were such as naphthalene, 1-methylnaphthalene, 2-methylnaphthalene, biphenyl, 2-ethylnaphthalene, 1,5-dimethylnaphthalene, 1,3-dimethylnaphthalene, 1,4-dimethylnaphthalene, 4-methylbiphenyl, 1-isopropylnaphthalene, 1,2,8-trimethylnaphthalene and fluorene. It was noticed that naphthalene was the main fraction in the WTPO. After hydrogenation catalyzed by NiMo and NiW, the concentration of PAHs in WTPO was significantly decreased to 0.56 wt% and 0.83 wt%, respectively. Some peak intensity such as 1-isopropylnaphthalene was disappeared after hydrogenation over both catalysts. However, the catalytic activity of NiMo catalyst was superior than that of NiW catalyst. This was due to the effect of high acidity and ability to reduction at lower temperature. Fig. 5 GC-MS chromatograms of PAHs in waste tire pyrolysis oil obtained fractionation by column chromatography (a) before and after hydrogenation over (b) NiMo and (c) NiW catalysts. 80

The 28th Special CU-af Seminar 2020 October 20, 2020 Table 6 Species and content of PAHs compounds found in WTPO before and after hydrogenation obtained from fractionation by column chromatography Conclusions This research comparatively studied the catalytic efficiency of Ni/γ-Al2O3 catalysts promoted by Mo, W or Pt for elimination of naphthalene (PAHs model compounds) and PAHs in the real WTPO. Without the use of promoter, 20 wt% Ni/γ-Al2O3 showed only 29.5% naphthalene conversion with tetralin yield of 24.4%. The addition of promoters at 0.27 atomic ratio could enhance the naphthalene conversion to 44.7, 38.8 and 55.8% for NiMo, NiW and NiPt catalyst, respectively. However, the unpromoted Ni catalyst and NiPt catalyst had lower sulfur tolerance when the system contained 1,400 or 10,000 ppm thiophene. Thus, NiMo and NiW catalysts were selected for WTPO hydrogenation. In the case of WTPO, it contained 1.81 wt% PAHs concentration with 1.56 wt% naphthalene content. After hydrogenation over NiMo and NiW catalysts, it was observed that NiMo catalyst had greater catalytic efficiency to decrease the concentration of PAHs to 0.56 wt% (68.8% PAhs reduction), while the use of NiW catalyst provided 54.2% PAHs reduction. The higher catalytic efficiency of NiMo catalyst was related to the higher acidity and reducibility that gave the benefit for PAHs hydrogenation. Acknowledgment The authors gratefully acknowledge The Asahi Glass Foundation and Center of Excellence on Petrochemical and Materials Technology, Chulalongkorn University for the financial support. 81

The 28th Special CU-af Seminar 2020 October 20, 2020 References 1) Ramos, A., Monteiro, E., Silva, V. and Rouboa, A., Renewable and Sustainable Energy Reviews, 2018(81): 380-398. 2) Oboirien, B.O. and North, B.C., Journal of Environmental ChemicalEngineering, 2017(5) : 5169-5178. 3) Islam, M.N. and Nahian, M.R., Journal of Renewable Energy,2016(2016): 1-8. 4) Edwin Raj, R., Robert Kennedy, Z. and Pillai, B.C., Energy Conversion and Management, 2013(67): 145-151. 5) Jantaraksa, N., Prasassarakich, P., Reubroycharoen, P. and Hinchiranan, N., Energy Conversion and Management, 2015(95): 424-434. 6) Chen, S.-J., Su, H.-B., Chang, J.-E., Lee, W.-J., Huang, K.-L., Hsieh, L.-T., Huang, Y.-C., Lin, W.-Y. and Lin, C.-C., Atmospheric Environment, 2007(41): 1209-1220. 7) Abdel-Shafy, H.I. and Mansour, M.S.M., Egyptian Journal of Petroleum, 2016(25): 107-123. 8) Williams, P.T. and Taylor, D.T., Fuel, 1993(72): 1469-1474. 9) Eliche-Quesada, D., Mérida-Robles, J., Maireles-Torres, P., Rodrı́guez-Castellón, E., Busca, G., Finocchio, E. and Jiménez-López, A., Journal of Catalysis, 2003(220): 457-467. 10) Jacinto, M.J., Santos, O.H.C.F., Landers, R., Kiyohara, P.K. and Rossi, L.M., Applied Catalysis B: Environmental, 2009(90): 688-692. 11) Feng, J., Liu, J., Tang, M., Zhou, L., Zhang, X., Wen, X., Ge, H., Li, D. and Li, X., International Journal of Hydrogen Energy, 2017(42): 6727-6737. 12) He, T., Wang, Y., Miao, P., Li, J., Wu, J. and Fang, Y., Fuel, 2013(106): 365-371. 13) Zhang, X., Hu, R., Liu, Y., Zhao, J., Hou, L., Yang, D., Hao, Z. and Su,H., International Journal of Hydrogen Energy, 2015(40): 10026-10032. 14) Tan, P., Applied Catalysis A: General, 2019(580): 111-120. 15) Spojakina, A., Palcheva, R., Jiratova, K., Tyuliev, G. and Petrov, L., Catalysis Letters, 2005(104): 45-52. 16) Zhao, X., Wu, K., Liao, W., Wang, Y., Hou, X., Jin, M., Suo, Z. and Ge, H., Green Energy & Environment, 2019(4): 300-310. 17) El Doukkali, M., Iriondo, A., Miletic, N., Cambra, J.F. and Arias, P.L., International Journal of Hydrogen Energy, 2017(42): 23617-23630. 18) Fang, M., Ma, S., Wang, T., Xia, Z., Tang, W., Xia, L. and Luo, Z., RSC Advances, 2017(7) : 54512-54521. 19) Sangnikul, P., Phanpa, C., Xiao, R., Zhang, H., Reubroycharoen, P., Kuchonthara, P., Vitidsant, T., Pattiya, A. and Hinchiranan, N., Applied Catalysis A: General, 2019(574) : 151-160. 20) Liu, H., Liu, C., Yin, C., Liu, B., Li, X., Li, Y., Chai, Y. and Liu, Y., Catalysis Today, 2016(276): 46-54. 21) Fang, Z., Shi, D., Lin, N., Li, A., Wu, Q., Wang, Q., Zhao, Y., Feng,C., Jiao, Q. and Li, H., Applied Catalysis A: General, 2019(574): 60-70. 22) Kumar, V.V., Naresh, G., Deepa, S., Bhavani, P.G., Nagaraju, M., Sudhakar, M., Chary, K. V.R., Tardio, J., Bhargava, S.K. and Venugopal, A., Applied Catalysis A: General, 2017(531) : 169-176. 23) Mihet, M. and Lazar, M.D., Catalysis Today, 2018(306): 294-299. 24) He, L., Huang, Y., Wang, A., Liu, Y., Liu, X., Chen, X., Delgado, J.J., Wang, X. and Zhang, T., Journal of Catalysis, 2013(298): 1-9. 25) Khzouz, M., Wood, J., Pollet, B. and Bujalski, W., International Journal of Hydrogen Energy, 82

The 28th Special CU-af Seminar 2020 October 20, 2020 2013(38): 1664-1675. 26) Lee, J.-H., Lee, E.-G., Joo, O.-S. and Jung, K.-D., Applied Catalysis A: General, 2004(269) : 1-6. 27) Tong, R.-l., Wang, Y.-g., Zhang, X., Zhang, H.-y., Dai, J.-z., Lin, X.-c. and Xu, D.-p., Journal of Fuel Chemistry and Technology, 2015(43): 1461-1469. 28) Badoga, S., Sharma, R.V., Dalai, A.K. and Adjaye, J., Industrial & Engineering Chemistry Research, 2014(53): 18729-18739. 29) Ning, Q., Liu, S.-y., Zhang, H.-k., Chen, Z.-q. and Ren, J., Journal of Fuel Chemistry and Technology, 2018(46): 1454-1461. 30) Alsobaai, A.M., Zakaria, R. and Hameed, B.H., Chemical Engineering Journal, 2007(132) : 173-181. 31) Zhang, J., Chen, T., Jiao, Y., Wang, L., Wang, J., Chen, Y., Zhu, Q. and Li, X., Applied Surface Science, 2020(507): 145113. 32) Li, J., Xiang, L., Feng, X. and Wang, Z., Applied Surface Science, 2008(254): 2077-2080. 33) Hu, S., Xue, M., Chen, H. and Shen, J., Chemical Engineering Journal, 2010(162): 371-379. 34) Bjelić, A., Grilc, M., Huš, M. and Likozar, B., Chemical EngineeringJournal, 2019(359) : 305-320. 35) Betti, C., Badano, J., Maccarrone, M.J., Mazzieri, V., Vera, C. and Quiroga, M., Applied Catalysis A: General, 2012(435-436): 181-186. 83



Development of natural rubber composite for carbon dioxide adsorbent material Assoc. Prof. Dr. Sirilux Poompradub



The 28th Special CU-af Seminar 2020 October 20, 2020 Development of natural rubber composite for carbon dioxide adsorbent material Assoc. Prof. Dr. Sirilux Poompradub Abstract Herein this research, the significant challenge of CO2 capture and recent advantages of natural rubber (NR) to produce NR adsorbent materials was focused. To improve CO2 adsorption capacity of NR, the silica (SiO2) based adsorbents was introduced. NR latex was modified by a diallylamine using tert-butyl hydroperoxide (t-BHP) as an initiator. The SiO2 particles were also modified by using 3-aminopropyltriethoxysilane (1N-ES) 3-aminopropyltrimethoxysilane (1N- MS), N-[(3-trimethoxysilyl)propyl]ethylenediamine (2N-MS) or N-[(3-trimethoxysilyl)-propyl] diethylenetriamine (3N-MS) via a sol-gel reaction. The chemical structure, physical and thermal properties, and morphologies of modified NR (MNR) and modified SiO2 particles were investigated. Then, the MNR was mixed with modified SiO2 particles to develop the adsorbent material for CO2 adsorption. CO2 adsorption capacity increased from 0.79 mg/g (for NR) to 1.26 (for MNR) and 6.11 mg/g (for MNR with modified silica particles by 3N-MS) at the ambient temperature. By increasing temperature up to 100 oC, the CO2 adsorption capacity of MNR foam composite increased because the structure of MNR foam composite and the structure of modifier on modified silica surface become more flexible. Finally, the kinetic adsorption of adsorbent materials fitted with Avrami’s model indicated that MNR foam composite showed either physisorption or chemisorption. Department of Chemical Technology Faculty of Science, Chulalongkorn University Bangkok, Thailand 87

The 28th Special CU-af Seminar 2020 October 20, 2020 Introduction and Objectives The world’s energy consumption has been increased with quickly rate because of population growth and economic development [1]. Human activities, such as the burning of fuels, deforestation and hydrogen production from hydrocarbons have increased carbon dioxide (CO2) concentration in the atmosphere. CO2 is one of the major greenhouse gases responsible for global warming. Thus, the control of CO2 emission has become a serious and challenging research topic [2]. CO2 capture and storage (CCS) technology is required and contributed to reduce CO2 concentration in the atmosphere. Currently, CCS technologies are focused on reducing CO2 emission from industrial and other plants which were operated at high temperature. However, CO2 concentration in the atmospheric environment is also seemed to be high. Accordingly, the development of CO2 adsorption material at an ambient temperature was focused. Synthetic polymer foams such as poly (ethylene vinyl acetate) [3], poly(ethylene propylene diene) [4], acrylonitrile butadiene rubber [5] and polyurethane foam [6] have been garnering great interest due to their significant potential in industrial and commercial applications. However, the chemical toxicity and environmental safety should be concerned. Then, to make the eco-friendly environment, the natural polymer is an alternative to replace the synthetic polymer. Natural rubber (NR) is a natural commodity that has tremendous economic and strategic importance due to its unique characteristics such as high strength, flexibility, elasticity. NR has many advantages compared to synthetic polymer such as biodegradability and renewability [7]. In addition, NR foam can be used as a good absorbent material for chemical [8] and oil [9]. In our previous research [10], NR foam was developed as a CO2 adsorbent material. The CO2 adsorption capacity of NR foam was not so high, while by including modified zeolite with tetraethylenepentamine (TEPA), the CO2 adsorption capacity of NR composite material was significantly increased. However, the use of zeolite was limited for industrial section due to its high cost. In this study, the modified NR and modified SiO2 particles were prepared to improve the CO2 adsorption capacity of sorbent material. NR was modified by diallyamine and the SiO2 particles were modified by 3-aminopropyltriethoxysilane (1N-ES), 3-aminopropyltrimethoxysilane (1N-MS), N-[(3-trimethoxysilyl)propyl]ethylenediamine (2N-MS) or N-[(3-trimethoxysilyl) propyl]diethylenetriamine (3N-MS) via a sol-gel reaction. The modified NR was then characterized by Fourier transform infrared spectroscopy (FT-IR), thermogravimetric analysis (TGA) and scanning electron microscopy (SEM). The characterization of modified SiO2 particles was evaluated by FT-IR, elemental analysis, TGA, SEM and surface area and porosity analyzer. Then, the modified NR adsorbent material was filled with amino-organoalkoxysilane modified SiO2 to prepare the rubber composite foam. The CO2 adsorption capacity of modified NR adsorbent material was determined by a stainless steel reactor at an ambient temperature and pressure. The morphology of composite material was examined. Finally, the kinetic adsorption of modi- fied NR adsorbent material was evaluated by using pseudo-first-order, pseudo-second order and Avrami models, respectively. Materials and Methods Materials NR latex (60 wt% dry rubber content) was obtained from the Rubber Research Institute, Thailand. Tetraethyl orthosilicate (TEOS), diallylamine, t-butyl hydroperoxide, tetraethylene pentamine and modifiers (1N-ES, 1N-MS, 2N-MS and 3N-MS) were purchased from Sigma-Aldrich, USA. Ammonium hydroxide (28.0 % purity) was purchased from Qrec chemicals, 88

The 28th Special CU-af Seminar 2020 October 20, 2020 Thailand. Acetone (AR grade) was purchased from Mermaid, Thailand. N,N-dinitrosopentam- enthylenetetramine was supplied from Shaanxi Pioneer Biotech, China. Stearic acid, N-cyclo- hexyl-2-benzothiazole sulfernamide (CBS), sulfur and zinc oxide were purchased from the PAN Innovation, Thailand. All materials were used without further purification. Modification of NR latex by diallylamine 100 g of NR latex was diluted with 0.26 %w/w ammonium hydroxide solution (37 mL) in a closed glass reactor under a nitrogen atmosphere. 5 g of diallylamine in 110 mL of 0.26 %w/w ammonium hydroxide solution was added to the reactor and mixed for 30 min under N2 atmosphere at room temperature. 1 g of tert-butyl hydroperoxide (an initiator) was then added and mixed for 15 min at room temperature. Finally, tetraethylenepentamine as an activator at 0.5 wt% of rubber was added and the temperature was heated up to 50 °C. The reaction was continuously stirred at 50 °C for 24 h. Finally, the modified NR was coagulated with acetone and dried at 60 °C in vacuum condition. Figure 1 shows the reaction or NR modification by diallamine. Preparation of modified SiO2 particles The SiO2 particles were modified via a sol-gel reaction by mixing two solutions at ambient temperature. The first solution was added into the second solution under stirring. The first solution was the mixture of 3.0 mL TEOS and 23.0 mL EtOH, and the second solution was the mixture of 1.0 mL 28% ammonia solution, 7.0 mL deionized water and 18.0 mL EtOH. Then, the modifier (0.4 times of TEOS mole) was added into the solution. The reaction was continuously stirred at room temperature for 24 h. For unmodified SiO2 particles, the total reaction time was 24 h without adding the modifier. The SiO2 particles with/without modifier were separated by centrifugation. The obtained SiO2 particles were washed by deionized water and dried until the weight was constant. The modification of SiO2 particles via a sol-gel reaction was shown in Figure 2. Preparation of NR and modified NR foam composites The formulation of rubber compounding is listed in Table 1. All ingredients were mixed by a two-roll mill. The samples were then pressed by a compression mold under the temperature of 100 °C for 9 min. Finally, the rubber foam sheet was further heated in an oven at temperature of 140 °C for 9 min. Figure 1 Reaction of NR modification by diallylamine. 89

The 28th Special CU-af Seminar 2020 October 20, 2020 Figure 2 Modification of SiO2 particles by amino-organoalkoxysilane via a sol-gel reaction. Table 1 Formulation of rubber compounding. 90

The 28th Special CU-af Seminar 2020 October 20, 2020 CO2 adsorption of modified NR vulcanizates NR or modified NR adsorbent material in a stainless steel reactor was treated under a vacuum condition at 60 °C for 20 min to remove the adsorbed gas inside the adsorbent material and the sample was cooled down to ambient temperature. Nitrogen (N2) gas was then feed into the stainless steel reactor by specifying in the flow rate of 50 mL/min to expel CO2 from the system. The mixed gas of CO2 12% by volume and 88% nitrogen gas by volume was flowed in the reactor by controlling the gas flow of 50 mL/min. The CO2 concentration and gas temperature at the outlet were detected by CO2 sensor. Characterization FT-IR spectra of modified NR, SiO2 and modified SiO2 samples were recorded with resolution of 4 cm-1 using Nicolet Nexus 670, USA, operating in Attenuated Total Reflectance (ATR) mode with a diamond crystal having a refractive index of 2.4 at the incident angle (θ) of 45° The contents of carbon, hydrogen and nitrogen on the SiO2 surface were confirmed by CHNS/O analyzer (PE-2400, Perkin-Elmer, USA). The samples of 2-3 mg were added into tin capsule, and then burnt in pyrolysis reactor at temperature of 1000 °C under a constant flow helium stream. The mixing combustion was separated by chromatographic column and detected with thermo conductivity detector. N2 adsorption-desorption isotherms were measured using MicromeriticsASAP-2020 at liquid N2 temperature (77 K).All samples were degassed at 150 °C for 12 h under evacuation prior to analysis.The specific surface areas were evaluated using the Brunauer– Emmett–Teller (BET) method, and pore size distributions were calculated using the Barret–Joyner– Halenda (BJH) model on the desorption branch. The thermal stability measurement of modified SiO2 samples were determined by a Perkin–Elmer Pyris Diamond TG/DTA, USA. TGA analyses were carried out under the N2 flow rate of 50 mL/min and a heating rate of 10 °C /min. The temperature was programmed from 30 °C to 900 °C. The morphology of unmodified and modified SiO2 particles was investigated by field-emission scanning electron microscope (FE-SEM) Hitachi, S-4800, Japan, operated at an acceleration voltage of 5 kV. The morphology of modified NR filled with/without modified SiO2 particles was examined by using the SEM JEPL, JSM-6480LV, Japan at the voltage of 15 kV. The samples were coated with gold before testing. Results and Discussion Characterization of modified NR Figure 3 shows the FT-IR spectra of diallylamine, NR and MNR. The distinct transmittance band of diallylamine was observed at 3300 cm-1 attributed to the N-H stretching vibration of amine. The spectra at 1642 and 1544 cm-1 corresponded to N–H bending of amine group [11]. The peaks at 993 and 915 cm-1 were assigned to C=C bending in alkene. For the spectra of NR, the characteristic peaks at 2960 and 2853 cm-1 corresponding to C-H stretching of -CH3 groups and -CH2- groups. The transmittance band at 1447 cm-1 was C–H deformation of –CH2– group. Moreover, the bands at 1664 and 836 cm-1 were assigned to C=C stretching of alkene [12]. Comparing with NR, the modified NR showed greater absorbance at 3300 and 1664 cm−1 relating to the grafting of diallylamine [13]. Accordingly, the FT-IR results confirmed that NR could be modified by diallylamine. 91

The 28th Special CU-af Seminar 2020 October 20, 2020 Characterization of SiO2 and modified SiO2 FTIR spectra of the unmodified and modified SiO2 particles are showed in Figure 4. Unmodified SiO2 particles showed the bands at 1064 cm-1 and 786 cm-1 associated with the asymmetric and symmetric stretching of Si-O-Si. The band at 948 cm-1 was due to the O-H stretching vibration in Si-OH group, whilst those at 3377 and 1630 cm-1 were attributed to O-H stretching and O-H bending from the adsorption water by the Si-OH groups on the SiO2 surface. After modification, the band of silanol group at 948 cm-1 was disappeared because the silanol groups on silica surface were bonded with modifier. Moreover, the FTIR spectra of modified silica particles showed the asymmetric and symmetric stretching and bending vibrations of the aliphatic amine (N-H) groups at about 3400 - 3300 cm-1 and 1650 – 1580 cm-1, respectively. Unfortunately, these were not evident following the SiO2 modification by these modifiers, because their absorption peaks appeared at the same wavelength as that of the adsorbed water. By comparing the FT-IR spectra of unmodified SiO2, the new bands at 2935, 2881 and 1450 cm-1 were observed for amine functionalized SiO2 which were assigned to symmetric and asymmetric stretching vibrations and bending vibration of -CH2-, respectively [14]. The presence of vibrations of -CH2- on the modified SiO2 particles indicating that the grafting of amino-organoalkoxyl groups of 1NES, 1NMS, 2NMS and 3NMS was successfully on the surface of SiO2. Figure 3 FT-IR spectra of diallylamine, NR and MNR. 92