Providing Vegetable Stearic Acid Super V 1895 S From Hidrogenated Of Crude Vegetable Stearic Acid HCV 1895 S With A Single Fractional Distillation Column
lnternational Journal of Engineering & Technology IJET-UENS vol; 1l No; 04
PROVIDING VEGETABLE STEARIC ACID SUPER V 1895 S FROIVI HYDROGENATED OF CRUDE VEGETABLE STEARIC ACID HCV 1895 S
WITH A SINGLE FRACTIONAL DISTILI-ATION COLUMN
Muhammad yusuf. Ritongal
l,Chemical Engineering Department, Faculty of Engineering, University of Sumatera Utara, Jl. Almamater Kampus USU, Medan 20155, Indonesia
E-mail : yu s ufrit @ gmail. com
Abstract
ln 2005 stearic acid V 1895 S is manufactured for l*ttime in Indonesia base on the natural vegetable oil, Crude palm Cil to apply Twin Rivers's demand. One of it specific quality (iodine value 0.15 gll00 g maximum) is very influence to ,re purifying steps, one of the steps is purifying of hydrogenated crude V 1895 S (HCV 1895 S) feed by a single
:olumn of fractional distillation to produce V 1895 S with the lowest possible of it iodine value, lower than the
naximum value. The results of application and small adjustments of determination operating conditions of the single ;olumn priority based on the feed composition, product composition of V 1895 S and it pressure drop character are ;ompared based on it quality standards. This is the method used in this research to provide V 1895 S and conducted in ihe plant scale of PT. XXX factory with production yield 88 - 89 Vo of 0.90mt HCV 1895 S, unsarurated content is ioiver than 0"15 g/100 g and higher than 98 7o wtw Cl8 purity. In general the specific quality standards of V 1895 S
can be met"
Keywords: quality standards, fractionation, composition, unsaturated contenl, adjustments
1. Introduction
Fany acid ofV 1895 Sis meansuper fatty acidbased
onnatural vegetableoilslfots with stearic acid
C17I{35COOH or C1s compositionis minimw 95 Vo wlw. Ithas been never manufactured at all, although
oleo chemical industryhad been growth in Indonesia in
the period 1989-2004. This fatty acid had been
produced in America and Europeto consume for:one of the consumption asthe raw material of the cosmetic manufacturing that have the lowest skin effect, for
exampleto manufacture the premium shampoo Pantene.Thts stearic acidhas very low unsaturated
content.Sincethe beginning of the year 2005,South East Asia's begunto see as a V 1895 S producer. In October 2005Twin Riversas a buyer gave an idea, a standard
quality that had to be followed and produce it according to the process sequences which were
determined and do by PT. XXX alone(Figure 1), a
company that had been exist in operation inoleo
chemical industry, since the last of the year 1998. Quality of V 1895 S produced had to meet the quality standard which is shown in Table 2 below (settled by
Twin Rivers).The same quality standard which was produced in America and Europe by the method was
different at all.This was because of the different raw material and the used technology"
In this research is useda natural vegetable oil, Crude Palm Oil (CPO) as raw material for the manufacturing and refining of stearic acid super V I 895 S as shown in Figure lbelow. Manufacturing and refining technique of CPO to produce stearic acid super V 1895 S (Figure
l)is so different compare tothe general alternative
manufacturing and refining techrique of fatty acid in Figwe 2 below [12]. Figure 2 is modified to be Figure I to produce V 1895 S to apply Twin Rivers's demand
with the lowest possible it unsaturated content, as the first technique in the world. Refining of raw material on left side of Figure land Figure 2 can be done by degumming only and/or combined together with bleaching and or deodorization process alternatively
according to raw material used, product quality have to be met and equipment capability. Refining of fatty acid
on the left side of the both figures can be proceed through distillation and./or combined with fractional distillation alternatively, again it's very depend on the
I 10404-3737 IJE'I-IJENS (d Augusr 2011 I.lflNS
@
--
lntemational Journal of Engineering & Technology IJET-IJENS Vol: 11 No: 04
raw material specs, product quality have to be met and present equipment capability.
{+Re+fr-ni-n--g*CPo'.-\-* eqfiu$g
{JSplit
Sepaiation
Hvdrogenauon
Glycerin,* Sil,,
r*tty roal-r R,*** \*rri*r.,,
+ + {Fauvaud
IH1'drogenahon Refine glvcenn Refitntiunrgg
Hytuse-
+
IpfllcLfilUl
Hydrogenated fattv acic
flfr\4 {
_lDistillahon
IFrtytyrrcdid I
1850
L+ Resldue
Fnctiopation
pl"lmid.q acid
-flv 1$9J S
-|
iuci;issfrf
ll,I SFractiohation-------* \'1895
, I
+toils and
Re;-turing
fats
ra$-'m=et\e_ria-lr
Br.Eflat
r+$plit
Hvdrogenation
+S.pJrtior, Gl1'ceru:ra t{rt
r*uu e.I,al R.t*s \*.,*Luo'
+ J {Fanv ecid
Hy&ogenation Refine glycerin Refimng
\/
Figure 2. General alternative of fatty acid refining
steps
Palm oil is one of the source of stearic acid
(CrzH:aCOOH) or Crs also as a source of unsaturated
fatty acid oleic acid (CtrHgsCOOH) or C1s-1 and linoleic acid (CrzHgzCOOH) or C1s-2 that could be hydrogenated to be stearic acid as also stated by t7ltl0lll5l till the total amount of Crs is around 50 7o
w/w. See the raw material quality CPO that used in this
study. The fatty acid distillate which is obtained of
splittedCPO is called fatty acid 1850[7]. Crude Palm
Oil has the highest C1s content than the others natural
vegetable oils such as: crude palm kernel oil and
coconut oil. This's the reason the raw material CPO is
used in this research.
Table 1 .Thespecific quality of CPO to provide V 1 895 S
S.awrrai+rial A\tr SY E?O
cro ?.6 101: 0:9
cru
0.:
c14 8tr6 *13 f,!&-X C:8-! 1.1 43.l ,1.4 39.1 10.3
c]0
0.4
"u-Ilk 0.8
0fiio rrcg ." F?ora $4Xg1f4 C*emirudo,J
{-qLS-:, AV= aeidvrlue; SV =saponificationvelue;I];ftri=unkno*ncomponenUmsterial
Refining steps that are started from CPO refinery till
distillation step in Figure I above are intended to
separate the impurities of CPO (gum, trace metals),
reduce of ; pigments, odors, short carbon chain of C6 till C14 fatty acid (as the experience CPO contain trace
C6-10 in the light end product of distillation process),
aldehydes, ketones, iodine value (through
hydrogenation and distillation) to provide of fatty acid
feed quality (HCv 1895 S in Table 2 below) that
refined in the end step, in a single fractional distillation column (Figure l). Each step affect to the each others,
agreed also by the researcher [8].
Fractional distillation
In this manuscript is discussed priority the results of
this study'in the end step of the refining Hydrogenated
Crude V 1895 S (HCV 1895 S) to produce stearic acid super V 1895 S through a single fractional distillation
column in Figure 3which is equipped a packed of structured packing and falling film reboiler [6]. The
using this column rvill be obtained the better saving of
energy consumption, equipment, operating cost, flexible and product quality compared to the using of
two fractional distillation columnsthat utilized trays
and live steam (Figure 4) tl3l. This technology in
Figure 4 was mostly used in Indonesia since 1989, in
oleo chemical industries.
I 10.104--i7J7 IJDT-IJENS {!) Augast 20ll UENS
ITN
lntemational Journal of Engineering & Technology UET-UENS Vol: I I No: 04
\recuum rystem Cl6purge rslight end V 1895 S asside product
Frsn* akier of
Sottorepro*mt orresid*e
D
Figure 3. The refining of HCV 1895 S by a single fractional distillation column
Please compare this new technology in Figure 3 above
with the old one in Figure 4 below. Refining of
blended fatty acid that content of at least two main
fatty acid componentswas usually used two fractionation columns. The l't colurnn is used to
separate light end and 2od colurnn is used to separate
the top product with bottom product. Light end contain
any substances or impurities which have the lower
boiling point than boi-ling point of the main product in 2'o column (the side or bottom product). The bottom
product has the higher boiling point than the boiling
point of the side product in the same column.
Separation of the light end in I't column and bottom product in 2od column affect much better to the main
products quality.
Vacuum pressure fractional distillation is applied in this study as done for distillation because of fatty acid
included unsaturated bond inside are so sensitive for
oxidation tl3ltl4l (whichis so influence into the bed
of fatty acid color and odor). This is also according to the reseachers tl-3lt9l. Unsaturated bond in fatty acid is more sensitive. Ifunsaturated bond is higher in fatty acid,it's easier to oxidize [10]. Prevention of fatty acid oxidationis mean to prevent the increment of fatty acid iodine value or unsaturated content, colors and odors. It is mean also control closely fatty acid quality which is
made" In this research wasV 1895 S quality" A
researcher explained in his disertasion: reduction of bottom temperature of column can reduce stearic acid
iodine value or unsaturated content or vice versa [18].
Fractional distillation is shown in Figure 3 above is intended to reduce unsaturated content of stearic acid C16" Separation of oleic acid Cls-1 in the residue of single fractional distillation column will reduce oleic acid (C1s.1) content in stearic acid Clsas a side product" It is mean the lower color and odors in stearic acid Clsor V 1895 S. Reduction of fatty acid iodine value,
color and odors is not only can be done by
hydrogenation priority, but can be done also through
fractional distillation process to achieve the lower
point. How it can be done?
Tab!* fu Quality :*aadsrd o f *gg{i$ ecid. s up e* V I B 9 5 $
FattS' o*id comy o s itioa,
$i'+-Jw) < ctr4
clf
c:6-t
c1?
c!8 cl8-I C}8.]
C:O
cI0-l
Arr]lrditr'rane*
0;5 lorrimus I0.0 msriuurr
9fi.{ luinillu.m
s,44 artlmuu
Figure 4. The refining of blended fatty acid through 2 fractionation columns
fiqgs6; flo I o fi fir[ilfl $:andu da, J 005
.[ehlCJ" Quality of feed ]ICV 1E95 S
Met*iaX(ltl C14 CI6 q:8 Clg-l C30 Seh Remark* Quality rtatus Status
-$qur*E; fforo $flslIs C*aml *da,261 5
Unsaturated content of HCV 1895 S is higher than the maximum standard (see Table 3 above), because ofit value is over than O.l5 Vo. So it has to be reduced till meet the standard, Reduction of C1s-lamount in C1s is also decrease of arachidic acid (C26) composition and the others impurities that have the same or higher than
stearic acid C18 boiling point which is so positive influence to the V 1895 S iodine value, unsaturated content, color and odors, to be lower. This will be influence to the lower production yield of V 1895 S
absolutely, but hopefully with the better quality V 1895
Il0il04-3737 UU1-IJENS (qi Auglst201I UENS
ME
lntemational Journal of Engineering & Technology IJET-IJENS Vo1: 1l No: 04
S than HCV 1895 S will be achieved' The above
question is answered'
Operating conditions of single fractional distillati,on co'lumn ias tobe determined and predict carefully
bstheeafnodsriettiobvyeieRtooautmhlte'asinoLxacidowanst[i5iod]ne[1raa6tn]io.dnOhileenaictthinaisgci'dsItnuccdoryne'tmeIntetni.sCt lsoso-f unsaturated content or iodine value should be
protected, should be lower in V 1895 S' The prediction
Lf the single fractional distillation column operating
conditionsis done base on the capacity and column
characters (the pressure drop on the top, bottom.and
feed tray location of the column),
HreCsivduferiSpSroSd, upirtodcoucmt pcoomsitpioosnitiiosn
feed compostuon
of V 1895 S and
predicted'-.{ctual
application of this prediction will change a little bit
depend on the actual product quality is made' The
2" Method Semi trial and error method is used in this researctr'
Operating conditions are determined, predict first-and
uubap'"sploetrydOtloongi-*itt"o.qttuhVealia1tyc8t9us5atalSntdesatporedlctts(p' VrTioh1rei8tg9y5laitlSCclionsmCop.biostasiifntoieondne''
changing
;;ft"?
have to be predicted and make better
to the quality standard' The understanding of
fru.iionuf distillation, physical and chemical properties
oi th" tobrtur.e, e*pJti"nc" in t"urr-ftur" the big conffibution
this to
process and filling do the changing of
tfr"
operaatrinJgdectoenrmdiintieodnsantdl{p' redTichte1-".
operati^ng
for 807o
"porroOJlr,"iorin",t yield of V 1895 S according to the specs- of
V rasS S composition to get in specs and the higher reflux rate in ihe single column, according to C13-1
ptop.tti"t
it.'trigt
that so reflux
sensitive rate has
to theoxidation and
to be made as it
heating' value is
;;hi";."d on the actual 807o production leld of V 1895
S or higher, that's limited by *rut*u',"4'"ontent, colorar;
Vach1i8e9v5ed-Socnomthpeosatctitouna'l
operating conditions are applied., to^ get the higher
yield. What are the suitable conditions?
unsaturated content) and production yield' See Figure 5 b"lo,". Proc"dure of anaiysis refers to AOCS Ce l3e-
Oiana fg l-64 to do iodine value analysis of fatty acid
eO6S 1-62 to do fatty acid composition analysis "t4f.t."
Check tl.e troruogenous feed tank sample; GC, !\r and rclor
Ttre rtain oPer=iing cenditia* are detemine* and Fredict ba3€d on tl.e ccmPo:iticn (GC) of feed. pu.ge, side and bottamProdircts
*e Pr*dicted
Appliad tlle main caemting co*&trs** t* *.:in5!e tellmr::o:erd remallised om 8O 96 cf Y 1895 S
gm&xt*teis de.ters*red
Dislillate a' 189f $ PuitY conttol
Product:V1895S Ilurit]' : 90 i'b minirrum C:s-t : fi,15 0,6maximum Crc : 1O -qio maxirctr'
Reduce side stream &aw off rste
Stored as ttrc in rpe* V 1895 S in fhe fiael
Figor. s=u.ti;;uto# "or
vand quality 1895 s as it standard
110104--173? IJI]llJtjNS (O August 20ll IIENS
UN
Intemational Journal of Engineering & Technology IJET-UENS vol: I I No: 04
3. Results and Discussions
uT-Vr_r1P89p5
,l intended to separate
Sandfarty acid residue
stearicacid (C1s) C1s_2s of the feed
super
UiV
1895 S to reduce unsaturated content of V lg95 S.
lhere is no light end draw off in this study because of
color (not shown here), C16 and C1s of HCV 1g95 S is
in specs. The final quality of V 1895 S is shown in
Table 4 below. The off specs one V 1895 S in the final
storage tank is caused of by production yield increment over than 87 Vo (see Table 5 and Figure 7 below).
I+h}fl.*" Unsaturated content o f HCV I I 9 5 S and. \r 1 895 S (96:*rsr)
1S
\r1s95 S
9S,06
0.s9 1"12 !S
.Qgs.r.;5g; f'lora $mCr{s S&fil{n{dg, .?00i
The results of, this research show that unsaturated
content in this step came down to beaverage of 0.19
Vo(based on Twin River's analysis)
\/ 1895 S came down ro be average
or of
iodine value of 0.12 gr l2fi00
g. Comparison of unsaturated content Clg-l of HCV 1895 S and V 1895 S can be seen in Table 4 above"
Actual color of V 1895 S is in specs, but not shown in
Table 4 above, discussionis centered to the unsaturated
trB-t >ff, cl8-1 >
tr$-1. C30 are C18-l is
.lPees
Feed
content which is off specs. See inTable 3 and Table 4
above.
The operating conditions are determined and predict, can bee seen in Figure 6 below.This prediction is made
base on the fatty acid compositiononly. please,
compare this one with the actual operating conditions in Table 6, in discussion 3.3.
iEl-l rrra,
eoffdsrr:rer Li_Eht Eud
lnol
rs'ei-e]rt 1',) e
35f
184
c14 0.{0
cx6 ct E
]..05 0.4_i
total 1,,50
C,.0S
76.S0 3fl.08
to0.0s
F€€d" HgV 1895 S
CI4 O, cl6 ?",6 fl.?6 cI I 980.5 98.S9 ci&l ].-l 0-11 c:0 9,: 0.9? total 999.7 l{0.0
34.r-r45 i$
iliii iir* *t*
:63-X64:9.
ri.r l895 S ?i,f*ad]
cI6 cl8
6_55 s.8 x9
?!1.}0 98.S4I
ccl380-r
ff.33 *.53
s.s4 o,30
totstr ?99.?6 I00.0tl
cls o
s.g5
c!3 18s.86 95.03
ct8-1 I.9S x.ffo
ct0 ?.6$ 3_ E3 tot,sl EgS.4O IflS.8
Figure 6.The operating conditions design forthe providing ofHCV 1895 S by a single fractional distillationcolumn
Il{)40.+-3737 tJEl,ltUNS @ Augusr20II IJENS
lrlilIm
lnternational Joumal of Engineering & Technology UET-IJENS Vol: 11 No: 04
10
3.1. How to minimize unsaturated content in V1895 S?
Based on Table 5 and Figure 7 below no doubt at all
that Cl8 purityis over than 95 7o, althoughv 1895 S production yield are (80.0 - 95'0 %). This is caused by
bt6 feed is lower than I Va (Table 4 above).If C16
amount in the feed HCV 1895 S is kept as is in V 1895 S feed and no light end draw offor total reflux in this system./study, ClS purity will not be lower than
90 7o wlw. This also affect to the bigger C16
concentration in V 1895 S that can minimize unsaturated (Cl8-1) content in V 1895 S, the others
components too, C20 and unknown (Unk) (in Table 5)'
The higher S production yield of V 1895 affect to the
smaller C16 content and the higher C20 and unknown
relatively. It affected to the off specs unsaturated content in V 1895 S. It can't be allowed. Please, see also both figures, Figure 7 and Figure 8 below'
s*foh+Eff.ectof,*B*if,:oati:omgreductitr:,xi€K*]s#wr,ra{a.ratesccnlerr1oflxlEEs$
s7 In *pecs
50,3 99,.13 0"04 0,0: 0,16 0,11 111:
51.1 9?,58 CIJ6 oJ3 0,f s
0s-1.0
4 fl"39 1,Xl 97
0.11 6.8-0-9
8?"0
s3$
95
3rr specs
$ff sprcr off
Sawqe: F/ara Sqrtt'ifs C]e*ind+. lS$J
100 99"J
99
fl ss.:
:E99
3 91"5
@
96.J 9d
95"5
91.J8)
*o rbditryfifd tJ lti [!gJE $ii
Figure 7. The influence ofV 1895 S production yield on C18 purity in V 1895 S
Separation of amount C18 into fatty acid residue (see Figure 5 above) affect to the lower amount ofV 1895 S.
Reduction the production yield of V 1895 S. This is
cause of C16 amount ratio to the total amount of side
product V 1895 S is bigger. Finally although without iight end draw off, color of V 1895 S met the quality
standard (not shown in discussion). So the decision not to draw off light end is the exact consideration in the beginning of this study, one technique to minimize the unsaturated content or not reach 0.15 Vo. The ratio can
be changed depend on the changing of V 1895 S
production yield. So in this study reduction of V 1895
S production yield is also the other technique to control
or reduce unsaturated (C18-1) content as it specs (see
Figure 8).
The influence of the reduction of V 1895 S production yield on C18-l content in V 1895 S and reflux rate in Figure 7 can be used as the evidence to support to
state; reduction of V 1895 S production yield will reduce unsaturated content in the main product'
Reduction of V 1895 S production yield is cause of the increasing of reflux rate which reduced the BT of fractionation column (in Figure 9 at discussion 3.4)" The same thing that had been done by a researcher to
reduce stearic acid iodine value or unsaturated content in stearic acid distillate; by reducing BT of distillation
column [18]. The reduction of BT affect to the
reduction of evaporation rate of unsaturated content,
reduce it amount into V 1895 S, of course also the
other components. This is according to the distillation
principles t5ltl llt17l.
1.8
? 1.6
c
cl
> 0.8
: 0.6
E 0.4
*.l* Production yield vs Reflux rale ---*- Production yield vs C18-l conrcnt
(95;0,38)
UO
-0.2 p.od,iJtion yi"ro ,li$ rsss s i&i)
Figure 8. The influence of V 1895 S production yield on C18-lcontent in V 1895 S and reflux rate
I I 0 104--1737 Ll U I''lJ IjNS (cl,\u gu sl l0 I I IJ E^-S
@
International Journal of Engineering & Technology IJET-IJENS Vol: I 1 No: 04
11
-1,i. Prediction versus actual V 1895 S production yield
; ::-:uon of V 18955 unsaturated content was done in
,: . ,:.rdy.Theproduction yield of V 1895 S is tried first
r. '-: asthe design and prediction of the operating
. ,::::rons(Figure 5 above). ReJlux rate is also made so
.:-, rgh in the beginning, on the intervall5OO - 1600 .; -,- This reflux rate is intended to suppress the : .:,:iation rate of unsaturated component as low as
:..,::le because of the boiling point of unsaturated
. -. .cid Cls-1is lower than C1s boiling point[16]. This l-r-: rate is made also because of the amount of C1s-1
: ---: feed HCV 1895 S is so small (0.23 7o in Table 4
-:,-,:r. also the other components (Table 4). On the
:--l'::,rnretofluthxerabteo,ttoitm's
more difficult to be separated product or residue (Figure 5
-;.,r3). Of course the amount of Cra will be bigger in --. ::sidue. It's a risk of a high reflux rate. The feed --.:,s ried first onl.000 kg/hr. These predictions are
.:: according to the experience in the fatty acid ' .:..ation and fractionation in the previous"
'-. :roduction yield of V 1895 S is then increased
:ii-)'(as shown in Table 5 above) after fractional
- .:-iation conditions werestable for 80.0 7o
r-:,:jction yield. This simulation is done for knowing
--. .;rfluence of the increment of V 1895 S on the -::.ment of unsaturated content in V 1895 S. The
':, -.:s rvithin Table 5 proved the prediction above. See
.. Figure 8 below.The higher possible production ..: can be achieved but at the same time unsaturated
- t:,3nt is closer to the maximum point of unsaturated
: --,ent and finally of specs on 9l Vo,93 % and 95 Vo :: -,:.rction yield (Table 5 above and Figure 8
:t.:.r').The in specs one ofV 1895 S should be made in
,- : research. How many percentthe maximum :::'r.rction yield has to be achieved? How many
:r::ent the safest production yield? It has to be met.
-:: rend in this Figure 9 can be used to predict the , --"ble production yield of V 1895 S to provide
The off specs one stearic acid V 1895 S in Table 5
above is sent back the other feed tank as described in Figure 5 above to avoid the higher unsaturated content
in the same feed tank and in the next V 1895 S will be produced.Based on data in this table, it is so clear the influence of the higher production yield V 1895 S to
the higher it unsaturated content. See Figure 8 below,
Base on this figure, V 1895 S production yield is predicted 89.5 Vo to achieve O.l5 7o unsaturated content in V 1895 S (point A). So the safest
production yield should be under 89.5 Vo. Based on this
predictionv 1895 S production yield should be
adjusted to be lower than 89.5 % (see Figure 9 below).
It could be 88 Vo or 89 Vo. Whyit can be happened? It has to be considered strongly the unsaturated content should be achieved first lower than 0.15 Vo. lt is very
important to meet the specs first. If the production
yield is kept constantly as is (89.5 Vo), therc is a big possibility that V 1895 S will be off specs.
Beside this reason the stability of this process is considered strongly. It is not guaranteed all the instru:mentations have the straight line response,
according to the experience it's normally has the small fluctuate on the same amplitude.
0.45
3 0.1 ; 0 35
6 o.r
6 0?5 6 ^i n r{
: 0.r 5 o.o5
,5,
B (91, 0.19 !
a{s9.i,0.ri)
80 *5 90 9'
lro&rtionyi{d of Y lS95 S G6)
t00
.:. rodine value and unsaturated content of V 1895 S
-- Table 5 above) came up to be higher. This is cause
:. *nsaturated component is evaporated more together
, :. \/ 1895 S as the impact of the increment of V -.:: S production yield and/or decreasing of reflux
:r,: see Figure 8 above). The composition of C20 is : :-ler also, caused of the same reasons. This is
.: - :rding to the distillation principles t5l t I 1l I171.
Figure 9. The influence of V 1895 S production yield on unsaturated content in V 1895 S
The actual average production yield is achieved 85.0 Vo. Why is it different? The production yields in Table 5 are not straight since the increment is done, It is done slowly, step by step.It so logic the actual production yield is lower than the predicted one.
i.3. Actual versus predicted operating conditions
-:.: reduction of V 1895 S production yield mean the
-: :3mperature of the fractionation column is adjusted .-.rer. This temperature could be determined and ::::rct practicallyby Roult's Law but have to be : =l-red actually through this research. The predicted
one is used as a main guide and very important to
conduct this research. This prediction was limited by
unsaturated content or double bond of C13-1 in the
quality standard. If the consideration was based only of C18 purity (see Table 3 above), V 1895 S can be draw
I 10.104'3737 lJIll'-lJIlNS @ Augusl 20ll IJENS
@
lnternirtional Journal of Engineering & Technology IJET-IJENS Vol: I I No: 04
t2
off 98 Vo or it production yield 98 70 as it composition
in the feed but actually unsaturated content is off specs when it production yield achieved 93 7o and95 Vo. So
it must belower because of timited by unsaturated content in the quality standard. It's mean also the
actual operating conditions are different with the
predicted operating conditions. See Table 6 until Table
9 below.
simulation data on the same yields from Table 7 below. The data for 93 7o and 95 Vo production yield didn't
take more because of the conditions were up set
already.
The average conditions for the production yields 88 7o, 89 Vo in Table 10 were the average of the simulation data on the same yields from Table 11 below.
The average conditions for the production yields 80 7o,
87 Vo and 93 Va (Table 6) were the average of the
Table 6. The actual operating conditions to produce V 1895 S of l't test
IJroductron TP" mbar TT. { C SPTT. FT, o C BT, o C RR, U:t& E}, mbar
Remarks
v'ipld l9,ir\
o fl
80.0 19.3 .I3X.S l4j{.4 231"4 ?66"6 1.1-1.1 50-J4 Cg-1 0"15 o]$
temperstue; ET:
Table 7. The actual simulation ofthe operating conditions to produce V 1895 S of l't test
IPryrooided8uutcl0cft"i.o.0.nt)TPl,9,.m.Jbar
TT,DDCC t32_S
SPTT,
oc
:44.4
FII,,{{CC
:3?.4
BT,sCC }R-R,E*rii,hh BpP,m,mhh4rar
256"6 1.1-11 5g:54
s0-0 80,0
87,0 s7.0 B?"0
19"5 19-1 25.8 21"6 ?6.0
]]J -0 23t.6
344,2 :.{.1.6
13?.6 ?3?3
?67.0 266.2
1-1-1.: 1.1-1.19
51_54 49-54
?14.5 f,43.6 :36.S 268"0 i.g9_1"1 5S-53
X14-7 214-3
?44.0 t4t.I
J37.0 236.6
?68..+ 1.0S-1.11 ?67.6 1.09,1.1
51_51 49-52
93,0 93,0
:?.: 27.25
203.$ 145.1 :38"9 ?03"7 :44.9 :38.9
2?4-1 0.8.1_0.9 50_51 ??3.9 0.8$_0"39 50-51
9t"0 ts.l :04-5 J44-5 :38"9 ]?i
.83 50-51
Remzrks
Cts-l< 015 0/6 c.8-1{O-t5 9d c.8-1{8-15 9/a cB-:{ *-15 9t *8-:{0-1"5 *6 c*-14 0.1,J % c8-1> 0-15 $6 cg-1> 0"15 BL
tr8-1> 0.15 }t
Soruce: SIora SqU*;g SftS4i*f{k 3
I'{ote: IP - top prersurs; TT: tcp t€nlperaturB; SPTT = side pro'duct ua}'trr.$erertue; r-I: feed iemperaft$e; ET:
Itltgm terRperature ; RR = r,eflux ret* ; BF = boitorn pre*:,r:re
The actual operating conditions on 80.0 7o production yield has a small different than the predicted one (see Figure 5). The actual suitable prediction of production
yield was 89.5 Vo (discussion 3.2 above). It's so
different also than the prediction was done only base on the C18 composition in the feed. Feed temperature can't be reached because of the limitation Oil Thermal
Heater system capacity (on 850-900 kgs/trour feed rate), to conduct heating in a heater and a reboiler of
fractionation column.
Reducing of TT and SPTT trends that almost constant are according to the simulation of Roult's Law (not
shown here) I5l. SPTT trend as is due to Cl8
concentration is almost constant around 98 Vo on the same pressure 27 mbat. Reducing of TT is due to increasing Cl6 or reducing C18 composition in the
total reflux of light end on the same pressure, around
25 mbar, as affect of V 1895 S production yield increment or decreasing of total reflux into the column, This increment is cause of reducing C15
composition in V 1895 S that is added major from light
end and also C18 composition increment that is added
major from bottom product. This increment is also
cause of the increment of BT that is effect to the higher
C18 evaporation so it composition is higher at the production yield 87 Vo (see Figure 8 in the next
discussion 3.4). [ncrement of production yield (priority
93 Vo and 95 Vo) is cause of the increment of BT that is
so effect to the higher content of Cl8-1, C20 and
unknown components, unsaturated content is off specs (0.36 and 0.38 Vo). This is also so affect to the lower
C18 content in V 1895 S. That is why both the last 2
actual operating conditions in Table 6 can't be applied
110.10.1-37.17 IJET'IJENS (E Aueus. 201 I UtjNS
@
International Journal of Engineering & Technology UET-IJENS
11 No: 04
l3
,he next test, but the higher yield should be pointed
., shown as discussion 3.2, based on Figure 7. The : ,"iuction yield 88 and 89 Vo are met graphically and
the operating conditions have to be predicted by Roult's Law, next has to be tested actually, and
compared how precise it.
- ":le 8. The predicted operating conditions to produce V lg95 S of l,ttest
r{d*iolTn)
ss.0
TP, mbar
2==3.3. 6
TT, o C
2, 1,7-
SPTT,
oc
?44
118 ?45
BT,{C RR, ?47 264
, mbar R
9t (beee on ligure 9 of
:l1l&?- ?14445- ?s3-89.0 ?3 J6
246_ 24"7
I st tes$
:.01-1.s.1+
50*
{C8-1= 0.14 +6 CI"15
264 flo, (base on Frgule I
\ote:T}=tgpPr*sure;TT:toFtempi[etue;S}T:=*i*uperEtue;BT=
cflsttest
hSffnS tenper*true ; RR = reflti:r rate ; EP = bottompre:ssire ; ; base on the highest agtual 1.ecu11111Irregflrre
l:ble 9" The predicted of V I 895 S quality on l,t tesr
-i c:6
18-1 C?S
Reflux rate,
ductio* yield
s
. ? 0,76 99.6? 0,06 0"32 0.30* - 1.0I-1.04+ 89$
In specs
Table 10. The actual operating conditions to produce V 1895 S of 2nd test
Production TP,mbar TT,oC SPTT, m rield (Va) 88.0 25.8 214.5 243.6 236.9 268.2 LOg_Lts
5A-52
Remarks C8-1= 0 < 0.15 7p (base
89.0
23.2 229.5 242.7 242.5 265.5 1.03-1.1
50-52
on Table 12) CS-l= 0.02
PROVIDING VEGETABLE STEARIC ACID SUPER V 1895 S FROIVI HYDROGENATED OF CRUDE VEGETABLE STEARIC ACID HCV 1895 S
WITH A SINGLE FRACTIONAL DISTILI-ATION COLUMN
Muhammad yusuf. Ritongal
l,Chemical Engineering Department, Faculty of Engineering, University of Sumatera Utara, Jl. Almamater Kampus USU, Medan 20155, Indonesia
E-mail : yu s ufrit @ gmail. com
Abstract
ln 2005 stearic acid V 1895 S is manufactured for l*ttime in Indonesia base on the natural vegetable oil, Crude palm Cil to apply Twin Rivers's demand. One of it specific quality (iodine value 0.15 gll00 g maximum) is very influence to ,re purifying steps, one of the steps is purifying of hydrogenated crude V 1895 S (HCV 1895 S) feed by a single
:olumn of fractional distillation to produce V 1895 S with the lowest possible of it iodine value, lower than the
naximum value. The results of application and small adjustments of determination operating conditions of the single ;olumn priority based on the feed composition, product composition of V 1895 S and it pressure drop character are ;ompared based on it quality standards. This is the method used in this research to provide V 1895 S and conducted in ihe plant scale of PT. XXX factory with production yield 88 - 89 Vo of 0.90mt HCV 1895 S, unsarurated content is ioiver than 0"15 g/100 g and higher than 98 7o wtw Cl8 purity. In general the specific quality standards of V 1895 S
can be met"
Keywords: quality standards, fractionation, composition, unsaturated contenl, adjustments
1. Introduction
Fany acid ofV 1895 Sis meansuper fatty acidbased
onnatural vegetableoilslfots with stearic acid
C17I{35COOH or C1s compositionis minimw 95 Vo wlw. Ithas been never manufactured at all, although
oleo chemical industryhad been growth in Indonesia in
the period 1989-2004. This fatty acid had been
produced in America and Europeto consume for:one of the consumption asthe raw material of the cosmetic manufacturing that have the lowest skin effect, for
exampleto manufacture the premium shampoo Pantene.Thts stearic acidhas very low unsaturated
content.Sincethe beginning of the year 2005,South East Asia's begunto see as a V 1895 S producer. In October 2005Twin Riversas a buyer gave an idea, a standard
quality that had to be followed and produce it according to the process sequences which were
determined and do by PT. XXX alone(Figure 1), a
company that had been exist in operation inoleo
chemical industry, since the last of the year 1998. Quality of V 1895 S produced had to meet the quality standard which is shown in Table 2 below (settled by
Twin Rivers).The same quality standard which was produced in America and Europe by the method was
different at all.This was because of the different raw material and the used technology"
In this research is useda natural vegetable oil, Crude Palm Oil (CPO) as raw material for the manufacturing and refining of stearic acid super V I 895 S as shown in Figure lbelow. Manufacturing and refining technique of CPO to produce stearic acid super V 1895 S (Figure
l)is so different compare tothe general alternative
manufacturing and refining techrique of fatty acid in Figwe 2 below [12]. Figure 2 is modified to be Figure I to produce V 1895 S to apply Twin Rivers's demand
with the lowest possible it unsaturated content, as the first technique in the world. Refining of raw material on left side of Figure land Figure 2 can be done by degumming only and/or combined together with bleaching and or deodorization process alternatively
according to raw material used, product quality have to be met and equipment capability. Refining of fatty acid
on the left side of the both figures can be proceed through distillation and./or combined with fractional distillation alternatively, again it's very depend on the
I 10404-3737 IJE'I-IJENS (d Augusr 2011 I.lflNS
@
--
lntemational Journal of Engineering & Technology IJET-IJENS Vol: 11 No: 04
raw material specs, product quality have to be met and present equipment capability.
{+Re+fr-ni-n--g*CPo'.-\-* eqfiu$g
{JSplit
Sepaiation
Hvdrogenauon
Glycerin,* Sil,,
r*tty roal-r R,*** \*rri*r.,,
+ + {Fauvaud
IH1'drogenahon Refine glvcenn Refitntiunrgg
Hytuse-
+
IpfllcLfilUl
Hydrogenated fattv acic
flfr\4 {
_lDistillahon
IFrtytyrrcdid I
1850
L+ Resldue
Fnctiopation
pl"lmid.q acid
-flv 1$9J S
-|
iuci;issfrf
ll,I SFractiohation-------* \'1895
, I
+toils and
Re;-turing
fats
ra$-'m=et\e_ria-lr
Br.Eflat
r+$plit
Hvdrogenation
+S.pJrtior, Gl1'ceru:ra t{rt
r*uu e.I,al R.t*s \*.,*Luo'
+ J {Fanv ecid
Hy&ogenation Refine glycerin Refimng
\/
Figure 2. General alternative of fatty acid refining
steps
Palm oil is one of the source of stearic acid
(CrzH:aCOOH) or Crs also as a source of unsaturated
fatty acid oleic acid (CtrHgsCOOH) or C1s-1 and linoleic acid (CrzHgzCOOH) or C1s-2 that could be hydrogenated to be stearic acid as also stated by t7ltl0lll5l till the total amount of Crs is around 50 7o
w/w. See the raw material quality CPO that used in this
study. The fatty acid distillate which is obtained of
splittedCPO is called fatty acid 1850[7]. Crude Palm
Oil has the highest C1s content than the others natural
vegetable oils such as: crude palm kernel oil and
coconut oil. This's the reason the raw material CPO is
used in this research.
Table 1 .Thespecific quality of CPO to provide V 1 895 S
S.awrrai+rial A\tr SY E?O
cro ?.6 101: 0:9
cru
0.:
c14 8tr6 *13 f,!&-X C:8-! 1.1 43.l ,1.4 39.1 10.3
c]0
0.4
"u-Ilk 0.8
0fiio rrcg ." F?ora $4Xg1f4 C*emirudo,J
{-qLS-:, AV= aeidvrlue; SV =saponificationvelue;I];ftri=unkno*ncomponenUmsterial
Refining steps that are started from CPO refinery till
distillation step in Figure I above are intended to
separate the impurities of CPO (gum, trace metals),
reduce of ; pigments, odors, short carbon chain of C6 till C14 fatty acid (as the experience CPO contain trace
C6-10 in the light end product of distillation process),
aldehydes, ketones, iodine value (through
hydrogenation and distillation) to provide of fatty acid
feed quality (HCv 1895 S in Table 2 below) that
refined in the end step, in a single fractional distillation column (Figure l). Each step affect to the each others,
agreed also by the researcher [8].
Fractional distillation
In this manuscript is discussed priority the results of
this study'in the end step of the refining Hydrogenated
Crude V 1895 S (HCV 1895 S) to produce stearic acid super V 1895 S through a single fractional distillation
column in Figure 3which is equipped a packed of structured packing and falling film reboiler [6]. The
using this column rvill be obtained the better saving of
energy consumption, equipment, operating cost, flexible and product quality compared to the using of
two fractional distillation columnsthat utilized trays
and live steam (Figure 4) tl3l. This technology in
Figure 4 was mostly used in Indonesia since 1989, in
oleo chemical industries.
I 10.104--i7J7 IJDT-IJENS {!) Augast 20ll UENS
ITN
lntemational Journal of Engineering & Technology UET-UENS Vol: I I No: 04
\recuum rystem Cl6purge rslight end V 1895 S asside product
Frsn* akier of
Sottorepro*mt orresid*e
D
Figure 3. The refining of HCV 1895 S by a single fractional distillation column
Please compare this new technology in Figure 3 above
with the old one in Figure 4 below. Refining of
blended fatty acid that content of at least two main
fatty acid componentswas usually used two fractionation columns. The l't colurnn is used to
separate light end and 2od colurnn is used to separate
the top product with bottom product. Light end contain
any substances or impurities which have the lower
boiling point than boi-ling point of the main product in 2'o column (the side or bottom product). The bottom
product has the higher boiling point than the boiling
point of the side product in the same column.
Separation of the light end in I't column and bottom product in 2od column affect much better to the main
products quality.
Vacuum pressure fractional distillation is applied in this study as done for distillation because of fatty acid
included unsaturated bond inside are so sensitive for
oxidation tl3ltl4l (whichis so influence into the bed
of fatty acid color and odor). This is also according to the reseachers tl-3lt9l. Unsaturated bond in fatty acid is more sensitive. Ifunsaturated bond is higher in fatty acid,it's easier to oxidize [10]. Prevention of fatty acid oxidationis mean to prevent the increment of fatty acid iodine value or unsaturated content, colors and odors. It is mean also control closely fatty acid quality which is
made" In this research wasV 1895 S quality" A
researcher explained in his disertasion: reduction of bottom temperature of column can reduce stearic acid
iodine value or unsaturated content or vice versa [18].
Fractional distillation is shown in Figure 3 above is intended to reduce unsaturated content of stearic acid C16" Separation of oleic acid Cls-1 in the residue of single fractional distillation column will reduce oleic acid (C1s.1) content in stearic acid Clsas a side product" It is mean the lower color and odors in stearic acid Clsor V 1895 S. Reduction of fatty acid iodine value,
color and odors is not only can be done by
hydrogenation priority, but can be done also through
fractional distillation process to achieve the lower
point. How it can be done?
Tab!* fu Quality :*aadsrd o f *gg{i$ ecid. s up e* V I B 9 5 $
FattS' o*id comy o s itioa,
$i'+-Jw) < ctr4
clf
c:6-t
c1?
c!8 cl8-I C}8.]
C:O
cI0-l
Arr]lrditr'rane*
0;5 lorrimus I0.0 msriuurr
9fi.{ luinillu.m
s,44 artlmuu
Figure 4. The refining of blended fatty acid through 2 fractionation columns
fiqgs6; flo I o fi fir[ilfl $:andu da, J 005
.[ehlCJ" Quality of feed ]ICV 1E95 S
Met*iaX(ltl C14 CI6 q:8 Clg-l C30 Seh Remark* Quality rtatus Status
-$qur*E; fforo $flslIs C*aml *da,261 5
Unsaturated content of HCV 1895 S is higher than the maximum standard (see Table 3 above), because ofit value is over than O.l5 Vo. So it has to be reduced till meet the standard, Reduction of C1s-lamount in C1s is also decrease of arachidic acid (C26) composition and the others impurities that have the same or higher than
stearic acid C18 boiling point which is so positive influence to the V 1895 S iodine value, unsaturated content, color and odors, to be lower. This will be influence to the lower production yield of V 1895 S
absolutely, but hopefully with the better quality V 1895
Il0il04-3737 UU1-IJENS (qi Auglst201I UENS
ME
lntemational Journal of Engineering & Technology IJET-IJENS Vo1: 1l No: 04
S than HCV 1895 S will be achieved' The above
question is answered'
Operating conditions of single fractional distillati,on co'lumn ias tobe determined and predict carefully
bstheeafnodsriettiobvyeieRtooautmhlte'asinoLxacidowanst[i5iod]ne[1raa6tn]io.dnOhileenaictthinaisgci'dsItnuccdoryne'tmeIntetni.sCt lsoso-f unsaturated content or iodine value should be
protected, should be lower in V 1895 S' The prediction
Lf the single fractional distillation column operating
conditionsis done base on the capacity and column
characters (the pressure drop on the top, bottom.and
feed tray location of the column),
HreCsivduferiSpSroSd, upirtodcoucmt pcoomsitpioosnitiiosn
feed compostuon
of V 1895 S and
predicted'-.{ctual
application of this prediction will change a little bit
depend on the actual product quality is made' The
2" Method Semi trial and error method is used in this researctr'
Operating conditions are determined, predict first-and
uubap'"sploetrydOtloongi-*itt"o.qttuhVealia1tyc8t9us5atalSntdesatporedlctts(p' VrTioh1rei8tg9y5laitlSCclionsmCop.biostasiifntoieondne''
changing
;;ft"?
have to be predicted and make better
to the quality standard' The understanding of
fru.iionuf distillation, physical and chemical properties
oi th" tobrtur.e, e*pJti"nc" in t"urr-ftur" the big conffibution
this to
process and filling do the changing of
tfr"
operaatrinJgdectoenrmdiintieodnsantdl{p' redTichte1-".
operati^ng
for 807o
"porroOJlr,"iorin",t yield of V 1895 S according to the specs- of
V rasS S composition to get in specs and the higher reflux rate in ihe single column, according to C13-1
ptop.tti"t
it.'trigt
that so reflux
sensitive rate has
to theoxidation and
to be made as it
heating' value is
;;hi";."d on the actual 807o production leld of V 1895
S or higher, that's limited by *rut*u',"4'"ontent, colorar;
Vach1i8e9v5ed-Socnomthpeosatctitouna'l
operating conditions are applied., to^ get the higher
yield. What are the suitable conditions?
unsaturated content) and production yield' See Figure 5 b"lo,". Proc"dure of anaiysis refers to AOCS Ce l3e-
Oiana fg l-64 to do iodine value analysis of fatty acid
eO6S 1-62 to do fatty acid composition analysis "t4f.t."
Check tl.e troruogenous feed tank sample; GC, !\r and rclor
Ttre rtain oPer=iing cenditia* are detemine* and Fredict ba3€d on tl.e ccmPo:iticn (GC) of feed. pu.ge, side and bottamProdircts
*e Pr*dicted
Appliad tlle main caemting co*&trs** t* *.:in5!e tellmr::o:erd remallised om 8O 96 cf Y 1895 S
gm&xt*teis de.ters*red
Dislillate a' 189f $ PuitY conttol
Product:V1895S Ilurit]' : 90 i'b minirrum C:s-t : fi,15 0,6maximum Crc : 1O -qio maxirctr'
Reduce side stream &aw off rste
Stored as ttrc in rpe* V 1895 S in fhe fiael
Figor. s=u.ti;;uto# "or
vand quality 1895 s as it standard
110104--173? IJI]llJtjNS (O August 20ll IIENS
UN
Intemational Journal of Engineering & Technology IJET-UENS vol: I I No: 04
3. Results and Discussions
uT-Vr_r1P89p5
,l intended to separate
Sandfarty acid residue
stearicacid (C1s) C1s_2s of the feed
super
UiV
1895 S to reduce unsaturated content of V lg95 S.
lhere is no light end draw off in this study because of
color (not shown here), C16 and C1s of HCV 1g95 S is
in specs. The final quality of V 1895 S is shown in
Table 4 below. The off specs one V 1895 S in the final
storage tank is caused of by production yield increment over than 87 Vo (see Table 5 and Figure 7 below).
I+h}fl.*" Unsaturated content o f HCV I I 9 5 S and. \r 1 895 S (96:*rsr)
1S
\r1s95 S
9S,06
0.s9 1"12 !S
.Qgs.r.;5g; f'lora $mCr{s S&fil{n{dg, .?00i
The results of, this research show that unsaturated
content in this step came down to beaverage of 0.19
Vo(based on Twin River's analysis)
\/ 1895 S came down ro be average
or of
iodine value of 0.12 gr l2fi00
g. Comparison of unsaturated content Clg-l of HCV 1895 S and V 1895 S can be seen in Table 4 above"
Actual color of V 1895 S is in specs, but not shown in
Table 4 above, discussionis centered to the unsaturated
trB-t >ff, cl8-1 >
tr$-1. C30 are C18-l is
.lPees
Feed
content which is off specs. See inTable 3 and Table 4
above.
The operating conditions are determined and predict, can bee seen in Figure 6 below.This prediction is made
base on the fatty acid compositiononly. please,
compare this one with the actual operating conditions in Table 6, in discussion 3.3.
iEl-l rrra,
eoffdsrr:rer Li_Eht Eud
lnol
rs'ei-e]rt 1',) e
35f
184
c14 0.{0
cx6 ct E
]..05 0.4_i
total 1,,50
C,.0S
76.S0 3fl.08
to0.0s
F€€d" HgV 1895 S
CI4 O, cl6 ?",6 fl.?6 cI I 980.5 98.S9 ci&l ].-l 0-11 c:0 9,: 0.9? total 999.7 l{0.0
34.r-r45 i$
iliii iir* *t*
:63-X64:9.
ri.r l895 S ?i,f*ad]
cI6 cl8
6_55 s.8 x9
?!1.}0 98.S4I
ccl380-r
ff.33 *.53
s.s4 o,30
totstr ?99.?6 I00.0tl
cls o
s.g5
c!3 18s.86 95.03
ct8-1 I.9S x.ffo
ct0 ?.6$ 3_ E3 tot,sl EgS.4O IflS.8
Figure 6.The operating conditions design forthe providing ofHCV 1895 S by a single fractional distillationcolumn
Il{)40.+-3737 tJEl,ltUNS @ Augusr20II IJENS
lrlilIm
lnternational Joumal of Engineering & Technology UET-IJENS Vol: 11 No: 04
10
3.1. How to minimize unsaturated content in V1895 S?
Based on Table 5 and Figure 7 below no doubt at all
that Cl8 purityis over than 95 7o, althoughv 1895 S production yield are (80.0 - 95'0 %). This is caused by
bt6 feed is lower than I Va (Table 4 above).If C16
amount in the feed HCV 1895 S is kept as is in V 1895 S feed and no light end draw offor total reflux in this system./study, ClS purity will not be lower than
90 7o wlw. This also affect to the bigger C16
concentration in V 1895 S that can minimize unsaturated (Cl8-1) content in V 1895 S, the others
components too, C20 and unknown (Unk) (in Table 5)'
The higher S production yield of V 1895 affect to the
smaller C16 content and the higher C20 and unknown
relatively. It affected to the off specs unsaturated content in V 1895 S. It can't be allowed. Please, see also both figures, Figure 7 and Figure 8 below'
s*foh+Eff.ectof,*B*if,:oati:omgreductitr:,xi€K*]s#wr,ra{a.ratesccnlerr1oflxlEEs$
s7 In *pecs
50,3 99,.13 0"04 0,0: 0,16 0,11 111:
51.1 9?,58 CIJ6 oJ3 0,f s
0s-1.0
4 fl"39 1,Xl 97
0.11 6.8-0-9
8?"0
s3$
95
3rr specs
$ff sprcr off
Sawqe: F/ara Sqrtt'ifs C]e*ind+. lS$J
100 99"J
99
fl ss.:
:E99
3 91"5
@
96.J 9d
95"5
91.J8)
*o rbditryfifd tJ lti [!gJE $ii
Figure 7. The influence ofV 1895 S production yield on C18 purity in V 1895 S
Separation of amount C18 into fatty acid residue (see Figure 5 above) affect to the lower amount ofV 1895 S.
Reduction the production yield of V 1895 S. This is
cause of C16 amount ratio to the total amount of side
product V 1895 S is bigger. Finally although without iight end draw off, color of V 1895 S met the quality
standard (not shown in discussion). So the decision not to draw off light end is the exact consideration in the beginning of this study, one technique to minimize the unsaturated content or not reach 0.15 Vo. The ratio can
be changed depend on the changing of V 1895 S
production yield. So in this study reduction of V 1895
S production yield is also the other technique to control
or reduce unsaturated (C18-1) content as it specs (see
Figure 8).
The influence of the reduction of V 1895 S production yield on C18-l content in V 1895 S and reflux rate in Figure 7 can be used as the evidence to support to
state; reduction of V 1895 S production yield will reduce unsaturated content in the main product'
Reduction of V 1895 S production yield is cause of the increasing of reflux rate which reduced the BT of fractionation column (in Figure 9 at discussion 3.4)" The same thing that had been done by a researcher to
reduce stearic acid iodine value or unsaturated content in stearic acid distillate; by reducing BT of distillation
column [18]. The reduction of BT affect to the
reduction of evaporation rate of unsaturated content,
reduce it amount into V 1895 S, of course also the
other components. This is according to the distillation
principles t5ltl llt17l.
1.8
? 1.6
c
cl
> 0.8
: 0.6
E 0.4
*.l* Production yield vs Reflux rale ---*- Production yield vs C18-l conrcnt
(95;0,38)
UO
-0.2 p.od,iJtion yi"ro ,li$ rsss s i&i)
Figure 8. The influence of V 1895 S production yield on C18-lcontent in V 1895 S and reflux rate
I I 0 104--1737 Ll U I''lJ IjNS (cl,\u gu sl l0 I I IJ E^-S
@
International Journal of Engineering & Technology IJET-IJENS Vol: I 1 No: 04
11
-1,i. Prediction versus actual V 1895 S production yield
; ::-:uon of V 18955 unsaturated content was done in
,: . ,:.rdy.Theproduction yield of V 1895 S is tried first
r. '-: asthe design and prediction of the operating
. ,::::rons(Figure 5 above). ReJlux rate is also made so
.:-, rgh in the beginning, on the intervall5OO - 1600 .; -,- This reflux rate is intended to suppress the : .:,:iation rate of unsaturated component as low as
:..,::le because of the boiling point of unsaturated
. -. .cid Cls-1is lower than C1s boiling point[16]. This l-r-: rate is made also because of the amount of C1s-1
: ---: feed HCV 1895 S is so small (0.23 7o in Table 4
-:,-,:r. also the other components (Table 4). On the
:--l'::,rnretofluthxerabteo,ttoitm's
more difficult to be separated product or residue (Figure 5
-;.,r3). Of course the amount of Cra will be bigger in --. ::sidue. It's a risk of a high reflux rate. The feed --.:,s ried first onl.000 kg/hr. These predictions are
.:: according to the experience in the fatty acid ' .:..ation and fractionation in the previous"
'-. :roduction yield of V 1895 S is then increased
:ii-)'(as shown in Table 5 above) after fractional
- .:-iation conditions werestable for 80.0 7o
r-:,:jction yield. This simulation is done for knowing
--. .;rfluence of the increment of V 1895 S on the -::.ment of unsaturated content in V 1895 S. The
':, -.:s rvithin Table 5 proved the prediction above. See
.. Figure 8 below.The higher possible production ..: can be achieved but at the same time unsaturated
- t:,3nt is closer to the maximum point of unsaturated
: --,ent and finally of specs on 9l Vo,93 % and 95 Vo :: -,:.rction yield (Table 5 above and Figure 8
:t.:.r').The in specs one ofV 1895 S should be made in
,- : research. How many percentthe maximum :::'r.rction yield has to be achieved? How many
:r::ent the safest production yield? It has to be met.
-:: rend in this Figure 9 can be used to predict the , --"ble production yield of V 1895 S to provide
The off specs one stearic acid V 1895 S in Table 5
above is sent back the other feed tank as described in Figure 5 above to avoid the higher unsaturated content
in the same feed tank and in the next V 1895 S will be produced.Based on data in this table, it is so clear the influence of the higher production yield V 1895 S to
the higher it unsaturated content. See Figure 8 below,
Base on this figure, V 1895 S production yield is predicted 89.5 Vo to achieve O.l5 7o unsaturated content in V 1895 S (point A). So the safest
production yield should be under 89.5 Vo. Based on this
predictionv 1895 S production yield should be
adjusted to be lower than 89.5 % (see Figure 9 below).
It could be 88 Vo or 89 Vo. Whyit can be happened? It has to be considered strongly the unsaturated content should be achieved first lower than 0.15 Vo. lt is very
important to meet the specs first. If the production
yield is kept constantly as is (89.5 Vo), therc is a big possibility that V 1895 S will be off specs.
Beside this reason the stability of this process is considered strongly. It is not guaranteed all the instru:mentations have the straight line response,
according to the experience it's normally has the small fluctuate on the same amplitude.
0.45
3 0.1 ; 0 35
6 o.r
6 0?5 6 ^i n r{
: 0.r 5 o.o5
,5,
B (91, 0.19 !
a{s9.i,0.ri)
80 *5 90 9'
lro&rtionyi{d of Y lS95 S G6)
t00
.:. rodine value and unsaturated content of V 1895 S
-- Table 5 above) came up to be higher. This is cause
:. *nsaturated component is evaporated more together
, :. \/ 1895 S as the impact of the increment of V -.:: S production yield and/or decreasing of reflux
:r,: see Figure 8 above). The composition of C20 is : :-ler also, caused of the same reasons. This is
.: - :rding to the distillation principles t5l t I 1l I171.
Figure 9. The influence of V 1895 S production yield on unsaturated content in V 1895 S
The actual average production yield is achieved 85.0 Vo. Why is it different? The production yields in Table 5 are not straight since the increment is done, It is done slowly, step by step.It so logic the actual production yield is lower than the predicted one.
i.3. Actual versus predicted operating conditions
-:.: reduction of V 1895 S production yield mean the
-: :3mperature of the fractionation column is adjusted .-.rer. This temperature could be determined and ::::rct practicallyby Roult's Law but have to be : =l-red actually through this research. The predicted
one is used as a main guide and very important to
conduct this research. This prediction was limited by
unsaturated content or double bond of C13-1 in the
quality standard. If the consideration was based only of C18 purity (see Table 3 above), V 1895 S can be draw
I 10.104'3737 lJIll'-lJIlNS @ Augusl 20ll IJENS
@
lnternirtional Journal of Engineering & Technology IJET-IJENS Vol: I I No: 04
t2
off 98 Vo or it production yield 98 70 as it composition
in the feed but actually unsaturated content is off specs when it production yield achieved 93 7o and95 Vo. So
it must belower because of timited by unsaturated content in the quality standard. It's mean also the
actual operating conditions are different with the
predicted operating conditions. See Table 6 until Table
9 below.
simulation data on the same yields from Table 7 below. The data for 93 7o and 95 Vo production yield didn't
take more because of the conditions were up set
already.
The average conditions for the production yields 88 7o, 89 Vo in Table 10 were the average of the simulation data on the same yields from Table 11 below.
The average conditions for the production yields 80 7o,
87 Vo and 93 Va (Table 6) were the average of the
Table 6. The actual operating conditions to produce V 1895 S of l't test
IJroductron TP" mbar TT. { C SPTT. FT, o C BT, o C RR, U:t& E}, mbar
Remarks
v'ipld l9,ir\
o fl
80.0 19.3 .I3X.S l4j{.4 231"4 ?66"6 1.1-1.1 50-J4 Cg-1 0"15 o]$
temperstue; ET:
Table 7. The actual simulation ofthe operating conditions to produce V 1895 S of l't test
IPryrooided8uutcl0cft"i.o.0.nt)TPl,9,.m.Jbar
TT,DDCC t32_S
SPTT,
oc
:44.4
FII,,{{CC
:3?.4
BT,sCC }R-R,E*rii,hh BpP,m,mhh4rar
256"6 1.1-11 5g:54
s0-0 80,0
87,0 s7.0 B?"0
19"5 19-1 25.8 21"6 ?6.0
]]J -0 23t.6
344,2 :.{.1.6
13?.6 ?3?3
?67.0 266.2
1-1-1.: 1.1-1.19
51_54 49-54
?14.5 f,43.6 :36.S 268"0 i.g9_1"1 5S-53
X14-7 214-3
?44.0 t4t.I
J37.0 236.6
?68..+ 1.0S-1.11 ?67.6 1.09,1.1
51_51 49-52
93,0 93,0
:?.: 27.25
203.$ 145.1 :38"9 ?03"7 :44.9 :38.9
2?4-1 0.8.1_0.9 50_51 ??3.9 0.8$_0"39 50-51
9t"0 ts.l :04-5 J44-5 :38"9 ]?i
.83 50-51
Remzrks
Cts-l< 015 0/6 c.8-1{O-t5 9d c.8-1{8-15 9/a cB-:{ *-15 9t *8-:{0-1"5 *6 c*-14 0.1,J % c8-1> 0-15 $6 cg-1> 0"15 BL
tr8-1> 0.15 }t
Soruce: SIora SqU*;g SftS4i*f{k 3
I'{ote: IP - top prersurs; TT: tcp t€nlperaturB; SPTT = side pro'duct ua}'trr.$erertue; r-I: feed iemperaft$e; ET:
Itltgm terRperature ; RR = r,eflux ret* ; BF = boitorn pre*:,r:re
The actual operating conditions on 80.0 7o production yield has a small different than the predicted one (see Figure 5). The actual suitable prediction of production
yield was 89.5 Vo (discussion 3.2 above). It's so
different also than the prediction was done only base on the C18 composition in the feed. Feed temperature can't be reached because of the limitation Oil Thermal
Heater system capacity (on 850-900 kgs/trour feed rate), to conduct heating in a heater and a reboiler of
fractionation column.
Reducing of TT and SPTT trends that almost constant are according to the simulation of Roult's Law (not
shown here) I5l. SPTT trend as is due to Cl8
concentration is almost constant around 98 Vo on the same pressure 27 mbat. Reducing of TT is due to increasing Cl6 or reducing C18 composition in the
total reflux of light end on the same pressure, around
25 mbar, as affect of V 1895 S production yield increment or decreasing of total reflux into the column, This increment is cause of reducing C15
composition in V 1895 S that is added major from light
end and also C18 composition increment that is added
major from bottom product. This increment is also
cause of the increment of BT that is effect to the higher
C18 evaporation so it composition is higher at the production yield 87 Vo (see Figure 8 in the next
discussion 3.4). [ncrement of production yield (priority
93 Vo and 95 Vo) is cause of the increment of BT that is
so effect to the higher content of Cl8-1, C20 and
unknown components, unsaturated content is off specs (0.36 and 0.38 Vo). This is also so affect to the lower
C18 content in V 1895 S. That is why both the last 2
actual operating conditions in Table 6 can't be applied
110.10.1-37.17 IJET'IJENS (E Aueus. 201 I UtjNS
@
International Journal of Engineering & Technology UET-IJENS
11 No: 04
l3
,he next test, but the higher yield should be pointed
., shown as discussion 3.2, based on Figure 7. The : ,"iuction yield 88 and 89 Vo are met graphically and
the operating conditions have to be predicted by Roult's Law, next has to be tested actually, and
compared how precise it.
- ":le 8. The predicted operating conditions to produce V lg95 S of l,ttest
r{d*iolTn)
ss.0
TP, mbar
2==3.3. 6
TT, o C
2, 1,7-
SPTT,
oc
?44
118 ?45
BT,{C RR, ?47 264
, mbar R
9t (beee on ligure 9 of
:l1l&?- ?14445- ?s3-89.0 ?3 J6
246_ 24"7
I st tes$
:.01-1.s.1+
50*
{C8-1= 0.14 +6 CI"15
264 flo, (base on Frgule I
\ote:T}=tgpPr*sure;TT:toFtempi[etue;S}T:=*i*uperEtue;BT=
cflsttest
hSffnS tenper*true ; RR = reflti:r rate ; EP = bottompre:ssire ; ; base on the highest agtual 1.ecu11111Irregflrre
l:ble 9" The predicted of V I 895 S quality on l,t tesr
-i c:6
18-1 C?S
Reflux rate,
ductio* yield
s
. ? 0,76 99.6? 0,06 0"32 0.30* - 1.0I-1.04+ 89$
In specs
Table 10. The actual operating conditions to produce V 1895 S of 2nd test
Production TP,mbar TT,oC SPTT, m rield (Va) 88.0 25.8 214.5 243.6 236.9 268.2 LOg_Lts
5A-52
Remarks C8-1= 0 < 0.15 7p (base
89.0
23.2 229.5 242.7 242.5 265.5 1.03-1.1
50-52
on Table 12) CS-l= 0.02