1

Crude glycerol purification by adsorption via activated carbon

derived from sewage sludge
Chaowat Autthanit
1*
, Mali Hunsom
1, 2

1
Department of Chemical Technology, Faculty of Science, Chulalongorn !ni"ersity, #ango, 1$%%$ Thailan&
2
Center of '(cellence on )etrochemical an& Materials Technology, Chulalongorn !ni"ersity, #ango 1$%%$, Thailan&
*
Author for correspon&ence*

'+Mail, ch-resol"es.gmail-com, Tel- /00 $2 211 2310, Fa(- /00 $2 233 31%1
Abstract: This work was carried out to investigate the
effect of parameters on the preparation of activated
carbon from sewage sludge for crude glycerol
purification. Types of activating agent (KO! "#O$ and
K%CO"&! ratio of activating agent to char ('.()*.( w+w&!
impregnation time (,)%, hrs& and carboni-ation
temperature (,(().((
o
C& were e/plored. The preliminary
results indicated that among the utili-ed activating
agents! the activated carbon activated by KO provided
the best efficiency for crude glycerol purification. The
purity of crude glycerol increased from ", to 0. wt.1 in
the presence of activated carbon activated by KO at
ratio of ,:'! impregnation time of %, hr and
carboni-ation temperature of 0((
o
C. 2oreover! the
o/ygen)containing surface groups on the surface of
activated carbon played a key role on the purity of crude
glycerol.
'. 3ntroduction
4nterest in 5io&iesel 5ase& on fatty aci& methyl
esters has grown consi&era5ly worl&wi&e- This is also
the case in Thailan& where the recent increase in
5io&iesel pro&uction is pre&icte& to 5e at least four+
fol& o"er 6ust four years, from less than 2-171$
0
l8&ay
in 2$$1 to an e(pecte& 1-971$
0
l8&ay in 2$12 :1;- The
pro&uction of 1$ g of 5io&iesel yiel&s appro(imately
1 g of cru&e glycerol :2;, 4ncreasing the share of
5io&iesel in transportation fuels may &ecrease the net
emission of C<2 an& alle"iate the glo5al warming
pro5lem- Howe"er, the increase in 5io&iesel
pro&uction rates will significantly raise the =uantity
an& surplus of cru&e glycerol an& partially waste in the
en"ironment- This is 5ecause the cru&e glycerol
&ischarge& from 5io&iesel pro&uction plants consists
not only of glycerol 5ut also many other chemicals
such as water, salts, soap, alcohol, traces of glyceri&es,
an& "egeta5le colors :%;- 4n a&&ition, high+purity
glycerol is still re=uire& as it is an important in&ustrial
fee&stoc for applications in the foo&, cosmetic an&
pharmaceutical in&ustries, as well as other more minor
uses :9;- #ecause the separation of glycerol "ia
con"entional &istillation processes is not economically
"ia5le, remo"al of impurities "ia a&sorption 5y a
selecti"e a&sor5ent is one of the possi5le ways to
purify cru&e glycerol to accepta5le stan&ar&s :3;-
Consi&ering the potential commercial application
of selecti"e a&sorption of glycerol, it is certainly
necessary to fin& a cheaper an& efficient a&sor5ent for
treatment of cru&e glycerol :0;- The a=ueous+phase
a&sorption of cru&e glycerol has 5een concerne&
mainly on acti"ate& car5on-
4n this wor, the purification of cru&e glycerol
&eri"e& from a waste use&+oil utili>ing 5io&iesel
?methyl ester@ pro&uction plant was carrie& out using
acti"ate& car5on &eri"e& from in&ustrial sewage slu&ge
5y chemical acti"ation metho&- 'ffects of ratio of
acti"ating agent to char, impregnation time an&
car5oni>ation temperature on the properties of
acti"ate& car5on as well as the cru&e glycerol
purification efficiency were in"estigate&
%. 2aterials and 2ethods
2-1 )re+treatment of cru&e glycerol
4nitially, the free fatty aci& an& salt contents were
re&uce& 5y aci&ification- 4n each trial, %$$ g of cru&e
glycerol was aci&ifie& 5y the a&&ition of phosphoric
aci& ?13A H%)<9@ to the &esire& pH 2-3, an& then left
for 12 h until the solution ha& phase separate& into
three &istinct layers, that is a top layer of free fatty
aci&s, the mi&&le glycerol+rich layer an&, if present,
the 5ottom inorganic salt+rich layer- The top layer was
remo"e& 5y slow &ecantation an& the mi&&le glycerol+
rich layer was separate& from the 5ottom inorganic
salt+rich layer, if present, 5y filtration- )rior to remo"al
of the resi&ue salts from glycerol+rich layer, it was
neutrali>e& 5y the a&&ition of 12-3 M Ba<H to pH 2-$,
left for a while an& then filtere& to eliminate the
precipitate& salt :1;-
2-2 )reparation of acti"ate& car5ons
Acti"ate& car5ons from sewage slu&ge were
prepare& in a two+step process, pyrolysis an& chemical
acti"ation processes- 4n the pyrolysis process, sewage
slu&ge was first washe& with water to remo"e the
water+solu5le impurities, &rie& at 1$3
o
C for 29 hrs-
Then the precursor was heate& to 9$$
o
C at the rate of
1$
o
C8min for 2 hrs un&er B2 atmosphere at flow of 3$
ml8min, an& then coole& to room temperature at the
same rate- The resulting material was calle& chars-
The chars were then su56ecte& to chemical
acti"ation- 4n the chemical acti"ation process, 1$-$ g
of the pyroly>e& car5on were mi(e& with 1$$ mC of
&istille& water an& amount of acti"ation agent,
&epen&ing on the ratios of acti"ation reagent8pyroly>e&
car5ons &esire& ?1,1, 2,1, %,1, 9,1, 3,1 an& 0,1 5y
weight@ for each acti"ation agent ?D<H, H%)<9 an&
D2C<%@- The mi(ture was shaen at 2$$ rpm
&epen&ing on the set time ?3+23 hrs@- Finally, the
car5oni>ation was con&ucte& at one of fi"e &ifferent
temperatures ?3$$+E$$
o
C@ un&er nitrogen flow at
heating rate of 1$
o
C8min, an& hel& at %$ min- After
#ure and Applied Chemistry 3nternational Conference %('" (#ACCO4 %('"&
2
acti"ation, the &eri"e& acti"ate& car5ons were remo"e&
from the reactor- Then, it was rinse& with %-$ M HCl
an& &eioni>e& water for se"eral times to remo"e other
inorganic matter until the pH 5ecome neutral, filtere&
an& finally &rie& at 1$3
o
C for 29 hrs-
2-% A&sorption with pro&uce& acti"ate& car5on
A&sorption of pretreate& cru&e glycerol was
con&ucte& at room temperature in conical flass with
continuous shaing- )rior to use, the pro&uce&
acti"ate& car5on was &rie& at 1$3 FC for %$ min to
eliminate free moisture- 4n the a&sorption stage,
acti"ate& car5on was mi(e& with pretreate& cru&e
glycerol at ratio of 02 g8l pretreate& cru&e glycerol,
mi(e& at 23$ rpm for E$ min an& then su56ecte& to
"acuum filtration to remo"e the acti"ate& car5on-
2-9 Analysis
The glycerol content in purifie& cru&e glycerol was
measure& accor&ing to the stan&ar& metho& ?ASTM
D20%2+1$@ :1; 5y a titration with so&ium
metaperio&ate- This test metho& is 5ase& on the col&
o(i&ation of the glycerol 5y so&ium metaperio&ate in a
strong aci&ic me&ium- The glycerol content can 5e
calculate& 5y using '=- ?1@
Glycerol content ?A@ H 1-13123 7 ?I1+I2@ ?1@
m
Jhere,
I1 H "olume of so&ium hy&ro(i&e use& in glycerol
titration ?mC@
I2 H "olume of so&ium hy&ro(i&e use& in 5lan
titration ?mC@
m H weight of glycerol sample use& ?g@
Ash content was analy>e& accor&ing to the
Stan&ar& metho& ?4S< 2$E1+1E22@ 5y 5urning 1 g
glycerol in a muffle furnace at 23$ FC for % h- The
M<BG ?Matter <rganic Bon+glycerol@ le"els an&
water was measure& in terms of contaminants, which
were calculate& 5y the &ifference from a hun&re& of
the pre"ious three compositions :1$$+?A glycerol
content/A ash content@;- The contents of o(ygen+
containing surface functional groups with "arying
aci&ity were &etermine& 5y the #oehmKs metho& of
titration with 5asic solutions of &ifferent 5ase strengths
?BaHC<%, Ba2C<% an& Ba<H@ :2;- The superiority of
sewage slu&ge for acti"ate& car5on pro&uction was
first chece& using pro(imate analysis-
". 5esults and 6iscussion
%-1 )roperties of cru&e glycerol
The original cru&e glycerol was a &ar 5rown
li=ui& with a pH of 1+E- 4t containe& a low content of
glycerol 5ut a relati"ely high ash an& contaminant
contents ?Ta5le 1@- The ash content was largely
compose& of inorganic matter such as so&ium salts that
originate& from the utili>e& catalyst ?Ba<H@ in the
transesterfication process- #y far the largest
contaminant was M<BG, which in&ee& e(cee&e& the
glycerol le"els, an& was generate& 5y the
contamination of soap, methanol an& methyl esters in
the glycerol resi&ue from the 5io&iesel pro&uction
process- During the pre+treatment stage, some of the
fatty aci&s were release& as solu5le soap an& some of
methyl esters &issol"e& or suspen&e& in the glycerol
solution- These free fatty aci&s an& methyl esters then
reacte& with the e(cess Ba<H in the su5se=uent
neutrali>ation step to re+form soap which remaine& in
the glycerol resi&ue :1;-
%-2 'ffect of parameters on cru&e glycerol purification
5y a&sorption with acti"ate& car5on &eri"e& from
slu&ge
%-2-1 'ffect of weight ratio of acti"ating agent to char
'ffects of weight ratio of acti"ating agent to char
on cru&e glycerol purification were carrie& out at the
weight ratio of 1 5y using three types of acti"ating
agent inclu&ing D<H, D2C<% an& H%)<9 at constant
acti"ating time of 2$ hr an& car5oni>e& in B2
atmosphere at 3$$
o
C for %$ min- The surface o(ygen
containing functional groups inclu&ing car5onyl,
car5o(yl, hy&ro(yl an& lactone were first measure&-
As shown in Figure 1, large =uantity of car5onyl was
generate& on the surface of acti"ate& car5on compare&
with other types of acti"ating agent- This group
increase& as the increase of weight ratio of D<H to
char while the other measure& groups remaine&
Ta5le 1- Characteristics of cru&e glycerol, pretreate& cru&e glycerol an& purifie& cru&e glycerol 5y a&sorption
process
)arameters #S 2021,1E2E
:E;
Cru&e
glycerol
)retreate&
cru&e glycerol
)urifie& cru&e glycerol 5y
a&sorption process
Acti"ate&
car5on &eri"e&
from slu&ge
a
Commercial
acti"ate&
car5on
pH 1-$+E-$ 2-$ 2-$ 2-$
Glycerol content ?wt-A@ L 1$ 22-2% M $-19 %3-20 M $-22 1E-%3 M 1-22 19-1 M $-13
Ash content ?wt-A@ N 1$ %0-11 M $-2$ 2E-$2 M $-31 E-E9 M 1-21 1%-1$ M $-1%
Contaminant ?wt-A@ N 1$ %0-3E %3-12 $-21M $-3$ 2-E M $-22
Color ?)t+Co unit@ 122,132 139,$%1 2,920 M %E 90,022 M 2021
a
At the acti"ate& car5on &ose of 02 g8l glycerol at 23$ rpm for E$ min ?D<H 3,1, 23 hr an& 1$$
o
C@
#ure and Applied Chemistry 3nternational Conference %('" (#ACCO4 %('"&
constant or &ecrease slightly- The similar tren& of
=uantity change was o5ser"e& in the case of D2C<%-
Howe"er, in the presence of H%)<9, more =uantities of
hy&ro(yl an& car5o(yl were generate& compare& with
car5onyl an& lactone- The hy&ro(yl an& car5o(yl
groups were 9-E an& 9-1 m'=8g at weight ratio of 0,
while only 2-1 an& 1-3 m'=8g of car5onyl an& lactone
were generate&- This might 5e attri5ute& to the fact that
D<H an& D2C<% are the 5asic chemicals, which can
enhance high generation of 5asic o(ygen functional
groups ?car5onyl group@, while H%)<9 is the aci&
chemical which can enhance a more generation of aci&
o(ygen functional groups ?car5o(yl, hy&ro(yl an&
lactone@-
Figure 1- Contents of o(ygen containing surface
functional groups of the acti"ate& car5on with D<H at
&ifference acti"ating ratios-
Figure 2- 'ffect of weight ratio of acti"ating agent to
char on the purity of cru&e glycerol-
Figure 2 shows "ariation of glycerol content in
cru&e glycerol purifie& 5y a&sorption with acti"ate&
car5on acti"ate& 5y D<H, D2C<% an& H%)<9- 4t was
foun& that the acti"ate& car5on acti"ate& 5y D<H at
the weight ratio of D<H to char of 3,1 pro"i&e& the
5est purity of cru&e glycerol compare& with other types
of acti"ating agent- 4t can enhance the purity of cru&e
glycerol from %3-20 to 01-$1 wt-A, while D2C<% an&
H%)<9 can increase the purity of cru&e glycerol to 31-2
an& 9$-1 wt-A at the weight ratio of 0 an& 9,
respecti"ely- This is 5ecause the D<H can enhance a
high formation of car5onyl group on the surface of
acti"ate& car5on- As propose& 5y )ereira et al- :1$;, the
car5onyl group is the 5asic functional group containing
high =uantity of electron acceptor :11; :12;, which can
5on& as well to the anionic charge of FAM' or FFAs
?O+C<<
+
@, resulting to the &ecrease of contaminant in
purifie& cru&e glycerol- Thus, it fa"ors to 5on& with the
negati"e charge& aci& groups- 4n this case, the cru&e
glycerol containe& large =uantity of free fatty aci&,
which has negati"ely charge& aci& groups :1%;- Thus,
the presence of high =uantity car5onyl group can
re&uce high =uantity of fatty aci& containing in cru&e
glycerol- Howe"er, if the ratio of D<H was increase&
to 0,1, the purity of cru&e glycerol also &ecrease& from
01-$1 to 01-12 wt-A- This was possi5ly &ue to the
&ecomposition of e(cessi"e D<H molecules to water
which can gasify with car5on to form C< an& H2 :19;-
<"er+gasification might re&uce the specific surface
area an& a&sorption of cru&e glycerol-
%-2-2 'ffect of impregnation time
'ffect of impregnation time on cru&e glycerol
purification 5y the a&sortion with acti"ate& car5on
acti"ate& 5y D<H at the weight ratio of 3,1 an&
car5oni>e un&er B2 atmosphere at 3$$
o
C for %$ min
was in"estigate& in the range of 3+23 hrs- The glycerol
content in purifie& cru&e glycerol increase& as the
increase of impregnation time ?Figure %@- This is might
5e attri5ute& to the fact that long impregnation time
woul& promote the &iffusion of D<H in the char,
resulte& to the formation of large =uantity of car5onyl
group from 1-E1 m'=8g at 3 hr to 9-00 m'=8g at 23 hr-
Hence, the impregnation time of 23 hr was re=uire&
an& was use& in su5se=uent e(periments-
Figure %- 'ffect of impregnation time on the purity of
cru&e glycerol purifie& 5y a&sorption with D<H at
weight ratio to char of 3-
%-2-% 'ffect of car5oni>ation temperature
'ffect of car5oni>ation temperature on cru&e
glycerol purification 5y the a&sortion with acti"ate&
car5on acti"ate& 5y D<H at the weight ratio of 3,1 an&
car5oni>e un&er B2 atmosphere at &ifferent temperature
in the range of 3$$+E$$
o
C for %$ min was in"estigate&
at constant impregnation time of 23 hr- As shown in
Figure 9, the glycerol content in purifie& cru&e glycerol
increase& from 22-01 to 1E-%3 wt-A as the increase of
car5oni>ation temperature of 3$$ to 1$$
o
C- This
correspon&s to the increase of car5onyl group on the
surface of acti"ate& car5on from 9-02 m'=8g to 2-$1
m'=8g- Howe"er, further increase the car5oni>e&
temperature from 1$$ to E$$
o
C results to the &ecrease
of glycerol content to 02-$1 wt-A, consistent with the
&ecrease of the =uantity of car5onyl group to 9-13
m'=8g- This is 5ecause the car5onyl groups can
&ecomposition aroun& 13$
o
C :11;, resulting to the
&ecrease of the a&sorption capacity of contaminants
from cru&e glycerol- 4n a&&ition, D<H can con"ert to
potassium car5onates at higher temperature :13;,
resulting to the &ecrease of surface area as well as the
low the capacity to a&sor5 contaminants in cru&e
glycerol-
Figure 9- 'ffect of car5oni>ation temperature on the
purity of cru&e glycerol purifie& 5y a&sorption with
D<H at weight ratio to char of 3-
The properties of purifie& cru&e glycerol purifie&
5y the a&sorption with acti"ate& car5on &eri"e& from
slu&ge an& commercial one were &emonstrate& in Ta5le
1- 4t was clearly seen that the cru&e glycerol purifie&
5y the acti"ate& car5on &eri"e& from slu&ge ha& high
glycerol content than that purifie& 5y commercial
acti"ate& car5on- 4n a&&ition, the color of the former
case was lower than the latter case as &emonstrate& in
Figure 3-


Figure 3- Color of ?a@ cru&e glycerol, ?5@ pretreatment
cru&e glycerol an& purifie& cru&e glycerol 5y
a&sorption with ?c@ commercial acti"ate& car5on an&
?&@ acti"ate& car5on &eri"e& from slu&ge-
$. Conclusions
4n this stu&y, sewage slu&ge has 5een shown to 5e
a feasi5le source for cru&e glycerol purification-
Among utili>e& chemical agent, D<H showe& the
performance compare& with D2C<% an& H%)<9 5ecause
it can enhance the large generation of car5onyl group
on the surface of acti"ate& car5on- Thus, it was chosen
as the most effecti"e acti"ation reagent for acti"ation
of sewage slu&ge for pro&ucing the 5est result in
percentage of glycerol- The optimum con&ition for
preparation of acti"ate& car5on from sewage slu&ge
was o5taine& at acti"ate& car5on acti"ate& 5y D<H at
ratio of 3,1, impregnation time of 23 hr an&
car5oni>ation temperature of 1$$
o
C- This con&ition
pro"i&e& the 5est acti"ate& car5on which ga"e the most
efficiency for cru&e glycerol purification-
Acknowledgements
The authors acnowle&ge the Department of
Chemical Technology, Faculty of Science,
Chulalongorn !ni"ersity for financial support, the
Banyang Te(tile Co-, Ct& for raw material support-
5eferences
:1; O- Manosa, S- Cimpattayanate an& M- Hunsom, Fuel
Processing Technology, .% ?1@, 2$11, E2+EE-
:2; P- Chi, D- )yle, P- Jen, C- Frear an& S- Chen, Process
Biochem- $% ?2$$2@ 13%2+1393-
:%; M- Ha6e an& F- Sopal, Bioresource Technol '('
?2$1$@ %292+%293-
:9; D-T- Qohnson an& D-A- Taconi, Eng. Progress %* ?2$$2@
%%1+%90-
:3; F- Daraosmanoglu, D-#- Cigi>oglu, M- Tuter, an& S-
'rtein, Energy Fuels '( ?1EE0@, 1E$)1E3-
:0; C- )eere5oom, #- Doenigsnecht, M- Hunter, Q-'-
Qacson, an& D-Q- Miller, Carbon $, ?2$$2@ 32E+310-
:2; R- Chen, S- Qeyaseelan an& B- Graham, Waste Manage
%% ?2$$2@ 233+20$-
:1; D-T- Qohnson an& D-A- Taconi, Eng. Progress %* ?2$$2@
%%1+%90-
:E; T-C- <oi, D-C- Song, D- D>ulefly, J-M-P- Jan Sunus
an& A-H- Ha>imah, J. Palm Oil Res- '" ?2$$1@ 10+22-
:1$; M- F- O- )ereira, S- F- Soares, Q- Q- M- <rfao, an& Q- C-
Figueire&o, Carbon $' ?2$$%@ 111+121-
:11; B-D- 'piotis, Journal of the American Chemical ociety
.$ ?1E22@ 1E90-
:12; G-R- Su, S-R- Cu, H- Chen an& P-B- Phu, Carbon $"
?2$$3@ 2213-
:1%; G- Ii"eanan&, urface mo!ification of acti"ate!
carbon for the remo"al of #ater im$urities, The Center
for Microfi5rous Material Manufacturing, Department
of Chemical 'ngineering, Au5urn !ni"ersity, !SA
?2$$1@-
:19; Q- G!< an& A- C- C!A, Chemical Engineering
Research an! %esign 0' ?2$$%@ 313+3E$-
:13; A- Basser an& A- 'l- Hen&awy, A$$lie! urface cience
%,, ?2$$E@ %22%+%2%$
?a@ ?5@ ?c@ ?&@