Fuzel alkollerinden asetat esterleri üreten reaksiyonlu distilasyon prosesinin inceloenmesi ve kontrolü
Başlık çevirisi mevcut değil.
- Tez No: 55593
- Danışmanlar: PROF.DR. DURSUN ALİ ŞAŞMAZ
- Tez Türü: Doktora
- Konular: Kimya Mühendisliği, Chemical Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1996
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 128
Özet
Özet yok.
Özet (Çeviri)
INVESTIGATION AND CONTROL OF THE REACTING DISTILLATION PROCESS PRODUCING ACETATE ESTERS FROM THE FUSEL ALCOHOLS SUMMARY Being chemical engineers, our job is to produce the most of materials that people need for the purposes of their life. By now, the raw materials for these productions are decreasing continuously so, in relation to this, prices are steadily going up. Thus, this makes it obligatory to evaluate ali kinds of possible materials even wastes and by-products in order to obtain valuable products. in addition, it is also necessary both to make use of ali kinds of raw materials in conscious a way and to evaluate ali wastes and by-products in such a way that enables them to reach theirs highest possible commercial value. Unfortunately, abuses are frequently met in this area that can easily be named as prodigality. A typical example to this is the state of fiısel oil that is a by-product of ethyl alcohol production by fermentation process. Although it contains quite valuable alcohols it is used in such nonproductive areas as4efoamer-or foel additive. in this study, the chosen by-product to evaluate, the füsel oil, was first examined to clarify its structure by means of laboratory scale simple distillations. After some primary experiments it was decided to realize a detailed distillation in which the distillate was to be drawn in constant volumes for which the starting and end temperatures were also recorded. Then, a 500 mi portion of füsel oil was distilled and the distillate was dravvn in 40 mi portions each. The fractions obtained in this way were analyzed by gas chromatography using HP 5 890A gas chromatograph with a TCD detector and a Chromosorb 101 column. Results of these analyses can be seen on Table 1. in the results there is a remarkable point to evaluate. it can easily be seen that, though there is a temperature diîference of exceeding 50 °C betvveen the starting and end points of the distillation, approximately 70% of the distillate is passed within first 20 °C. This means that distillation curve is rather flat. The reason for this is the azeotropes formed by the füsel alcohols and the water it normally contains. These are aH4ow4?oiling azeotropes. VII*o 1 ; in «* ov oo t-c.* vo O.* oo, ^* y-, -» fi in t--' od -H \o «X \d vo od m' S*,^'“ ^H _, _ -ı c-ı CN r ır> t~- oo ON | ”o o O 3 C? __! tf^ oe o oo o\ x~' f>if5r4fififi'*ınm'»n^:fi m /-V.5,^ ^r O o ^i ^ ^ **\ **> **\“*. ”“l ^ **i ^ ^*. ı Şl - J -' -' ”-' -' ~ -'“ ”°' ^“o JZ o o ^ ^vûfN'*v£>Ort^poO, -- >~/ oo'.**' -»n'l''*r|? ^ ”“ **. ”* °\ ^ ''î O“* ”^ '“l R °°.ı O ^ ^ o r^ od od ON ö o vd in \d es ö 'S3îs;r-]«SfSrvir4r^mfS^< ”^ £ T3 on S l l.a s O -7; ooooooooooooo ^ O t-*-*Th^j--*Tj-'!t'*'*'* 4-< d ı 2uO O O f>.J3 -yr«~, vo O f^ O“i p| VC; od oo' 00 fS,. £r^r^'*>nxOOo”Ö'OO'- rt“invor~oooN2^2'55 2Bgpij H ^ vmTo break-up this azeotropes a set of azeotropic distillation experiments with benzene as additives were carried out. Drawing of the distillates were observed in theirs volumes and temperatures. When the distillation curves were formed it was seen that ali of these were not too much different from the distillation curve of the original material with respect to theirs characters. Two remarkable points were recognized. First, as the ratio of benzene was increased the more of the distillate was drawn out at a lower temperature (below 90 °C). Second, increased amount of benzene lowered the starting temperature of the distillation asymptotically. The reason for this is the binary and ternary azeotropes formed by benzene with water and alcohols that ali have quite lower boiling points than ordinary alcohol-water azeotropes. The first coming approximately of 24% to 32% portions (according to benzene ratio) was ali two phase distillates with lower phase consisting practically of water. The second coming portions (above 110 ”C approximately) was single phase distillates consisting of practically i-butyl and i-amyl alcohols. in doing so the water of fusel oil was separated efficiently. But meanwhile ali phases were found to be contaminated by benzene. Because it is a strong carcinogen, benzene contamination is fatal when it is took under consideration that ali fusel acetates especially heavy ones are frequentiy used for domestic purposes. On the other hand separation of water in this way, also causes the separation of heavy and light alcohols. This, as it will be clear later, is undesirable for the practice of reacting distillation process, for it is necessary for the feed stream to contain some amount of light alcohols in order 4o enable 4he reflux mechaaism 4o «perate. Following this, a set of experiments were carried out to draw out the water of fusel oil by use of anhydride Na2SÛ4. Here, optimum operation conditions were determined as 0.15 g Na2SC>4/cc Fusel oil salt döşe and öne hour mixing time. in these conditions the \vatercontent of raw fusel oil waslowered from 20% to 9%. This draw-out of water influenced the distillation curve of fusel oil greatly. But it was not found to be an industrially apph'cable method. Meanwhüe, such a decrease in the water content was not of great importance in the view of reacting distillation practice. After ali these experiments, it was seen that applying a preprocessing to fusel oil was not a fruitful alternative. Thus, use of fusel oil after removal of the non volatile residue by a simple distillation was accepted as the most reasonable way. The most important advantage of this method was that different fractions of fusel oil obtained in pretreatment could be experimented as feed stocks. IXit will be illuminating to çite about the conditions of the gas chromatography employed in the analyses of samples here. Primary works on this subject were canied out by a SH1MADZU GC 6A chromatograph with double“hollovv electrode”type and high sensivity FID detectors. The selected column for this purpose was a Carbowax 20M column. Its measures were 2.2 mm i.d. * 2 m anda 15%coverageon 80/100 mesh Chromosorb W/AW diatomite support. Because the FID detector was not sensitive to water, no remarkable problem occurred during the analyses of fusel fractions except the absence of the percentage ofwater. Later a new chromatograph, the HP 5890 A with a TCD detector was obtained. By use of this new instrument, it was seen that Carbowax 20M column had some serious problems. First, water peak comes between ethyl alcohol and n- propyl alcohol peaks and disturb the after comings by extensive tailing. Second, Carbowax 20M column was found to be unsatisfactory in the analyses of esterification products especially because of excessive distortion of acetic acid peak. Peak tailing was found to be result of high hydroxyl affinity of the support. This was tried to be corrected by silanisation with hegsamethyl disilasane (HMDS) but no satisfaction could be obtained. This situation obligated a search for a new column. By evaluation of acquired experience and under the lights of gas chromatography literatüre, it was decided that a Chromosorb 101 column might possibly give satisfactory results in the analyses of fusel fractions and the esterification products of them. After obtaining ali the necessary materials, a Chromosorb 101 column of 3mm id. * 2m measure was prepared in its proper way and tested after conditioning vvhich was absolutely necessary for ali such porous polymer columns (Chromosorb Century, Porapak and Haysep series). in tests realized with fusel fractions, Chromosorb 101 was found to be excellent. Because it has very low hydroxyl affinity the problem source, water, came first with no tailing and by means of a simple temperature program ali other peaks were obtained in quite well forms. On the side of esterification products acceptable results were also obtained. No problem was encountered with acetic acid. Some peak coincidence problems were observed but these negative situations did not cause too much serious problems because of proper distribution of esterification products along the esterification column. XBefore beginning to the primary esterification studies, it will be helpftıl to glance at the properties of fusel alcohol acetates. Here, the most remarkable point in respect to distillation practice is that these esters form lots of azeotropes with theirs alcohols and water. in fact, the cause which gives an excessive complexity to the multicomponent reactmg distillation process is the existence of these azeotropes. As knovvn, esterification is an equilibrium process. To drive it to completion it is necessary to draw out at least öne of products. According to the reaction realized, this may be ester ör water. in the production of fusel acetates, determination of the outgoing point of the water is öne of main problems. If it is decided to evaluate ali of the fusel oil, solution is rather easy. in such a process reaction water and the water from feedstock can be dravvn out as the top product along with light esters. in this case, top product will be two phases and lower water phase can be discarded by means of gravitational separation. Resulting ester phase can be both refluxed partially and drawn out as top product. Meanwhile heavy esters consisting of i-amyl and i-butyl acetates are stripped down to the reboiler and be dravvn out as bottom product. Now, it will be usefiıl to çite about design, manufacture and construction of the distillation column reactor used in this study. Among ali column types for distillation, a packed column model was chosen because of both economical constrains and ease of manufacture. Column body was consisted of four 40 cm long glass pipes of 8 cm diameter. Rasching rings of 5 mm diameter were chosen as packing material. These compartments vvere fitted öne above other by means of 10 cm long Teflon couplings. These couplings have five ports; two for feed streams, two for liquid and vapor sample drawing and öne for the temperature probe. At the top a total condenser was fitted after a U turn. Below condenser was a condensate reservoir to collect ali the top product and to let it to separate in to two phases. Ester phase was dravvn out from side port and send to a reflux divider. Water phase was dravvn from the4>ottoffl port of the collector. Beneath the column was a reboiler consisting of a 20 l glass flask and a 2.5 kW heat mantle fitted to it. Reboiler flask has a filling port at its upper side, a column connection at the top and a valved port for bottom product at its bottom. Temperature measurement was realized by a ten channel digital thermometer \vith 0. l °C accuracy. Probe locations are, ambient, reboiler liquid phase, reboiler vapor phase, 1.packing, 2.packing, 3.packing, 4.packing, reflux input to column, condenser water supply, condenser water return. XIAfter necessary preparations some experiments were realized to observe the behavior of the reboiler at heating and cooling periods. It was found that while they go in-step at heating period, reboiler liquid and vapor phase temperatures behave in different ways at a step down in the reboiler heat input. Vapor phase temperature was found to be ten times as much sensitive to any step down. So, this parameter was selected to be measured variable for the reboiler temperature control system. By evaluation of the data obtained, it was decided that a dual slope PID controller must be used. Well-known digital PID algorithm was rearranged and a polarity detector was joined. The controller routine was operated with different controller parameters according to the polarity of the error detected. Controller parameter sets for negative and positive errors were obtained from Cohen - Coon equations using step-up «id step-down 4ata. When experimented on the plant, the new PID controller was found quite satisfactory. After all step inputs, the controller brought the reboiler temperature to its set point with a good performance. The only drawback observed was that the PID controller became late to catch the set value even though its output goes near saturation when put on-line and when feed rate doubled. In this test, column head temperature was fallen below the critical value 79 °C, and this prevented the water to go out as part of top product, resulting a temporary spoilage in column operation. To overcome this problem a generalized controller was synthesized. This controller was synthesized in such a way that it was capable of totally removing the steady state and velocity errors and forcing the reboiler to behave like a second order process with damping factor Ç = 0.707. When experimented, this controller was found very satisfactory. This third order controller caught set value quickly enough and lead no spoilage at the regime of the column head, though produced an oscillation with amplitude of ± 2 °C around set point. This oscillation was partially a result of predetermined slight oscillatory ( C, = 0.707) second order behavior. In addition to this low input - output accuracy ( = 8 bits) of the control system used led the controller be late at both measurement and acting. So, it was decided that a control interface with 12 bit accuracy was necessary in order to get full satisfaction from such controllers. XIVAfter all these experiments it was decided that some reorganization was necessary to keep the study in right a way. At first, feed rates were guaranteed by use of reciprocating type dosage pumps with Teflon diaphragms. Second, column body was isolated to prevent the self reflux. Then whole system was repaired to prevent the leakage's that frequently encountered, and to enable the sample ports to operate which were all plugged because of heat effects. As a novelty it was decided to use a lighter fraction of fusel oil to increase the amount of the top product. In doing so, it was expected that reflux mechanism would be examined more conveniently. The composition of this new alcohol mixture is given in Table - 3. Table - 3 Composition of the New Alcohol Feedstock Component % Amount After prepared, a set of second stage esterification experiments were realized. First of all, the optimum feed rate was determined as 2 kg/h in alcohol basis. Second, it was found that to start continuous feed, it was necessary to wait the reboiler temperature reach 125 °C in order to minimize the shock phenomena encountered. After exerted rigorous efforts, it was decided that it is virtually impossible to provide a steady state at the reboiler temperature by manually adjusting the heat input. This situation obligated a robust controller for the reboiler. In addition currently used overflower type reflux divider was found to be unsatisfactory for its operation regime was heavily influenced by the changes in the rate of ester phase flow coming from condensate reservoir. Finally, it was decided that an automatic system to draw out the water continuously from the bottom of condensate reservoir was absolutely necessary to provide a continuous flow of ester phase to the reflux divider. Before starting the third stage esterification experiments a computer based control system with 8 channel input and 3 channel output was established. To the column were fitted a digital reflux divider, an automatic water drawing apparatus with infrared eye mechanism and a triac module with 25 A capacity to command the reboiler heater in order to control the reboiler temperature. All of these were designed and implemented in our laboratory. XIIIAfter necessary preparations some experiments were realized to observe the behavior of the reboiler at heating and cooling periods. It was found that while they go in-step at heating period, reboiler liquid and vapor phase temperatures behave in different ways at a step down in the reboiler heat input. Vapor phase temperature was found to be ten times as much sensitive to any step down. So, this parameter was selected to be measured variable for the reboiler temperature control system. By evaluation of the data obtained, it was decided that a dual slope PID controller must be used. Well-known digital PID algorithm was rearranged and a polarity detector was joined. The controller routine was operated with different controller parameters according to the polarity of the error detected. Controller parameter sets for negative and positive errors were obtained from Cohen - Coon equations using step-up «id step-down 4ata. When experimented on the plant, the new PID controller was found quite satisfactory. After all step inputs, the controller brought the reboiler temperature to its set point with a good performance. The only drawback observed was that the PID controller became late to catch the set value even though its output goes near saturation when put on-line and when feed rate doubled. In this test, column head temperature was fallen below the critical value 79 °C, and this prevented the water to go out as part of top product, resulting a temporary spoilage in column operation. To overcome this problem a generalized controller was synthesized. This controller was synthesized in such a way that it was capable of totally removing the steady state and velocity errors and forcing the reboiler to behave like a second order process with damping factor Ç = 0.707. When experimented, this controller was found very satisfactory. This third order controller caught set value quickly enough and lead no spoilage at the regime of the column head, though produced an oscillation with amplitude of ± 2 °C around set point. This oscillation was partially a result of predetermined slight oscillatory ( C, = 0.707) second order behavior. In addition to this low input - output accuracy ( = 8 bits) of the control system used led the controller be late at both measurement and acting. So, it was decided that a control interface with 12 bit accuracy was necessary in order to get full satisfaction from such controllers. XIV
Benzer Tezler
- An investigation on IMC based dual-phase PID controller
Dahili model kontrol bazlı iki fazlı PID kontrol ediciler üzerine bir araştırma
EBRU CEBECİ
Yüksek Lisans
İngilizce
2005
Kimya Mühendisliğiİstanbul Teknik ÜniversitesiKimya Mühendisliği Ana Bilim Dalı
DR. HİKMET İSKENDER
- Effects of various carbohydrates on electricity generation in microbial fuel cell
Çeşitli karbonhidratların mikrobiyal yakıt hücrelerinde elektrik üretimine etkileri
TUNÇ ÇATAL
Doktora
İngilizce
2008
Biyoteknolojiİstanbul Teknik Üniversitesiİleri Teknolojiler Ana Bilim Dalı
DOÇ. DR. HAKAN BERMEK
DOÇ. DR. KAİCHANG Lİ
- Fusel yağı alkollerinden asetik esterlerin elde edilmesi sürecinde optimum çalışma şartlarının tesbiti
Determination of optimum operation conditions in the process of the production of acetic esters from the fusel oil alcohols
SALİH YÜKSEL YEĞİN
- Buji ateşlemeli motorlarda amil alkol kullanımının deneysel ve yapay sinir ağlarıyla incelenmesi
Investigation of using amyl alcohol on spark ignition engine with experimental and artificial neural networks
SAMET USLU
Doktora
Türkçe
2018
Makine MühendisliğiKarabük ÜniversitesiMakine Mühendisliği Ana Bilim Dalı
PROF. DR. MUSTAFA BAHATTİN ÇELİK
- İçki ispirtosu üretiminde çavdar kullanımı üzerine bir araştırma
An Investigation on the use of rye in the production of ethanol for beverage
MÜGE KAYA