The Extraction and Precipitation of Pectin Citation Ferkel, Albert Jefferson (1925) The Extraction and Precipitation of Pectin. Bachelor's thesis, California Institute of Technology. doi:10.7907/ehz1-ee02. https://resolver.caltech.edu/CaltechTHESIS:08062024-184852993 Abstract No abstract. Item Type: Thesis (Bachelor's thesis) Subject Keywords: (Chemical Engineering) Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemical Engineering Thesis Availability: Public (worldwide access) Research Advisor(s): Lacey, William Noble Thesis Committee: Unknown, Unknown Defense Date: 1 January 1925 Record Number: CaltechTHESIS:08062024-184852993 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:08062024-184852993 DOI: 10.7907/ehz1-ee02 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 16613 Collection: CaltechTHESIS Deposited By: Benjamin Perez Deposited On: 06 Aug 2024 19:40 Last Modified: 06 Aug 2024 19:40 Thesis Files PDF - Final Version See Usage Policy. 6MB Repository Staff Only: item control page

THE EXTR.A.CTION fl.ND PP.ECI"PTT.ATION OF PECTIN Thesis by Albert J. Ferkel In partial fulfillment of the re~~irements of the Degree of :Bachelor of Science in Chemical Engineering. CALIFORNIA IUSTITUTE OF TECHNOLOGY Pasadena, Cal ifornis. 1925 Puruose of the Research The resea.rch wae undertaken in an effort to find an economical and ef : icient means of extracting pectin from orange peels and to find a method of precipitating the pectin and so obtaining it as a A secondary or preparatory purpose was to find a convenient solid. and accurate rm3·~hod of determining the amount of pectin present in a solution. Pectin is lmoYm to be a collald, but its exact cornposi tionis r,ot l Th. von Fellenburg says that pectin. is composed of a number known. of compounds and assigns the following composition to orange pectin: a.raban, 36.1 %; methyl pentosan 6%; gala.at an, 49. 3%; and methyl 1 alcohol, 11. 51:.. Colloids can ·i,e divided into two classes. The first of these imbibe and retain water and the colloidal particles contain much water. Sols of such colloids rna:r have high vhcosi ties and a.re only precipitated with difficulty. philic. Such cc,llloids are called hydro- Those of the second class are termed hydrophobic and are substances to which can be ascribed no special relationship with water. A sol of a hydrophobic colloid shows properties very li~tle different from those of the dispersion medium. The viscosity of .such sols is, for example, very little different fro~ that of water. Hydrophobic sole are very easily precipitated by the addition of an electrolyte or by properly adjusting the rr+ ion concentration. l J. c. s., A, 1915 i - 774. 2 Hydrophilic compounds a.re, on the other hari.d, not precipitated by electrolytes and not easily precipitated by adjusting the Ir+ concentration. The· follo·,ving experimental work shows pectin to be?'hydro- Sols of hydrophilic colloids can be precipitated philic compound. by and With other colloids of opposite charge. Pectin is a negative colloid and as such it is to be expected that it will be precipitated by a positive colloid. shown below. Thia is found to be the case, as Will be When one colloid is precipitated with another, the two are so closely intermingled as to make complete eeparation extremely difficult. This is particularly true in the case of pectin precipitates as solvents in which the precipitating colloid is soluble make a gel of the pectin. This applies particularly to aqueo'\18 sol vents. Previous Wo.J"~ Pectin has been sold for some years, but hae principally been marketed as a solution rather than as the solid. The advantages of solid pectin over the solution are nu.~crous, including reduced cost of packing and shipping, and the possibility of using small amounts from a large container without danger of the remainder of the lot spoiling. By far the greater part of the pectin of commerce is at present produced from non-citrus fruits, berries and apples being the principal sou~ces. There is one company, the Exchange Lemon Products Co., located at Corona, California, which is producing dry pectin from citrus fruit peels. The pectin ple.nt is operated in conj1.1nct1on with a c•tric acid and aroma.tic oil plant. '.rhe extraction 3 is carried out using a dilute acid solution a~ the extracting medium and the precipitation is effected by the double colloid method, using Al(OH) 3 , precipitated from an AlCl3 solution as the precipitating colloid. The preci~itated colloid mixture is then treated with al- coholic HCl to remove the .Al (OH) 3. This removal is not complete, tests made on the product showing around 10% Al{OH)3 in it. The alcoholic HCl washing is, in addition, rather expensive .and not very convenient. The water concentration nmst be very low to avoid forming a gel of the pectin and thus making it unhandleable. This makes it neceseary to use either low HCl concentration or to 1.,se gaseous HCl to further saturate the alcohol without adding more water. .lllC1 3 is, moreover, not very soluble in a. water-alcohol mixture which a.dda to the difficrul t :,, of effecting complete wal!hing. Some experimental work has been done to det ermine a satisfactory method of detemining the amount of pectin present in a solution. The four general methods which have been used for qusntitative determinations of pectin a.re: precipitation of pectin with C2l4. 5 0H, precipitation a-s calcium peotate, deter-mination of vhcosi ty of solution a.nd. taking this as a measure of the pectin present) and det ,3rmining the gelling power of the pectin present by forming a gel with suga.r. Each of these r.-:ethods pos sesses some disadvantages. The alcoholic preaipitation is said. to be inaccurate and the results of Frederick Ra.rdf 2 and Haynes and Carr~3 show this to be the case. The calciw.n pectate precipitation method., developed by Hc::.ynes and Carre"3 is apparently acct"<ra.te, but it :possesses the disadvantage of being a. 2Frederlck Hardy , Biochem. Journ., Vol. XVIII, no. 2 (1924). 3naynes and Carrl, Biochem. J curn., Vol. XVI, pg. 60-69 (1922). 4 vnry involved. procedure and req:-uiring a long time to carry u"l.l.t. The minimum possible time is 15 hours and the recommended time 2 days. The viscosity of a solution is not a true measure of the aznount of pectin present therein. In the first place most solutions to be teated contain other substances whfch affect their viscosities. Secondly pectin, beine; in colloidal suspension rather than in true solutiont does not govern the viscosity of its solutions in the manner i:r. which a true solute would.. Thirdly, unless a special forra of Vis- cosity pipette is used, the reaul ts are found to be absolutely worthlesa; due to turbulent ratll::,r than viscous flow in the stem of the pipette. Frederick Hardy¾as proved the invaliiity of viscosity fOOaaureinents taken with an ordinary pipette. Inasmuch as the value of pectin liee in its power of gel for~a- d i' -,-- .-.,: t tion the direc:t. r:..easurement of this power should be the best way of estimating the amount of ~ectin pregent in a solution. was used by Joh.."lstin and Denton4. This method The difficulties encountered in attempting to use this method are outlined ,mder exparimental work. The extraction of pectin from the orange peel is shown to be beat carried out in an acid solution. F. P.ardy2 made a number of teat runs, uaing various concentrations of HCl and citric acid, ranging from .002 N. to 1. 5 N. Re obtained the best results by the use of a 0.25 N. citric acid solution, carrying on the extraction for 20 minutes at 9S°C, then allowing the mixture to cool £or 10 minutes. Hardy's results also indicate that complete pectin recovery can not be expectad in one extraction, the first treatment 4Johnstin and Denton, Ind. Eng. Chem., 15, 780 (1923). 5 removing slightly over 50%of the e;ctra.~tabl -~ :pectin from the peel E:.n d ea.ch succeeding treatme11.t removing; about pectin left by the preceding extraction. 50%of the extractable The results of Hardy are particularly valuable aa they are cr_,_ i te extensive and as the 2mo·.mt of pectin pres ent in ea.ch extract was determined i n three ways by alcoholic precipitation, by preciyitation as calciu:.~ pectate, and by viacosi ty measurements. Experimental Work It was firat desired to det ermine the best means of estimating the amount of pectin present in a solution. As stated above, a direct measurement of the gelling power seemed to be the best way cf determining the value of a solution of pectin, so this method was tried. If it were possible to determine the relative rigiditiea of a series of gels rna.de from pectin solutione and sugar, thia should prove to befvery satisfactory way of detennining the relative gelling powers of a series of pectin solution,. A visoosimeter is clearly unsuited for this purpose, as once the structu.re of the gel is broken its viscosity means little. It was suggested by Dr. Lacey that it might be possible to determine the relative hardness or resistance to penetration of a s er ies of gels by the use of an instrument similar to the penetrometer us~d for asphalt testing. Acting on this suggestion an effort was made to adopt a penatrometer to gel testing. The needle used in asphalt test- ing is, of course, far too small in cross-section to give a.~y re~ding as it drops through the most rigid gels with almos t free fall velocity. 'A plunger of greater cross-aection area was tried, but found unsatis- 6 f actory , du e to the nskinll which forms at t he surface of the gal. }:Ven when this nskin" is removed a. plunger of large enough crosssection to move with a. measurable velocity doea not move at a ilniform speed when penetrating the gel. The motion when it is pene- trating tha upper layers is slower than when penetrating the 'bottom layers, in spite of the increased side friction. As the penetration tests were made on jellies which were about 20 hours old the variation of rigidity in the different layers can not be ascribed to incomplete "setting" of the jelly due to lack of time. Believing that a penetrometer test could not be used eatisfaetorily to deter:nine gel strength, it ·,r;as decided to rna..'lce a sta..vi.d.ard test se:r.ies cf gels from 1mown q_uanti ties of pectin and suga.r in such a way that a sample gel could be made under the aarne condi td.one from a pectin solution ,vhose strength it was desired to ascertain. 'The sample gel would then be compared to the standard teet series and it waJ. ld be considered to contain the same amol!lt?lt of pectin as the standard gel it most resembled in consistency. The pectin us ed for making the test gels was purified co mrLercial pectin. The co?nrter- cial pro duct, before purification showed over 10 % Al ( OH)] 'by co:nbustion anal ysis. T".nie commercial p ro duct wa~ waqh ad repeatedly w1 th HC1- C2 H5 0H mixture, the mos t efficient mixture conta ining 15%concentrated hydrochloric acid and 85%ethyl alcohol. S8lutione stronger in HCl could not be used because the water present partially dissolved the pectin making a gel wh ich, of course, e-ould not be wa shed. .After many washings :<1i th the ahove mixture, the .Al s alt content of the 1 pectin ,,,vas red-\lOed to 3.1 % , but could not be brou ght bel ow thi s figure. 7 This shows one of the disadvantagea of the do~ble colloid precipitation ruethod b y wh ich this commercial pectin was produced. were made from thie purified pectin as follows. The gels The amount of pecUn used for ea.ch gel was dissolved in 20 cc. of.05 N. citric acid. To this solution wae added 12 grams of sugar, the sugar dissolved and the mixture boiled until a boiling point of 105°c was reached. The solution was then poured into molds and allowed to sta.~d for 20 hours when the rigidity or texture of the reeulting gels was noted and recorded. Sample no. Table I gives the results for a series of standard gels. Grams pectin 1.0 Relative viscosity Rel. vis~osity divided by grams pectin . Consistency of gel after ·20 hours. .All pectin did not dissolve. 6.64 13. 3 elastic solid 3 0.2 3.09 15. 5 viecous solid 4 0.15 2.12 14.2 semi-solid.; w,:rn.ld not pour. 5 0.14 14. 5 just pour 6 0.13 15. 5 very viscous 7 0.1 17.2 entirely lig_uid 8 o.os 19.8 9 0.06 23.7 10 0.02 58.5 g Tha relative viscosities given in the tfiird column are the visc:osi tiea of the pectin solution before the sugar was added. Columns 3 and 4 are given to show that the viscosity of a pectin solution is :-iot a direct measure of the arooi.mt of pectin :present even though no other viscous materials be present. If the pectin present were in direct ratio to the viscosity, the figures of cclumn 4 worud have a constant value. In such a series of gels only a few are of value, that is to say, only a few have a consistency which can be definitely fixed and which is proportional to the aruo-:1nt of pecth. present. For example, if a gel is an elastic solid its consistency will apparently be the same even thcm.gh a large excess of pectin over the amount . needed to make the gel an elastic solid be present. Similarly, if the mixture is a liquid we only know, under the condi t1dms of this test, that there is less than 0.1 grams of pectin present in 20 cc. of the , soh1tioi1. Only those gels whose ccnsistency is on the borderline between solid and liq_uid can be referred to. arises here. A further diffic-.il ty The consistency of the gels changes on standing 1 even after the first 20 hours, so that either the tester mt.st ramernber the exact consistency of each of the sa...-rrple gels, or make up a new series each time it is desired to test some solutions for pectin; or, in the event that test& are being run continuously, make new standards at least once lia.ily. Possessing ,·vnat 8Pl)eared to be a fairly satisfactory method of determining the pectin present in solutions, experimental extraction run& were made. T11e first extractions were made using 0.25 N. and 0.1 N. citric acid as the extracting liquid. In each case 50 gme. of 9 fresh peel was extra ct ed. with 100 cc. of the citric acid s olution , the peel being c1.1t into strips roughly 1/8" wide before extract,iol'l and the extraction being carried out for 20 minu tes at 98°0. The re- sults of this ru..~ are given in Table II. Sample %:Extract no. in s ol 1n. * normality of extracting acid, Std. gel to which Indicated % saraple corresponds pectin in extract. ·- - -·· no. 2 l 100 0.25 2 75 0.25 3 50 0 .. 25 li quid 4 25 0.25 fl 5 100 0.13 no. 3 6 75 0.10 l i q_uid 7 50 0.10 II g 25 0.10 II II 3 1.0 - --- - - - - -·- The resu.1 ts of this r 11n show the hi gher acid concentra·t ion t o be the best and also show the ina ccuracy of the metho d us ed for determini ng t he p ectin concentration. T'ne figures in column 5 for samples 1 and 2 aho-u ld agree, but do not do so showing that, al though gals ,.-;ere prepa.re-i under idenUcal conditions. as far a s poss ible , and carefully cor.u-pc1red to the s t andard gels , a pe rfect or evel'l a close dete r uiination did not r esult. *The extract was diJ.uted aa s}1ovm in cohmm 2 to 1t18J-: e i t r-o ssi"ble to g et a ser ies of gels fro m e a ch extract. 10 Sir;ce the extraction hs.d oean so t horoughly i nvestigated by ria rdy a nd. s:i.nc e the time al l otted f or tne research waa r apidly pa ssing, i t was t hcc:ght advisable t o t ak e the extraction outli:c1ed ab:we, using O. 25 N citric acid, a s the standard l) rocedure ai1.d to pass on to the more interesting and 1.es s thoroughly investigated problem of precipitation. As mentioned a,bove, ·under "Theory", recth: i s 1:1 rJ.ega tive colloid and. , as such, it was thought that it might be preoipitatad by the simple addition of an electrolyte. If sucl-l s. precipitation wer-e poss i ble , it s:C.:.ould ·oe brought s,bout more reaclily by a.YJ. electrolyte containing a polyvalent positive ion than by one containing a monovalent positive ion. It would be cor::venient if the salt were s oluble in C2H50H so S0 L1tbns of va ri c,·,;. 13 s t r engths of the foll owir.g salts wa-:- s used.: FeClJ, AlCl.3 1 and 1'TiC1 2 , but the pectin was no t p recipit ated• . Some colloicl& are sensitive to changes in the HT ion concentra tion s o pectin so lutions ware titrated very slowly wi ti:l T-ICl Wld Ne.CH scJ.utions in ox, ende avor t o find a !tt" ion concentra tion w}:.ich would bririg about the precipitation. No sue)? concentration was found , the only preci1)i tat ion taking place in a NaOH solution of cons i <Iarable strength , t lLe precipitate being dark in color and ap})arently not pross essine t he properties of rectin . lytes The non-preci p it ation of positive i on electro- and ·oy W ion ad.Jil!5trnent substantiat e!3 t,h s theory that :pecti:,;i is a hydrophilic colloid. Ina smuch as the precip itation i:::o,.:..ld no~ ce succe sl:'. f1;lly n:_,1.cle by means of $ ingle salts or by the adj v,s tment of the H+ ion, i 'v waG de c i ded to hw estigate the ~,oas ibEi tie!.! of ::: 011":; i;;rci al alcohcl:lo 11 p recip itation. :-, f T}le double colloid precid t at io n was re jected. ·o ec aus e the ,iif:ficul t y cf sepuratin2; the col l oid mixture resulti ng from :n.1ch a p recip i tation. In order to p r oduce a :.c reci p i tate tha t wilJ. be cor.c.pletel y di ssolved when it is desired t o u.s e it , it is ess ential tha t the s olution from wh ich the preci ; i t at ion i ,;; t o be made be perfectly clear . It is only by filtering that a clear soJ.uo'c ion can be obtai ne d; for the suiill particles of :pv.lp wl":.ich a r e susp ended in the so lution extrac·ted from the peels are of such a density t hat they will not settle out from the so lution, eyen after prolonged standing. In c onnection with this filt r ation the advi sability of using filt e rcel, s. pu:dfied. for m of diatomaceous earth 1 was investigated. The first resu lts of this invest i gation are given in Table III . Table III :Ho. Acid used 1 0. 25 N. ..,. Pre coated filter paper. 3 n o. 25 n. Condition of f'il t. Wt . of C2HsOH ppt . from 10 cc . extr2.ct y es 6.5 cl ear , light in color 0.12 grams no 21.4 cloudy, light in color 0.28 " no 9.65 clouey, dark in color 0.41 " ci tric 2 --------Rel. visco s ity of filtrate HCl --- - -- ----- ·--·- -- - - --- - - - - ------- - - . The pap er for sample 1 was precoiited with fil tercel t o a depth of one to t v,o rr.illimet ers . .All viscositi es were obt ained. oy the use of a viscosity pipette, having a squat bulb and a long capill a r~r atem. The alcoholic pre cipitates were obtained by a.iding the filtrat e to an sg_ual a.::1o·unt of a lcohol, the addition of fur th er alcohol producing no 12 further p recip itate . The da ta of t his te,bl e h1dicates that the f ilt e r-- zel adsorbs a considerable portion cf the pectir.. out as onl y a relatively small amo1;nt of the so l ution was run t hr ough the filter, theee results must not be given t oo nm.ch weight. This tabla also shows t h at HCl , while giving a l a r ger alcoholic :precipitate than citric acid, is not s d table beca.1.1-se the filtrate and a.lCWJholie precipitate are very da rk in color. T'.cie extrG-ct. fil terecl through the preccated filter less rapidly than thr o1:-gh the filter which was not preccated. Further tests gave the same reaults. The proper ~no~nt of alcohol to be used to ins ure complete preci-pitation was determined in the fellowing manner. To a number of 10 cc. samples of the unfiltered extract, obtained by the method which was adopted as standard for extraction, were added vari ous amo,mts cf C2 HsOH and the p recipitate filtered out. :F\rrt.l1er C2 HsOH was then aa.ded to the filtrates and the filtrate originally containing the smallest amo.unt of C2 Hs0H which did not become cloudy on the further addit.ion of C2 HsOH was considered a s l1av i ng the proper amount of C2 R5 CR. By thi s r,rocedure it was f ound that e qual vol 'Llmes of extrac t and alcohol :produced corq)lete preci:: :tation without the -~se of excess alcohol. As the preci pi tation is dep endent on alcoholic concentration the arno1..1.nt of pectin present, does not seem to change the ratio of H2 0 snd C2 H5 0H necessary to produce con.i-plete precipitation. The alcohol precipitate was somewhat gelatinous bl. character, but if precip itated col d by a sufficient g:uanti ty o f C2 F.sOH it was easily sep arated fro m the liquid by fil tra,ti o~. If, however, it was p reci pi tated hot or b y an insufficient quanU ty of C2 HsOH it was extremel;r di fficult to hand.le. It was fo -;;:nd that a temperature of above 5() 0 0 cou ld not be used i n dry-ing the :preci ~)i tated pectin wi thout danger of it turning dark in color. Paper not being practical for a conYnercial filtering machine tests were run using duck ae the filtering medi'U?ll. 260 grar.ns of fre sh orange peels were he a ted with 520 cc. cf 0.25 N. citric acid for 20 minutes at 98°0, on a water bath. The solution was decanted and drained from the 1 urge pieces of peel and divided into three taroplea, each of 75 cc., wh ich were treated as followa: Sa1nple ~-• Filtered through precoated duck after ad- mixture of l gram cf filtercel. Alcohol precipitate - 0.22 gram Filtrate clear. Filtered easily. Sample g_. Filtered through precoated duck, no admixture of fil tercel. Filtrate clear. Filtered with considerable difficulty. Jicohclic preaipitate - 0.29 gram. Sampl e 3. Filtered througl--. plain duck. Filtrate cloudy, but cleare~ up by repeated filtration through same filter. Filtered very e&sily. Alcohol precipitate - 0.35 gram. These resu lts reaffil"""~ the adsorption cf pectin by filtarcel ~nd also show that its use is not essential t o a clear product. Some tests on the advisability of running a second extraction on the peels and which showed that this was high desirable concluded the laboratory work. Pectin, or as it is sometimes called, pectinogen, is a negative hydrophilic colloid whose sol is not sensitive towards hydrogen ion concentration. It can best be extracted from or..nge peels by the use of a 0.25 N. so l ution of citric acid and can be precipitated from this solution by the addition of a quantity cf C2 HsOH equal in volume to the extract. The extract should be clarified by filtratio~ through du.ck before the precipitation is niade. The use cf fil tercel or similar substances is not recommended, as they adsorb considerable quantities of pectin. The alcoholic preoipitation of pectin should be commercially :practicable, if it is put on a large enough scale to make possible the use of efficient machinery, solvent-recovery driers, and rectifying stills for the alcohol used. Of the methods proposed for the estimation of the amount of pectin in a solution only the calcium pectate preCi}) i tation method is accura te and it takes too long to be of great use. I wish to thank Dr. w. N. Lacey under whose direction this work was carried out, fot the !ni.ggesti on "to undertake the wo rk and for the suggesti ons he of.fered during the progres s of the inves tigation. 15 Summal';i_ An effort was ~;.a.de to find a method to comrnercially produce pure pectin. Extraction by means of 0.25 N. citric acid solution heated to 98°C for 20 rninutes in contact with the c·v.t-up orange peels was fo«nd to be satisfactory, but for good yield the extractio:n proces s sho~ld be repeated. The extract c~n be filtered through 'duck to give a clear solution. by the addition of an eg_ual extract. Precipitation can be ca~ried out volume of C2 H5 0H to the clarified T't1e precipitate is easily filtered and must be dried at a low temperature. The disadv2..ntages of the vari ous methods of r,ectin esti~.ation are pointed out. pectin can not be precipitated by the addition of electrolytes or by Ir'" ion concentra.tiom. adjustment. Pectin precipitated with a positive colloid can not be comple·tely freed· from this precipitating colloid. 16 131'blic_g~ Freundlich - Elements of Colloidal Chemistry. Chen:ical .Abstracts, Vol. 15, page 3349. J. Il'ld.. Eng. Chem., 14, 710-ll (1922). J. C. S. A, 1915 1 - 774. Hardy, Biochemical Journal XYIII, Ne. 2, 1924. Chem. and Met. 31, 198 (1924). J. Ind. Eng. Cbem., 16, 1042-1044 (1924). Carre (1922) Biochemical Journal, 16, 701J. Clayson, Norris and Schryr,er (1921), Bicche~ ical Journal, 15, 644. Haynes and Carre (1922) Biochemical Journal, 16, Johnstin a..--id Denton (1923) J. Ind. Eng. Chem., 60. 15, 750. Noyes, H. A. (1923) Ind. Eng. Chem. 15, News Ed., March 10, p. 3.