Conversion of Xylose to Furfural over Lignin-Based Activated Carbon-Supported Iron Catalysts
Full text
This is a sel -a chi ed e sion o an o iginal a icle. This e sion
may di e om he o iginal in pagina ion and ypog aphic de ails.
Au ho (s):
Ti le:
Yea :
Ve sion:
Copy igh :
Righ s:
Righ s u l:
Please ci e he o iginal e sion:
CC BY 4.0
h ps://c ea i ecommons.o g/licenses/by/4.0/
Con e sion o Xylose o Fu u al o e Lignin-Based Ac i a ed Ca bon-Suppo ed I on
Ca alys s
© 2020 he Au ho s
Published e sion
Rusanen, Annu; Kupila, Riikka; Lappalainen, Ka ja; Kä kkäinen, Johanna; Hu, Tao;
Lassi, Ulla
Rusanen, A., Kupila, R., Lappalainen, K., Kä kkäinen, J., Hu, T., & Lassi, U. (2020). Con e sion o
Xylose o Fu u al o e Lignin-Based Ac i a ed Ca bon-Suppo ed I on Ca alys s. Ca alys s, 10(8),
A icle 821. h ps://doi.o g/10.3390/ca al10080821
2020
ca alys s
A icle
Con e sion o Xylose o Fu u al o e Lignin-Based
Ac i a ed Ca bon-Suppo ed I on Ca alys s
Annu Rusanen 1,* , Riikka Kupila 2, Ka ja Lappalainen 1,2 , Johanna Kä kkäinen 1, Tao Hu 1
and Ulla Lassi 1,2
1Resea ch Uni o Sus ainable Chemis y, Uni e si y o Oulu, P.O. Box 4300, FIN-90014 Oulu, Finland;
[email p o ec ed] (K.L.); [email p o ec ed] (J.K.); [email p o ec ed] (T.H.);
[email p o ec ed] (U.L.)
2Uni o Applied Chemis y, Kokkola Uni e si y Conso ium Chydenius, Uni e si y o Jy askylä,
FI-67100 Kokkola, Finland; [email p o ec ed]
*Co espondence: [email p o ec ed]; Tel.: +358-294-48-1636
Recei ed: 30 June 2020; Accep ed: 20 July 2020; Published: 22 July 2020
Abs ac :
In his s udy, con e sion o xylose o u u al was s udied using lignin-based ac i a ed
ca bon-suppo ed i on ca alys s. Fi s , h ee ac i a ed ca bon suppo s we e p epa ed om hyd olysis
lignin wi h di e en ac i a ion me hods. The suppo s we e modi ied wi h di e en me al p ecu so s
and me al concen a ions in o i e i on ca alys s. The p epa ed ca alys s we e s udied in u u al
p oduc ion om xylose using di e en eac ion empe a u es and imes. The bes esul s we e
achie ed wi h a 4 w % i on-con aining ca alys , 5Fe-ACs, which p oduced a 57% u u al yield, 92%
xylose con e sion and 65% eac ion selec i i y a 170
◦
C in 3 h. The amoun o Fe in 5Fe-ACs was only
3.6
µ
mol and using his amoun o homogeneous FeCl
3
as a ca alys , educed he u u al yield, xylose
con e sion and selec i i y. Good ca aly ic ac i i y o 5Fe-ACs could be associa ed wi h i on oxide and
hyd oxyl g oups on he ca alys su ace. Based on he ecycling expe imen s, he p epa ed ca alys
needs some imp o emen s o inc ease i s s abili y bu i is a easible al e na i e o homogeneous
FeCl3.
Keywo ds: u u al; ca bon-suppo ed ca alys ; xylose con e sion; i on; he e ogeneous ca alys s
1. In oduc ion
Fu u al is an impo an biomass-based high- alue chemical wi h nume ous applica ions [
1
].
While i is widely used di ec ly as a sol en and a ungicide, i is mos commonly con e ed
in o pha maceu icals, chemicals and biopolyme s, many o which a e used as subs i u es o
pe ochemical-de i ed analogs [2,3].
App oxima ely 300,000 ons o u u al a e p oduced annually and he bigges manu ac u e is
China [
4
]. I is o med by he dehyd a ion o pen oses, mainly xylose, which can be ob ained h ough
he hyd olysis o di e en ag icul u al esidues (co n s o e , whea s aw, suga cane bagasse, ice
husk, oa hull) o o es indus y was es (bi ch o popla sawdus ). Cu en indus ial p ocesses use
mine al acids, such as sul u ic, phospho ic o hyd ochlo ic acid, as ca alys s, wi h an app oxima e
u u al yield o 50% [
5
,
6
]. The ela i ely low yield, high ene gy consump ion (caused by he high,
150–240
◦
C, eac ion empe a u e) and en i onmen al conce ns ela ed o acidic p ocess was es a e
d i ing scien is s o de elop ca alys s wi h high selec i i y o u u al o ma ion. A numbe o s udies
ha e used inexpensi e wa e -soluble ino ganic sal s (mainly chlo ides) as ca alys s ins ead o mine al
acids o xylose con e sion o u u al [
7
–
10
]. Many o hose s udies show ha FeCl
3
esul s in he
highes u u al yields compa ed o o he me al chlo ides [7,10].
Ca alys s 2020,10, 821; doi:10.3390/ca al10080821 www.mdpi.com/jou nal/ca alys s
Ca alys s 2020,10, 821 2 o 20
Compa ed o wa e -soluble sal s, solid ca aly ic sys ems ha e ewe en i onmen al impac s han
liquid ones, educe ope a ional cos s and a e echnically easible al e na i es o indus y [
11
]. A b oad
ange o di e en he e ogeneous ca alys s ha e been s udied o xylose con e sion o u u al, he mos
common o which a e acidic zeoli es and mesopo ous silicas, such as SBA-15 and MCM-41 [
11
,
12
].
Addi ionally, sul ona ed me al oxides (especially TiO
2
, Z O
2
) ha e been used, ei he as such o
combined wi h mesopo ous suppo s [
13
–
16
]. Many s udies ha e shown ha some B øns ed acidi y
is needed, ei he om me al oxide o suppo , in addi ion o me al oxides wi h a Lewis cha ac e , o
achie e good u u al yields [
17
,
18
]. The need o B øns ed acidi y is associa ed wi h he u u al
p oduc ion mechanism; while Lewis acid can isome ize xylose o xylulose, B øns ed acid is needed o
he dehyd a ion s ep (Scheme 1) [19].
Ca alys s 2020, 10, x FOR PEER REVIEW 2 o 21
Compa ed o wa e -soluble sal s, solid ca aly ic sys ems ha e ewe en i onmen al impac s
han liquid ones, educe ope a ional cos s and a e echnically easible al e na i es o indus y [11].
A b oad ange o di e en he e ogeneous ca alys s ha e been s udied o xylose con e sion o
u u al, he mos common o which a e acidic zeoli es and mesopo ous silicas, such as SBA-15 and
MCM-41 [11,12]. Addi ionally, sul ona ed me al oxides (especially TiO
2
, Z O
2
) ha e been used,
ei he as such o combined wi h mesopo ous suppo s [13–16]. Many s udies ha e shown ha some
B øns ed acidi y is needed, ei he om me al oxide o suppo , in addi ion o me al oxides wi h a
Lewis cha ac e , o achie e good u u al yields [17,18]. The need o B øns ed acidi y is associa ed
wi h he u u al p oduc ion mechanism; while Lewis acid can isome ize xylose o xylulose,
B øns ed acid is needed o he dehyd a ion s ep (Scheme 1) [19].
Scheme 1. Dehyd a ion o xylose o u u al by B øns ed acid and Lewis acid ca alys s.
Ca bon-based ca alys s a e a ac i e since he ca bon suppo s a e low-cos ma e ials wi h high
su ace a ea and good he mal s abili y and hey a e easily modi ied wi h unc ional g oups [20].
Ca bon su ace g oups con aining he e oa oms, such as oxygen, can ac as ancho ing si es o me al
pa icles and gene a e high me al dispe sion [21]. The di e se su ace also enables B øns ed acidi y
in addi ion o me al’s Lewis acidi y, which can u he p omo e xylose con e sion o u u al. To he
bes o ou knowledge, ew pape s ha e been published on ca bon-suppo ed me al oxide ca alys s.
Mazzo a e al. used a sul ona ed ca bonaceous ma e ial wi h TiO
2
si es, while Russo e al. used
TiO
2
/ca bon black in he con e sion o xylose o u u al [22,23]. Ba oso-Bogea e al. p epa ed a
class o di e en ac i a ed ca bon-me al oxide ca alys s (Fe, Al, Zn, Sn, Ti and W) bu did no use
hem in any eac ion [24–26]. The e o e, cheap me al oxides, such as i on oxide, ha e no been
u ilized in xylose con e sion o u u al, e en hough i on is common in homogeneous ca alys s. The
aim o his s udy was o c ea e i on oxide si es on a ca bon-suppo ed ca alys and apply he ca alys
in he con e sion o xylose o u u al. Ac i a ed ca bon suppo was de i ed om hyd olysis lignin,
which is a side s eam o cellulosic e hanol p oduc ion. Reac ions we e pe o med in biphasic media
o inc ease eac ion selec i i y.
2. Resul s and Discussion
2.1. P elimina y S udies
The pu pose o he p elimina y s udies was o op imize he eac ion media and selec he bes
ca aly ic me al o he e ogeneous ca alys s.
2.1.1. Fu u al Pa i ioning in Biphasic Reac ion Sys em
The eac ion sys em was op imized in e ms o he app op ia e o ganic sol en and
sol en :wa e a io. Expe imen s we e ca ied ou using a 4.7 w % u u al solu ion in wa e as eed
and oluene o me hyl isobu yl ke one (MIBK) as an o ganic sol en . The pu pose was o compa e
u u al pa i ioning om wa e in o hese wo sol en s. Sol en a ios o 1:1, 1:2 and 1:3 we e used
based on he li e a u e [17,27,28]. The esul s o pa i ioning expe imen s a e shown in Figu e 1, and,
as expec ed, he mo e o ganic sol en was added, he be e u u al was ex ac ed o he o ganic
Scheme 1. Dehyd a ion o xylose o u u al by B øns ed acid and Lewis acid ca alys s.
Ca bon-based ca alys s a e a ac i e since he ca bon suppo s a e low-cos ma e ials wi h high
su ace a ea and good he mal s abili y and hey a e easily modi ied wi h unc ional g oups [
20
].
Ca bon su ace g oups con aining he e oa oms, such as oxygen, can ac as ancho ing si es o me al
pa icles and gene a e high me al dispe sion [
21
]. The di e se su ace also enables B øns ed acidi y
in addi ion o me al’s Lewis acidi y, which can u he p omo e xylose con e sion o u u al. To he
bes o ou knowledge, ew pape s ha e been published on ca bon-suppo ed me al oxide ca alys s.
Mazzo a e al. used a sul ona ed ca bonaceous ma e ial wi h TiO
2
si es, while Russo e al. used
TiO
2
/ca bon black in he con e sion o xylose o u u al [
22
,
23
]. Ba oso-Bogea e al. p epa ed a
class o di e en ac i a ed ca bon-me al oxide ca alys s (Fe, Al, Zn, Sn, Ti and W) bu did no use
hem in any eac ion [
24
–
26
]. The e o e, cheap me al oxides, such as i on oxide, ha e no been u ilized
in xylose con e sion o u u al, e en hough i on is common in homogeneous ca alys s. The aim
o his s udy was o c ea e i on oxide si es on a ca bon-suppo ed ca alys and apply he ca alys in
he con e sion o xylose o u u al. Ac i a ed ca bon suppo was de i ed om hyd olysis lignin,
which is a side s eam o cellulosic e hanol p oduc ion. Reac ions we e pe o med in biphasic media o
inc ease eac ion selec i i y.
2. Resul s and Discussion
2.1. P elimina y S udies
The pu pose o he p elimina y s udies was o op imize he eac ion media and selec he bes
ca aly ic me al o he e ogeneous ca alys s.
2.1.1. Fu u al Pa i ioning in Biphasic Reac ion Sys em
The eac ion sys em was op imized in e ms o he app op ia e o ganic sol en and sol en :wa e
a io. Expe imen s we e ca ied ou using a 4.7 w % u u al solu ion in wa e as eed and oluene
o me hyl isobu yl ke one (MIBK) as an o ganic sol en . The pu pose was o compa e u u al
pa i ioning om wa e in o hese wo sol en s. Sol en a ios o 1:1, 1:2 and 1:3 we e used based on
he li e a u e [
17
,
27
,
28
]. The esul s o pa i ioning expe imen s a e shown in Figu e 1, and, as expec ed,
Ca alys s 2020,10, 821 3 o 20
he mo e o ganic sol en was added, he be e u u al was ex ac ed o he o ganic phase (Figu e 1a).
MIBK showed a be e abili y o ex ac u u al han oluene, as he u u al con en in wa e was
ela i ely highe in he oluene expe imen s han in he MIBK expe imen s. In he bes case, when 1:3
wa e :MIBK was used as a sol en , 96% o u u al was ex ac ed o he MIBK phase. Simila ly, MIBK
was epo ed o ex ac u u al be e han oluene in he li e a u e [
27
,
29
]. MIBK has a pola ca bonyl
g oup s uc u e ha may in e ac be e wi h u u al, which is an aldehyde, compa ed o nonpola
oluene. The pa i ioning coe icien o MIBK has been epo ed o be 7 [
29
], while in ou s udy i was 8
using a simila sol en a io (1:1) and 8.5 when he sol en a io was 1:3 (Figu e 1b). Fo oluene, he
pa i ioning coe icien has been epo ed o be 3 [
29
] bu in ou s udy i was signi ican ly highe (6)
wi h a 1:1 sol en a io. Di e ences in epo ed and measu ed pa i ioning coe icien s a e likely due o
di e en expe imen al se ups. In he ci ed s udy, he pa i ioning coe icien s we e de e mined wi hou
any hea ing, using only shaking as an ex ac ion echnique. In ou s udy, 5 min mic owa e hea ing
(a 160
◦
C) was used, as he highe empe a u e co esponds o he eac ion condi ions in con e sion.
Based on he pa i ioning esul s, MIBK was conside ed mos sui able o he u u al emo al om
wa e and i was used as an o ganic sol en in u he expe imen s.
Ca alys s 2020, 10, x FOR PEER REVIEW 3 o 21
phase (Figu e 1a). MIBK showed a be e abili y o ex ac u u al han oluene, as he u u al
con en in wa e was ela i ely highe in he oluene expe imen s han in he MIBK expe imen s. In
he bes case, when 1:3 wa e :MIBK was used as a sol en , 96% o u u al was ex ac ed o he MIBK
phase. Simila ly, MIBK was epo ed o ex ac u u al be e han oluene in he li e a u e [27,29].
MIBK has a pola ca bonyl g oup s uc u e ha may in e ac be e wi h u u al, which is an
aldehyde, compa ed o nonpola oluene. The pa i ioning coe icien o MIBK has been epo ed o
be 7 [29], while in ou s udy i was 8 using a simila sol en a io (1:1) and 8.5 when he sol en a io
was 1:3 (Figu e 1b). Fo oluene, he pa i ioning coe icien has been epo ed o be 3 [29] bu in ou
s udy i was signi ican ly highe (6) wi h a 1:1 sol en a io. Di e ences in epo ed and measu ed
pa i ioning coe icien s a e likely due o di e en expe imen al se ups. In he ci ed s udy, he
pa i ioning coe icien s we e de e mined wi hou any hea ing, using only shaking as an ex ac ion
echnique. In ou s udy, 5 min mic owa e hea ing (a 160 °C) was used, as he highe empe a u e
co esponds o he eac ion condi ions in con e sion. Based on he pa i ioning esul s, MIBK was
conside ed mos sui able o he u u al emo al om wa e and i was used as an o ganic sol en
in u he expe imen s.
(a) (b)
Figu e 1. Fu u al pa i ioning be ween wa e and me hyl isobu yl ke one (MIBK) and wa e and
oluene using di e en sol en a ios. (a) Pa i ioning calcula ed as he a io o moles o u u al in
he o ganic phase and in he wa e phase; (b) Pa i ioning calcula ed as he pa i ioning coe icien
(based on concen a ions).
2.1.2. Fu u al P oduc ion Using Homogeneous Ca alys s
A e choosing he eac ion media, homogeneous me al sal s (ZnCl2, AlCl3·6H2O, C Cl3·6H2O,
SnCl2·2H2O and FeCl3·6H2O) we e used as ca alys s o de e mine he bes me al o u u al
p oduc ion. Di e en me al sal s ha e been s udied in he li e a u e bu i was impo an o pe o m
es s in ou eac ion sys em, using mic owa e hea ing and wa e /MIBK media. Reac ions we e
pe o med a a empe a u e o 160 °C, a eac ion ime o 1.5 h and a ca alys amoun o 0.05 mmol,
based on he li e a u e [10]. Almos ull xylose con e sion (98%–99%) was achie ed using all
ca alys s o he han zinc chlo ide, which yielded only 91% con e sion (Figu e 2, as e isks). In he
u u al yields, he e was a ia ion be ween ca alys s and he yields a ied be ween 33% and 68%
(Figu e 2, ba s). Ch omium chlo ide and aluminum chlo ide p oduced he lowes u u al yields,
al hough he xylose con e sion was almos comple e. The e o e, hei selec i i ies (34% and 38%,
espec i ely) we e poo es among all ca alys s (Figu e 2, squa es). Tin chlo ide and zinc chlo ide
p oduced sligh ly highe u u al yields compa ed o ch omium o aluminum and hei selec i i ies
we e also highe (46% and 48%, espec i ely). Howe e , he signi ican ly highes u u al yield
(68%) and selec i i y (70%) we e achie ed using i on chlo ide as a ca alys . Ou esul ag ees wi h
he li e a u e, as many s udies ha e shown ha FeCl3 p oduces he highes u u al yield in wa e
Figu e 1.
Fu u al pa i ioning be ween wa e and me hyl isobu yl ke one (MIBK) and wa e and
oluene using di e en sol en a ios. (
a
) Pa i ioning calcula ed as he a io o moles o u u al in he
o ganic phase and in he wa e phase; (
b
) Pa i ioning calcula ed as he pa i ioning coe icien (based
on concen a ions).
2.1.2. Fu u al P oduc ion Using Homogeneous Ca alys s
A e choosing he eac ion media, homogeneous me al sal s (ZnCl
2
, AlCl
3·
6H
2
O, C Cl
3·
6H
2
O,
SnCl
2·
2H
2
O and FeCl
3·
6H
2
O) we e used as ca alys s o de e mine he bes me al o u u al p oduc ion.
Di e en me al sal s ha e been s udied in he li e a u e bu i was impo an o pe o m es s in ou
eac ion sys em, using mic owa e hea ing and wa e /MIBK media. Reac ions we e pe o med a a
empe a u e o 160
◦
C, a eac ion ime o 1.5 h and a ca alys amoun o 0.05 mmol, based on he
li e a u e [
10
]. Almos ull xylose con e sion (98%–99%) was achie ed using all ca alys s o he han
zinc chlo ide, which yielded only 91% con e sion (Figu e 2, as e isks). In he u u al yields, he e was
a ia ion be ween ca alys s and he yields a ied be ween 33% and 68% (Figu e 2, ba s). Ch omium
chlo ide and aluminum chlo ide p oduced he lowes u u al yields, al hough he xylose con e sion
was almos comple e. The e o e, hei selec i i ies (34% and 38%, espec i ely) we e poo es among
all ca alys s (Figu e 2, squa es). Tin chlo ide and zinc chlo ide p oduced sligh ly highe u u al
yields compa ed o ch omium o aluminum and hei selec i i ies we e also highe (46% and 48%,
espec i ely). Howe e , he signi ican ly highes u u al yield (68%) and selec i i y (70%) we e
Ca alys s 2020,10, 821 4 o 20
achie ed using i on chlo ide as a ca alys . Ou esul ag ees wi h he li e a u e, as many s udies
ha e shown ha FeCl
3
p oduces he highes u u al yield in wa e media compa ed o o he me al
halides [
7
,
10
]. The 68% u u al yield achie ed in ou s udy wi h FeCl
3
ca alys is also compa able
o ha ound in o he s udies. Fo example, E sho a e al. achie ed a 60.3% u u al yield in wa e
a 180
◦
C, while Zhang e al. epo ed a 77% u u al yield in gamma- ale olac one a 160
◦
C [
10
,
30
].
Howe e , in mos s udies NaCl has been used as a phase modi ie , and, because Cl
−
has been shown o
ca alyze u u al o ma ion, i is di icul o compa e hose esul s o ou s udy wi hou NaCl [31–34].
Ca alys s 2020, 10, x FOR PEER REVIEW 4 o 21
media compa ed o o he me al halides [7,10]. The 68% u u al yield achie ed in ou s udy wi h
FeCl3 ca alys is also compa able o ha ound in o he s udies. Fo example, E sho a e al. achie ed
a 60.3% u u al yield in wa e a 180 °C, while Zhang e al. epo ed a 77% u u al yield in
gamma- ale olac one a 160 °C [10,30]. Howe e , in mos s udies NaCl has been used as a phase
modi ie , and, because Cl− has been shown o ca alyze u u al o ma ion, i is di icul o compa e
hose esul s o ou s udy wi hou NaCl [31–34].
Figu e 2. Fu u al yield (ba ), espec i e con e sion (as e isk) and eac ion selec i i y (squa e)
du ing xylose dehyd a ion a 160 °C o 1.5 h using di e en me al chlo ides (0.05 mmol) as ca alys s.
2.2. P epa a ion and Cha ac e iza ion o He e ogeneous Ca alys s
A e he app op ia e eac ion media and ca aly ic me al we e chosen, h ee di e en ca alys
suppo s (ACz, ACzN, ACs) and i e di e en ca alys s (5FeNO3-ACz, 5Fe-ACz, 5Fe-ACzN, 5Fe-ACs
and 10Fe-Acs) we e p epa ed (Table 1). In he p epa a ions, di e en ac i a ion me hods, me al
p ecu so s, me al concen a ions and addi ional ea men s we e used. ACz suppo was p epa ed
applying common ZnCl2 chemical ac i a ion [35]. ACz-based ca alys s 5FeNO3-ACz and 5Fe-ACz
we e p epa ed using 5 w % o ei he FeNO3 o FeCl3 as an i on p ecu so , espec i ely. ACzN
suppo was p epa ed using HNO3 ea men a e ZnCl2 chemical ac i a ion and ACzN-based
ca alys 5Fe-ACzN by u he imp egna ion wi h FeCl3. The hi d suppo ype, ACs, was p epa ed
using he physical ac i a ion me hod. Compa ed o chemical ac i a ion, physical ac i a ion is
pe o med a a highe empe a u e bu only using wa e s eam as an ac i a ion agen [21,36].
ACs-based ca alys s 5Fe-ACs and 10Fe-ACs we e p epa ed using FeCl3 as a me al p ecu so and 5 o
10 w % as he me al concen a ion.
Figu e 2.
Fu u al yield (ba ), espec i e con e sion (as e isk) and eac ion selec i i y (squa e) du ing
xylose dehyd a ion a 160 ◦C o 1.5 h using di e en me al chlo ides (0.05 mmol) as ca alys s.
2.2. P epa a ion and Cha ac e iza ion o He e ogeneous Ca alys s
A e he app op ia e eac ion media and ca aly ic me al we e chosen, h ee di e en ca alys
suppo s (ACz, ACz
N
, ACs) and i e di e en ca alys s (5Fe
NO3
-ACz, 5Fe-ACz, 5Fe-ACz
N
, 5Fe-ACs and
10Fe-Acs) we e p epa ed (Table 1). In he p epa a ions, di e en ac i a ion me hods, me al p ecu so s,
me al concen a ions and addi ional ea men s we e used. ACz suppo was p epa ed applying
common ZnCl
2
chemical ac i a ion [
35
]. ACz-based ca alys s 5Fe
NO3
-ACz and 5Fe-ACz we e p epa ed
using 5 w % o ei he FeNO
3
o FeCl
3
as an i on p ecu so , espec i ely. ACz
N
suppo was p epa ed
using HNO
3
ea men a e ZnCl
2
chemical ac i a ion and ACz
N
-based ca alys 5Fe-ACz
N
by u he
imp egna ion wi h FeCl
3
. The hi d suppo ype, ACs, was p epa ed using he physical ac i a ion
me hod. Compa ed o chemical ac i a ion, physical ac i a ion is pe o med a a highe empe a u e
bu only using wa e s eam as an ac i a ion agen [
21
,
36
]. ACs-based ca alys s 5Fe-ACs and 10Fe-ACs
we e p epa ed using FeCl3as a me al p ecu so and 5 o 10 w % as he me al concen a ion.
Table 1.
Key ac o s in he p epa a ion o a ious suppo s and ca alys s and he me al con en s (w %)
o he p epa ed ca alys s measu ed by induc i ely coupled plasma op ical emission spec ome y
(ICP-OES).
Sample Ac i a ion
Me hod
O he
T ea men
Me al
P ecu so
Ini ial Fe
(w %)
Measu ed Me al (w %)
Zn Fe
ACs S eam (H2O) - - - 0.01 0.06
5Fe-ACs S eam (H2O) - FeCl35 - 4.0
10Fe-ACs S eam (H2O) - FeCl310 - 9.2
ACz ZnCl2- - - 8.2 0.08
5FeNO3-ACz ZnCl2- FeNO35 4.6 5.0
5Fe-ACz ZnCl2- FeCl35 3.8 4.5
ACzNZnCl2HNO3- - 0.07 0.06
5Fe-ACzNZnCl2HNO3FeCl35 0.08 5.5
Ca alys s 2020,10, 821 5 o 20
The me al con en s (acco ding o induc i ely coupled plasma op ical emission spec ome y
(ICP-OES)) o all suppo s and ca alys s a e lis ed in Table 1. ACs suppo was analyzed mos
comp ehensi ely since i was he leas ea ed and he e o e p o ided some indica ion abou he
me al con en o he biomass-based lignin. Acco ding o he esul s, ACs con ained mino amoun s
o me als, such as Ca (0.47 w %) and Na (0.57 w %) bu i s Fe con en was e y low (0.06 w %)
(Table S1). ACs-based i on-imp egna ed ca alys s 5Fe-ACs and 10Fe-ACs con ained 4.0 and 9.2 w %
i on, espec i ely. Based on hese alues, i on imp egna ion was conside ed success ul, as he a ge
amoun s we e 5 and 10 w %. The o he suppo , ACz, con ained a signi ican amoun o zinc (8.2
w %), which o igina ed om he chemical ac i a ion s ep in he p epa a ion p ocess. Zinc was also
na u ally p esen in ACz-suppo ed 5Fe-ACz and 5Fe
NO3
-ACz (3.8 and 4.6 w %, espec i ely) bu he
amoun dec eased du ing he imp egna ion s ep. The i on con en s o 5Fe-ACz and 5Fe
NO3
-ACz
we e 4.5 and 5.0 w %, espec i ely. Because o he emaining zinc, a hi d ca alys suppo ACz
N
was
p epa ed simila ly o ACz bu a e wa ds i was ea ed wi h HNO
3
in o de o emo e emaining zinc.
Acco ding o he ICP-OES esul s, his ea men did emo e zinc because he emaining amoun was
only 0.07 w %. Meanwhile, he i on con en o 5Fe-ACzNwas 5.5 w %.
Addi ionally, he su ace a ea (SA) acco ding o B unaue –Emme –Telle (BET) heo y, a e age
po e diame e and po e olume dis ibu ions acco ding o densi y unc ional heo y (DFT) we e
de e mined o all suppo s and ca alys s using N
2
-physiso p ion analysis (Table 2). As can be seen
om he esul s, he su ace a ea and po e olume o he chemically ac i a ed ACz and ACz
N
suppo s
we e highe (1470/1091 m
2
g
−1
and 0.72/0.49 cm
3
g
−1
) han o he ACs suppo p epa ed wi h s eam
ac i a ion (760 m
2
g
−1
and 0.47 cm
3
g
−1
) (en ies 1, 4 and 7). Fu he , he o al po e olume and he
ela i e amoun o mic opo es we e highe in ACz and ACz
N
han in ACs. T ea men wi h HNO
3
a e
chemical ac i a ion dec eased he suppo su ace a ea, a e age po e diame e and o al po e olume
bu he quan i ies we e s ill g ea e han wi h s eam ac i a ed ACs.
Table 2. N2-physiso p ion analysis om di e en ac i a ed ca bon (AC) suppo s and ca alys s.
En y Sample BET DFT
BET
SA (m2/g)
A g. Po e
Diam. (nm)
To al Po e
Volume
(cm3/g)
Mesopo es
(cm3/g)
Mic opo es
(cm3/g)
1 ACs 760 2.90 0.47 0.26 0.21
2 5Fe-ACs 455 3.44 0.34 0.23 0.11
3 10Fe-ACs 380 3.15 0.26 0.16 0.10
4 ACz 1470 2.29 0.72 0.31 0.41
5 5Fe-ACz 1000 2.28 0.48 0.19 0.29
6 5FeNO3-ACz 948 2.16 0.45 * 0.15 0.29
7 ACzN1091 2.15 0.49 0.16 0.33
8 5Fe-ACzN790 2.07 0.35 0.10 0.25
* also con ained mac o-po es 0.01 cm3/g (3%).
The su ace a ea and o al po e olume o all ca alys s dec eased while he i on was imp egna ed
on he ca bon su ace. This is easonable, as me al is deposi ed on he su ace and goes in o he po es
(Table 2). Fo ACs-based ca alys s, he su ace a ea dec eased om 860 o 455 o 380 m
2
/g and o al
po e olume om 0.47 o 0.34 o 0.26 cm
3
/g, depending on how much me al was imp egna ed on he
ca alys su ace (en ies 1–3). Fo ACz-based ca alys s, he su ace a ea dec eased om 1470 o 1000 o
948 m
2
/g and he o al po e olume om 0.72 o 0.48 o 0.45 cm
3
/g, depending on which i on p ecu so
was used (en ies 4–6). Fo ACz
N
-based ca alys s, he su ace a ea dec eased om 1091 o 790 m
2
/g
and he o al po e olume dec eased om 0.49 o 0.35 (en ies 7–8).
Absolu e olumes o he meso- and mic opo es dec eased wi h all ca alys s when i on was
imp egna ed bu he e we e di e ences in he inal po e olume dis ibu ions be ween meso- and
mic opo es, pa ly because o di e en ini ial po e olume dis ibu ions o he di e en suppo s.
Ca alys s 2020,10, 821 6 o 20
Fo s eam-ac i a ed ACs, he mic opo e olume dec eased in 5Fe-ACs and 10Fe-ACs ca alys s by
48%–52% compa ed o ACs. Wi h mesopo es, he dec ease was smalle , only 12%–38%, meaning
ha he mic opo es we e p ima ily illed in ACs when i on was imp egna ed. Because mic opo es
we e mo e illed wi h i on han mesopo es, he a e age po e diame e also inc eased wi h ACs-based
ca alys s when compa ed o plain ACs. Compa ing 5Fe-ACs and 10Fe-ACs, he di e ence in mesopo e
olume was no able, which mean ha 5% i on addi ion did no a ec he mesopo es signi ican ly
while 10% i on addi ion illed he mesopo es conside ably. Fo chemically ac i a ed ACz and ACz
N
,
he mic opo e olume dec eased only 29% o 24%, espec i ely, while he olume o he mesopo es
was dec eased 39%–52% o ACz and 38% o ACz
N
. Di e en p ecu so s did no ha e a signi ican
e ec on su ace a ea o po e olumes. Because he olume o he mic opo es dec eased less han he
olume o he mesopo es, he ela i e amoun o mic opo es was inc eased in chemically ac i a ed
i on ca alys s. In addi ion, he a e age po e diame e dec eased compa ed o he suppo s.
O he su ace p ope ies, such as unc ional g oups, acidi y and de ailed me al composi ion,
we e s udied using X- ay pho oelec on spec oscopy (XPS), he Boehm i a ion me hod and X- ay
di ac ion (XRD). These analyses we e only conduc ed o ACs, 5Fe-ACs and 10Fe-ACs, which we e
de e mined o be he mos p omising ones in he con e sion o xylose o u u al (see Sec ion 2.3).
XPS esul s show ha plain ACs suppo al eady con ains some oxygen unc ionali ies (hyd oxyl and
ca boxyl g oups) on he su ace bu i on addi ion inc eases he oxygen–ca bon a io o he ca alys s
compa ed o ACs suppo (Table 3, Table S2). The inc ease is also dependen on how much i on is
added, as he pe cen age o ca bon a oms dec eases om 96.8 o 93.6 o 86.3% and ha o oxygen
a oms inc eases om 2.9 o 4.1 o 9.5% when he amoun o i on inc eased om 0.06 o 4.0 o o 9.2
w %, espec i ely. In addi ion o C, O and Fe, Cl was de ec ed om i on-imp egna ed ca alys s. An
inc ease o oxygen unc ionali ies was de ec ed om an O1s scan a 531.0eV and a 532.5eV (Table
S2). The o me can o igina e om ca bonyl g oups o me al oxides and he la e , o example,
om hyd oxyl g oups o O-Fe bonds [
37
,
38
]. F om he XPS Fe2p spec a, a peak a 711.3 eV was
de ec ed om bo h 5Fe-ACs and 10Fe-ACs ca alys s, indica ing he p esence o oxidized i on, Fe
2
O
3
o
FeOOH [
37
,
39
]. Acco ding o XRD, he i on was mos ly p esen as oxides (Fe
3
O
4
and Fe
2
O
3
, Figu e S1),
so i is p oposed ha oxygen con en was inc eased oge he wi h i on con en as i on oxide. Acco ding
o XRD, 10Fe-ACs con ained mos ly Fe
2
O
3
(hema i e, 01-080-5405) and only small amoun s o Fe
3
O
4
(magne i e, 04-015-9120). Con e sely, 5Fe-ACs con ained clea ly mo e Fe
3
O
4
han Fe
2
O
3
. Howe e , no
i on chlo ides we e de ec ed wi h XRD measu emen s.
Table 3.
Su ace analysis o p epa ed ca bon suppo and ca alys s acco ding o X- ay pho oelec on
spec oscopy (XPS) and Boehm i a ion.
Sample XPS a
To al C-%
om C1s
To al O-%
om O1s
To al Fe-%
om Fe2p
To al Cl-%
om Cl2p
To al Acidic
G oups
(mmol/g) b
ACs 96.8 3.0 nd nd 0.07
5Fe-ACs 93.7 4.1 0.6 1.5 1.77
10Fe-ACs 86.2 9.6 2.4 1.6 1.95
aas a om-%, bby Boehm i a ion, nd =no de ec ed.
Acco ding o Boehm i a ion, plain ACs con ained a small amoun o acidic oxygen unc ionali ies
(0.07 mmol/g, Table 3, Figu e S2.). The amoun o acidic oxygen g oups inc eased when he me al was
added o 1.77 o 1.95, depending on he me al amoun . Acidic oxygen unc ionali ies a e p obably
o med du ing i on imp egna ion as a consequence o HCl o ma ion om FeCl
3
hyd a ion in wa e
solu ion. Ba oso-Bogea e al. showed ha me al ions ma kedly in luence he pH o he imp egna ion
solu ion and he eby he oxidizing powe o his solu ion owa d he ac i a ed ca bon suppo [
40
].
Since XPS e ealed he po en ial o he p esence o i on hyd oxides, i is possible ha B øns ed acidi y
Ca alys s 2020,10, 821 7 o 20
o 5Fe-ACs and 10Fe-ACs is induced by i on hyd oxides (e.g., FeOOH) [
25
,
41
]. Rega dless o he
speci ic na u e o he acidic oxygen g oups, hey a o me al adso p ion [42].
The mo phology o he physically ac i a ed suppo (ACs) and ca alys s (5Fe-ACs and 10Fe-ACs)
was obse ed using a scanning elec on mic oscope (SEM) and a scanning ansmission elec on
mic oscope (STEM). SEM images clea ly e ealed pa icles on he ca bon su ace o 5Fe-ACs and
10Fe-ACs, while plain ACs did no con ain any isible pa icles (Figu e 3). The suppo and bo h
ca alys s showed a e y po ous s uc u e in SEM as well as in STEM. Figu e 4(and Figu e S3) shows
chemical mapping o he elemen s C, O, Fe and Cl using ene gy-dispe si e x- ay spec oscopy in
scanning ansmission elec on mic oscopy (STEM-EDS). Compa ing he dis ibu ion o Fe and O, i
is clea ha bo h elemen s appea a he same loca ion, which indica es he p esence o i on oxide.
The e o e, he esul s ob ained om XPS and XRD showing ha he i on pa icles we e oxides we e
con i med. Mapping also showed ha signi ican amoun o esidual chlo ine was e enly dis ibu ed
on he su ace. No able chlo ine emains ha e been also de ec ed in he li e a u e when FeCl
3
has been
used as a me al p ecu so [43].
Ca alys s 2020, 10, x FOR PEER REVIEW 8 o 21
Figu e 3. Scanning elec on mic oscopy (SEM) images o 5Fe-ACs (a), 10Fe-ACs (b) and ACs (c). I on
pa icles a e clea ly isible in 5Fe-ACs and 10Fe-ACs while ACs shows a po ous s uc u e.
Figu e 4. Chemical mapping o 5Fe-ACs using ene gy-dispe si e x- ay spec oscopy in scanning
ansmission elec on mic oscopy (STEM-EDS). The igu e e eals ha i on is mos likely p esen as
i on oxide on he ca bon su ace, as i on and oxygen appea in he same loca ions.
Figu e 3.
Scanning elec on mic oscopy (SEM) images o 5Fe-ACs (
a
), 10Fe-ACs (
b
) and ACs (
c
). I on
pa icles a e clea ly isible in 5Fe-ACs and 10Fe-ACs while ACs shows a po ous s uc u e.
Ca alys s 2020,10, 821 8 o 20
Ca alys s 2020, 10, x FOR PEER REVIEW 8 o 21
Figu e 3. Scanning elec on mic oscopy (SEM) images o 5Fe-ACs (a), 10Fe-ACs (b) and ACs (c). I on
pa icles a e clea ly isible in 5Fe-ACs and 10Fe-ACs while ACs shows a po ous s uc u e.
Figu e 4. Chemical mapping o 5Fe-ACs using ene gy-dispe si e x- ay spec oscopy in scanning
ansmission elec on mic oscopy (STEM-EDS). The igu e e eals ha i on is mos likely p esen as
i on oxide on he ca bon su ace, as i on and oxygen appea in he same loca ions.
Figu e 4.
Chemical mapping o 5Fe-ACs using ene gy-dispe si e x- ay spec oscopy in scanning
ansmission elec on mic oscopy (STEM-EDS). The igu e e eals ha i on is mos likely p esen as
i on oxide on he ca bon su ace, as i on and oxygen appea in he same loca ions.
2.3. Fu u al P oduc ion Using He e ogeneous Ca alys s
Con e sion s udies we e s a ed wi h he con ol expe imen wi hou any suppo o ca alys
(Table 4). This so-called au oca alysis was able o p oduce a 12% u u al yield and 18% xylose
con e sion a 160
◦
C in 1.5 h. Au oca alysis is based on high empe a u e and p essu e, whe e he
dissocia ion cons an o wa e is inc eased [
44
]. In addi ion, o ma ion o o ganic acids (e.g., o mic and
lac ic acid) du ing he eac ion migh occu , which hen u he ca alyzes he hyd olysis eac ion [
45
].
Howe e , a 160
◦
C eac ion empe a u e and 1.5-h eac ion ime ep esen a he mild condi ions
and he e o e only low con e sion and yield we e achie ed. All suppo s and ca alys s we e able o
p oduce highe u u al yields and con e sions han he con ol expe imen . Fi s , chemically ac i a ed
(ZnCl
2
) ca bon suppo (ACz) was es ed and i esul ed in good con e sion (82%, Table 4, en y 2). The
u u al yield was also conside ably high (28%). The high con e sion is mos likely connec ed o he
high zinc con en o ACz (Table 1), which o igina ed om chemical ac i a ion. I on was imp egna ed
o he suppo using an FeNO
3
p ecu so and FeCl
3
p ecu so (Table 4, en ies 3
¨
C4, espec i ely).
NO
3
-based sal s a e commonly a o ed as p ecu so s since hey lea e no esidue on he ca alys [
46
,
47
].
Wi h a ni a e p ecu so , con e sion inc eased u he ( om 82 o 91%) compa ed o ACz suppo bu
he yield dec eased om 28 o 23% (Table 4, en ies 2–3). The e o e, he eac ion selec i i y owa d
u u al also dec eased ( om 36 o 27%). Wi h a chlo ide p ecu so , he yield inc eased ( om 28
o 32%) compa ed o he ACz suppo and he eac ion selec i i y also inc eased ( om 36 o 51%,
Table 4, en ies 2 and 4). Based on he highe u u al yield and eac ion selec i i y, he FeCl
3
p ecu so
was de e mined o be mo e sui able han FeNO
3
and was used in u he ca alys s. Simila esul s
we e ob ained by Cha eonlimkun e al., who disco e ed ha chlo ide-based p ecu so s esul ed in
highe eac i i y compa ed o ni a e-based p ecu so s in Z O and TiO ca alys s [
14
]. As men ioned in
Sec ion 2.1.2, chlo ide ions ha e been shown o enhance u u al o ma ion om xylose by a o ing
he 1,2-enediol o ma ion be o e dehyd a ion [
34
]. This is mos p obably eason why chlo ine-based
p ecu so s inc ease u u al yield and eac ion selec i i y compa ed o ni a e-based p ecu so s.
Ca alys s 2020,10, 821 15 o 20
wi h a Zeiss Sigma Field emission scanning elec on mic oscope (FESEM). In he sample p epa a ion, a
powde sample was placed on a conduc i e glue ape. The SEM images we e aken a a ol age o 5 kV
and a wo king dis ance a ound 5 mm.
Ca alys su ace acidi y was cha ac e ized by applying he Boehm i a ion me hod [
59
–
63
]. A
o al o 100 mg o ca alys was weighed and mixed wi h 50 mL 0.01 M NaOH. Samples we e shaken
(300 pm) in sealed ubes o 72 h a oom empe a u e and hen il e ed using a sy inge and sy inge
il e (0.45
µ
m, egene a ed cellulose). Ti a ion was ca ied ou using a back- i a ion me hod by
aking 10 mL o il a e, mixing i wi h 20 mL o 0.01 M HCl and inally back- i a ing wi h 0.01 M
NaOH. Acidic g oups we e calcula ed using Equa ion (2), based on he heo y ha NaOH neu alizes
all acidic oxygen g oups (including phenols, lac onic g oups and ca boxylic acids) p esen on ca bon.
Nonconsumed base con en was neu alized wi h acid and hen nonconsumed acid was quan i ied
h ough simple acid-base i a ion.
3.4. Fu u al P oduc ion om Xylose
In a con e sion eac ion, 0.25 mmol (37.6 mg) o xylose and 0.0036/0.050 mmol o homogeneous
me al sal (AlCl
3·
6H
2
O, ZnCl
2
, C Cl
3·
6H
2
O, SnCl
2·
2H
2
O o FeCl
3·
6H
2
O) o 5 mg he e ogeneous
ca bon-based ca alys we e placed in o a 5 mL eac ion ube. A magne ic s i ing ba , wa e (1 mL) and
MIBK (3 mL) we e added and he ube was sealed. The eac ion was ca ied ou in a Bio age Ini ia o
mic owa e eac o (Bio age, Uppsala, Sweden) a 160/170/180
◦
C o 30 min o 3.5 h. A e he eac ion,
app oxima e 1 mL samples om bo h laye s we e il e ed wi h a sy inge il e (an RC il e o he
o ganic laye and a PTFE il e o he wa e laye ) and hen analyzed wi h HPLC.
3.5. Ca alys Recycling
Ca alys ecycling expe imen s we e ca ied ou wi h 5Fe-ACs ca alys wi h a 3 h eac ion ime and
a eac ion empe a u e o 170
◦
C. A e he eac ion, liquid samples we e aken no mally o u u al
and xylose analyses. In addi ion, me al leaching (Fe and Zn) was moni o ed by measu ing he me al
con en o he wa e phase by AAS (see Sec ion 3.6) and he ca alys was collec ed using a PALL Easy
P essu e Sy inge Fil e Holde and hyd ophilic polyp opylene memb ane (GHP). The ca alys was i s
washed wi h me hanol (4 +10 mL) and wa e (3*10 mL) and hen d ied and weighed. Used ca alys s
we e analyzed wi h SEM and TEM (see Sec ion 3.3).
3.6. Analy ical Me hods o Con e sion S udies
Two di e en HPLC analyses we e used o de ec u u al and xylose in he samples. In he
analyses, calib a ions we e pe o med wi h comme cial u u al o xylose. HPLC analysis o u u al
was ca ied ou using a Wa e s 2695 sepa a ion module i ed wi h an A lan is T3 (3
µ
m, 4.6
×
150
mm) column and a Wa e s 996 pho odiode a ay (PDA) de ec o (Wa e s Co p., Mil o d, MA, USA).
A mix u e o wa e (0.1% TFA) and me hanol (0.1% TFA) (90:10) was used as he mobile phase, wi h
a low a e o 1 mL/min. The column empe a u e was kep cons an a 30
◦
C and he UV de ec ion
o u u al was pe o med a 277 nm. HPLC analysis o xylose was ca ied ou using a Shimadzu
LC-20AT liquid ch oma og aph ins umen i ed wi h an SIL-20A TH au osample , RID-20A e ac i e
index de ec o , SUGAR SH-G p e-column and Shodex SUGAR SH1821 column (8.0
×
300 mm). Sul u ic
acid (5 mM) was used as a mobile phase wi h a low a e o 0.8 mL/min and he column empe a u e
was kep cons an a 60 ◦C.
A omic abso p ion spec oscopy (AAS) was used o de e mine i on and zinc leaching om
5Fe-ACs and 5Fe-ACz ca alys s. Fi s , wa e phase samples o he eac ions we e dilu ed wi h wa e o
a minimum 10 mL. Then, de e mina ions we e made using Va ian AA240FS equipmen (Va ian Inc.,
Palo Al o, CA, USA), ai -ace ylene uel, a Va ian Spec AA lamp (Cu/Mn/Zn/Fe) and lame emission
wa eleng hs o 372.0 nm o Fe and 213.9 nm o Zn.
Ca alys s 2020,10, 821 16 o 20
3.7. Equa ions
The pa i ioning coe icien (P) was calcula ed using he ollowing o mula:
P=[ u u al]o g/[ u u al]aq, (1)
whe e [ u u al]
o g
is he concen a ion (g/l) o u u al in he o ganic laye and [ u u al]
aq
is he
concen a ion o u u al in he wa e laye .
The o al amoun o acid si es acco ding o Boehm i a ion was calcula ed as ollows:
n o al acids =[(CNaOH*VNaOH added - (CHCl*VHCl added - CNaOH*VNaOH i a ion)/1
5)-n o al acids in e e ence]/m, (2)
whe e c (NaOH and HCl) a e concen a ions in mol/L, V (NaOH and HCl) a e added olumes in mL, V
(NaOH i a ion) is he olume o NaOH in ml needed o achie e equilib ium in i a ion, m is he
mass o ca bon weighed and n ( o al acids in e e ence) ep esen s a blank solu ion wi hou ca bon.
The ac o 1
5is due o he measu emen o he 10 mL aliquo s ep esen ing 1
5o he eac ion base.
The yield o u u al was calcula ed as ollows:
Y u u al (%) =[c u meas o g/c u max]×100%, (3)
whe e c
u . meas. o g
is he measu ed u u al concen a ion in he o ganic phase o he sample and c
u max is he heo e ical maximum concen a ion o u u al in he sample.
The con e sion o xylose was calcula ed as ollows:
Cxylose (%) =[n xyl ini ial/nxyl inal]×100%, (4)
whe e n
xyl ini ial
is he ini ial amoun o xylose (in moles) ed o he eac ion and n
xyl inal
is he amoun
o xylose le in he eac ion mix u e a e he eac ion.
The selec i i y o he xylose o u u al con e sion was calcula ed as ollows:
S (%) =[(c u meas o al/c u max)/con e sion] ×100%, (5)
whe e c
u . meas o al.
is he measu ed o al u u al concen a ion in he o ganic and aqueous phases o
he sample and c u max is he heo e ical maximum concen a ion o u u al in he sample.
4. Conclusions
In his s udy, he con e sion o xylose o u u al was s udied using lignin-based ac i a ed
ca bon-suppo ed i on oxide ca alys s. Th ee di e en ac i a ed ca bon suppo s and i e di e en
ca alys s we e p epa ed and s udied in u u al p oduc ion. Di e en ac i a ion me hods, me al
p ecu so s and me al concen a ions we e used o he ca alys s and di e en empe a u es and eac ion
imes we e s udied in he con e sion eac ions. Chemical ac i a ion esul ed in a highe su ace a ea
and po e olume han physical ac i a ion bu in con e sion eac ions, physically ac i a ed ca alys s
p oduced be e eac ion selec i i y. FeNO
3
p ecu so yielded highe xylose con e sion han FeCl
3
p ecu so bu he u u al yield and selec i i y we e highe wi h FeCl
3
p ecu so . The bes esul s
o xylose con e sion o u u al we e achie ed wi h a 4 w % i on-con aining ca alys (5Fe-ACs),
which p oduced a 57% yield, 92% con e sion and 65% selec i i y a 170
◦
C in 3 h. The esul s wi h a
ca alys con aining mo e i on (9.2 w %) we e lowe (54% yield, 93% con e sion and 60% selec i i y) in
simila condi ions. The ca aly ic amoun o Fe in 5Fe-ACs was only 3.6
µ
mol and using his amoun o
homogeneous FeCl
3
as a ca alys , educed he u u al yield, xylose con e sion and selec i i y. Based
on ca alys cha ac e iza ion, i on was in he o m o i on oxide on he su ace o he he e ogeneous
ca alys , which may ha e a ec ed o i s ca aly ic ac i i y posi i ely compa ed o FeCl
3
. Mo eo e ,
hyd oxyl g oups we e de ec ed on he su ace o 5Fe-ACs, which inc eases ca alys B øns ed acid
Ca alys s 2020,10, 821 17 o 20
si es and he e o e can inc ease u u al p oduc ion. The ecycling expe imen s e ealed ha pa
o he i on is easily leached ou o he ca alys a a high empe a u e and in acidic condi ions and
he ca alys adso bed some eac ions p oduc s. These ac o s dec eased he u u al yield and xylose
con e sion a e he i s ound o ecycling bu hen hey emained cons an . Al hough he ac i a ed
ca bon-suppo ed i on oxide ca alys needs some imp o emen s o be e s abili y, i is a easible
al e na i e o homogeneous FeCl3.
Supplemen a y Ma e ials:
The ollowing a e a ailable online a h p://www.mdpi.com/2073-4344/10/8/821/s1,
Table S1: Me al analysis om ACs by ICP-OES, Table S2: XPS esul s o ACs, 5-Fe-ACs and 10Fe-ACs, Figu e S1:
XRD esul s o 5Fe-ACs and 10Fe-ACs, Figu e S2: Boehm i a ion cu es, Figu e S3: EDS spec a o a ea shown in
Figu e 4, Figu e S4: G aphical p esen a ion o he esul s p esen ed in Table 5, Figu e S5: HPLC ch oma og am o
wa e (a) and o ganic (b) phase o eac ion solu ion using 5Fe-ACz ca alys . G ams show inc easing side p oduc
peak a 3.1 min and u u al shoulde a 8.5 min, when 180
◦
C was used as eac ion empe a u e, Figu e S6: STEM
HAADF image o h ee imes used 5Fe-ACs, which shows la ge agglome a ed i on pa icles (diame e app ox.
15
¨
C40 nm) as well as small single pa icles (diame e app ox. 5 nm), Figu e S7: SEM images o unused 5Fe-ACs
(a,c) and used 5Fe-ACs (b,d) ca alys s.
Au ho Con ibu ions:
Concep ualiza ion, A.R.; me hodology, A.R. and R.K.; o mal analysis, A.R., R.K. and T.H.;
in es iga ion, A.R. and R.K.; da a cu a ion, A.R.; w i ing—o iginal d a p epa a ion, A.R.; w i ing— e iew and
edi ing, K.L., R.K., J.K., T.H. and U.L.; isualiza ion, A.R.; supe ision, K.L., J.K. and U.L.; unding acquisi ion,
A.R., U.L. and K.L. All au ho s ha e ead and ag eed o he published e sion o he manusc ip .
Funding:
This esea ch was unded by Fo um Founda ion, g an numbe s 201800022 and 20190005, he
EU/In e eg Bo nia-A lan ica, g an numbe 20201508, he Founda ion o Tauno Tönning, g an numbe 20190154
and Nessling Founda ion, g an numbe 201800070.
Acknowledgmen s:
Sa i Tuikkanen is acknowledged o comple ing pa o HPLC measu emen s and Riina
Hemmilä o AAS-measu emen s.
Con lic s o In e es : The au ho s decla e no con lic o in e es .
Re e ences
1.
We py, T.; Pe e sen, G. Top Value Added Chemicals om Biomass: Volume I–Resul s o Sc eening o Po en ial
Candida es om Suga s and Syn hesis Gas; U.S. Depa men o Ene gy: Washing on, DC, USA, 2004. [C ossRe ]
2.
Ko ke, R.H. Fu an de i a i es. In Encyclopedia o Chemical Technology; O hme , K., Ed.; John Wiley & Sons,
Inc.: New Yo k, NY, USA, 2000; Volume 12, pp. 259–286. [C ossRe ]
3.
Eseyin, A.E.; S eele, P.H. An o e iew o he applica ions o u u al and i s de i a i es. In . J. Ad . Chem.
2015,3, 42–47. [C ossRe ]
4. Win, D. Fu u al–Gold om Ga bage. Au J. Technol. 2005,8, 185–190.
5.
Zei sch, K. The Chemis y and Technology o Fu u al and i s Many By–P oduc s; Else ie : Ams e dam, The
Ne he lands, 2000; pp. 36–74.
6.
Dash ban, M.; Gilbe , A.; Fa ehi, P. P oduc ion o u u al: O e iew and challenges. J. Sci. Technol. Fo .
P od. P ocess 2012,2, 44–53.
7.
Yemis, O.; Mazza, G. Ca aly ic Pe o mances o Va ious Solid Ca alys s and Me al Halides o
Mic owa e–Assis ed Hyd o he mal Con e sion o Xylose, Xylan, and S aw o Fu u al. Was e Biomass
Valo i. 2019,10, 1343–1353. [C ossRe ]
8.
Le Guenic, S.; Delbecq, F.; Ceballos, C.; Len, C. Mic owa e–assis ed dehyd a ion o D–xylose in o u u al
by dilu ed inexpensi e ino ganic sal s solu ion in a biphasic sys em. J. Mol. Ca al. A Chem.
2015
,410, 1–7.
[C ossRe ]
9.
Liu, C.; Wyman, C.E. The enhancemen o xylose monome and xylo iose deg ada ion by ino ganic sal s in
aqueous solu ions a 180 ◦C. Ca bohyd . Res. 2006,341, 2550–2556. [C ossRe ]
10.
E sho a, O.; Nieminen, K.; Six a, H. The Role o Va ious Chlo ides on Xylose Con e sion o Fu u al:
Expe imen s and Kine ic Modeling. Chem. Ca . Chem. 2017,9, 3031–3040. [C ossRe ]
11.
Romo, J.E.; Bolla , N.V.; Zimme mann, C.J.; We s ein, S.G. Con e sion o Suga s and Biomass o Fu ans Using
He e ogeneous Ca alys s in Biphasic Sol en Sys ems. Chem. Ca . Chem. 2018,10, 4819–4830. [C ossRe ]
Ca alys s 2020,10, 821 18 o 20
12.
Da Cos a Lopez, A.M.; Mo ais, A.R.C.; Łukasik, R.M.; Fang, Z.; Smi h, R.L., J .; Qi, X. Sus ainable Ca aly ic
S a egies o C5-Suga s and Biomass Hemicellulose Con e sion Towa ds Fu u al P oduc ion. In P oduc ion
o Pla o m Chemicals om Sus ainable Resou ces; Sp inge Na u e: Singapo e, 2017; Volume 7, pp. 45–80.
[C ossRe ]
13.
Zhang, J.; Lin, L.; Liu, S. E icien P oduc ion o Fu an De i a i es om a Suga Mix u e by Ca aly ic. P ocess.
Ene g. Fuel 2012,26, 4560–4567. [C ossRe ]
14.
Cha eonlimkun, A.; Champ eda, V.; Sho ip uk, A.; Laosi ipojana, N. Reac ions o C5 and C6-suga s, cellulose,
and lignocellulose unde ho comp essed wa e (HCW) in he p esence o he e ogeneous acid ca alys s. Fuel
2010,89, 2873–2880. [C ossRe ]
15.
Shi, X.; Wu, Y.; Li, P.; Yi, H.; Yang, M.; Wang, G. Ca aly ic con e sion o xylose o u u al o e he solid acid
SO42-/Z O2–Al2O3/SBA-15 ca alys s. Ca bohyd . Res. 2011,346, 480–487. [C ossRe ]
16.
Dias, A.S.; Pillinge , M.; Valen e, A.A. Mesopo ous silica–suppo ed 12– ungs ophospho ic acid ca alys s o
he liquid phase dehyd a ion o d-xylose. Mic opo . Mesopo . Ma e . 2006,94, 214–225. [C ossRe ]
17.
Weinga en, R.; Cho, J.; Conne , W.C., J .; Hube , G.W. Kine ics o u u al p oduc ion by dehyd a ion o
xylose in a biphasic eac o wi h mic owa e hea ing. G een Chem. 2010,12, 1423–1429. [C ossRe ]
18.
Choudha y, V.; Pina , A.B.; Sandle , S.I.; Vlachos, D.G.; Lobo, R.F. Xylose Isome iza ion o Xylulose and i s
Dehyd a ion o Fu u al in Aqueous Media. ACS Ca al. 2011,1, 1724–1728. [C ossRe ]
19.
Choudha y, V.; Sandle , S.I.; Vlachos, D.G. Con e sion o Xylose o Fu u al Using Lewis and B øns ed Acid
Ca alys s in Aqueous Media. ACS Ca al. 2012,2, 2022–2028. [C ossRe ]
20.
Rod
í
guez-Reinoso, F. The ole o ca bon ma e ials in he e ogeneous ca alysis. Ca bon
1998
,36, 159–175.
[C ossRe ]
21.
Rod
í
guez-Reinoso, F.; Molina-Sabio, M. Ac i a ed ca bons om lignocellulosic ma e ials by chemical and/o
physical ac i a ion: An o e iew. Ca bon 1992,30, 1111–1118. [C ossRe ]
22.
Mazzo a, M.G.; Gup a, D.; Saha, B.; Pa a, A.K.; Bhaumik, A.; Abu-Oma , M.M. E icien Solid Acid Ca alys
Con aining Lewis and B øns ed Acid Si es o he P oduc ion o Fu u als. ChemSusChem
2014
,7, 2342–2350.
[C ossRe ]
23.
Russo, P.A.; Lima, S.; Rebu ini, V.; Pillinge , M.; Willinge , M.; Pinna, N.; Valen e, A.A. Mic owa e-assis ed
coa ing o ca bon nanos uc u es wi h i anium dioxide o he ca aly ic dehyd a ion o d-xylose in o u u al.
RSC Ad . 2013,3, 2595–2603. [C ossRe ]
24.
Ba oso-Bogea , A.; Alexand e-F anco, M.; Fe n
á
ndez-Gonz
á
lez, C.; G
ó
mez–Se ano, V. P epa a ion o
ac i a ed ca bon-me al oxide hyb id ca alys s: Tex u al cha ac e iza ion. Fuel P ocess. Technol.
2014
,126,
95–103. [C ossRe ]
25.
Ba oso-Bogea , A.; Alexand e-F anco, M.; Fe n
á
ndez-Gonz
á
lez, C.; G
ó
mez–Se ano, V. P epa a ion and
Mic os uc u al Cha ac e iza ion o Ac i a ed Ca bon–Me al Oxide Hyb id Ca alys s: New Insigh s in o
Reac ion Pa hs. J. Ma e . Sci. Technol. 2015,31, 806–814. [C ossRe ]
26.
Ba oso-Bogea , A.; Alexand e-F anco, M.; Fe n
á
ndez-Gonz
á
lez, C.; Mac
í
as-Ga c
í
a, A.; G
ó
mez-Se ano, V.
P epa a ion o Ac i a ed Ca bon-SnO
2
, TiO
2
, and WO
3
Ca alys s. S udy by FT-IR Spec oscopy. Ind. Eng.
Chem. Res. 2016,55, 5200–5206. [C ossRe ]
27.
Mi al, A.; Black, S.K.; Vinzan , T.B.; O’B ien, M.; Tucke , M.P.; Johnson, D.K. P oduc ion o Fu u al om
P ocess-Rele an Biomass-De i ed Pen oses in a Biphasic Reac ion Sys em. ACS Sus ain. Chem. Eng.
2017
,5,
5694–5701. [C ossRe ]
28.
Mo eau, C.; Du and, R.; Pey on, D.; Duhame , J.; Ri alie , P. Selec i e p epa a ion o u u al om xylose
o e mic opo ous solid acid ca alys s. Ind. C ops P od. 1998,7, 95–99. [C ossRe ]
29.
B ouwe , T.; Blahusiak, M.; Babic, K.; Schuu , B. Reac i e ex ac ion and eco e y o le ulinic acid, o mic
acid and u u al om aqueous solu ions con aining sulphu ic acid. Sep. Pu i . Technol.
2017
,185, 186–195.
[C ossRe ]
30.
Zhang, L.; Yu, H.; Wang, P.; Li, Y. P oduc ion o u u al om xylose, xylan and co ncob in
gamma- ale olac one using FeCl3·6H2O as ca alys . Bio esou . Technol. 2014,151, 355–360. [C ossRe ]
31.
Vom S ein, T.; G ande, P.M.; Lei ne , W.; Dominguez de Ma ia, P. I on-Ca alyzed Fu u al P oduc ion
in Biobased Biphasic Sys ems: F om Pu e Suga s o Di ec Use o C ude Xylose E luen s as Feeds ock.
ChemSusChem 2011,4, 1592–1594. [C ossRe ]
32.
Zhang, Y.; Chen, M.; Wang, J.; Hu, Q. Fu u al p oduc ion om dehyd a ion o xylose ca alyzed by ch omium
chlo ide in biphasic sys em. Chem. Eng. 2014,3, 54–58. [C ossRe ]
Ca alys s 2020,10, 821 19 o 20
33.
Yang, Y.; Hu, C.; Abu–Oma , M.M. Con e sion o ca bohyd a es and lignocellulosic biomass in o
5–hyd oxyme hyl u u al using AlCl
3·
6H
2
O ca alys in a biphasic sol en sys em. G een Chem.
2012
,
14, 509–513. [C ossRe ]
34.
Ma co ullio, G.; Jong, W.D. Chlo ide ions enhance u u al o ma ion om D–xylose in dilu e aqueous acidic
solu ions. G een Chem. 2010,12, 1739–1746. [C ossRe ]
35.
Va ila, T.; Be gna, D.; Lah i, R.; Roma , H.; Hu, T.; Lassi, U. Ac i a ed ca bon p oduc ion om pea using
ZnCl2: Cha ac e iza ion and applica ions. Bio esou ces 2017,12, 8078–8092. [C ossRe ]
36.
Lah i, R.; Be gna, D.; Roma , H.; Tuu ila, T.; Hu, T.; Lassi, U. Physico–chemical p ope ies and use o was e
biomass-de i ed ac i a ed ca bons. Chem. Eng. T ans. 2017,57. [C ossRe ]
37.
B ion, D. E ude pa spec oscopie de pho oelec ons de la deg ada ion supe icielle de FeS2, CuFeS2, ZnS e
PbS a l’ai e dans l’eau. Appl. Su . Sci. 1980,5, 133–152. [C ossRe ]
38.
Moulde , J.F.; S ickle, W.F.; Sobol, P.E.; Bomben, K.D. Handbook o X- ay Pho oelec on Spec oscopy; Pe kin-Elme
Co po a ion: Eden P ai ie, MN, USA, 1992; p. 45.
39.
Li, J.; Liu, J.; Zhou, H.; Fu, Y. Ca aly ic T ans e Hyd ogena ion o Fu u al o Fu u yl Alcohol o e
Ni ogen-Doped Ca bon-Suppo ed I on Ca alys s. ChemSusChem 2016,9, 1339–1347. [C ossRe ]
40.
Ba oso–Bogea , A.; Alexand e-F anco, M.; Fe n
á
ndez-Gonz
á
lez, C.; G
ó
mez-Se ano, V. Ac i a ed ca bon
su ace chemis y: Changes upon imp egna ion wi h Al(III), Fe(III) and Zn(II)-me al oxide ca alys p ecu so s
om NO3−aqueous solu ions. A ab. J. Chem. 2019,12, 3963–3976. [C ossRe ]
41.
Xiong, Y.; Tong, Q.; Shan, W.; Xing, Z.; Wang, Y.; Wen, S.; Lou, Z. A senic ans o ma ion and adso p ion
by i on hyd oxide/manganese dioxide doped s aw ac i a ed ca bon. Appl. Su . Sci.
2017
,416, 618–627.
[C ossRe ]
42.
Pakuła, M.; Biniak, S.; ´
Swia kowski, A. Chemical and Elec ochemical S udies o In e ac ions be ween I on(III)
Ions and an Ac i a ed Ca bon Su ace. Langmui 1998,14, 3082–3089. [C ossRe ]
43.
Chang, Q.; Lin, W.; Ying, W. P epa a ion o i on-imp egna ed g anula ac i a ed ca bon o a senic emo al
om d inking wa e . J. Haza d. Ma e . 2010,184, 515–522. [C ossRe ]
44.
Dallinge , D.; Kappe, C.O. Mic owa e–Assis ed Syn hesis in Wa e as Sol en . Chem. Re .
2007
,107,
2563–2591. [C ossRe ]
45.
An al, M.J.; Leesomboon, T.; Mok, W.S.; Richa ds, G.N. Mechanism o o ma ion o 2- u aldehyde om
d-xylose. Ca bohyd . Res. 1991,217, 71–85. [C ossRe ]
46.
Hu chings, G.J.; V
é
d ine, J.C. He e ogeneous Ca alys P epa a ion. In Basic P inciples in Applied Ca alysis;
Bae ns, M., Ed.; Sp inge : Be lin, Ge many, 2004; Volume 75, pp. 215–257. [C ossRe ]
47.
Jün gen, H. Ac i a ed ca bon as ca alys suppo : A e iew o new esea ch esul s. Fuel
1986
,65, 1436–1446.
[C ossRe ]
48.
Ma a o -Mey al, Y.I.; Shein uch, M. Ca aly ic Aba emen o Wa e Pollu an s. Ind. Eng. Chem. Res.
1998
,37,
309–326. [C ossRe ]
49.
K aeme , S.M. I on oxide dissolu ion and solubili y in he p esence o side opho es. Aqua . Sci.
2004
,66,
3–18. [C ossRe ]
50.
Cheng, B.; Wang, X.; Lin, Q.; Zhang, X.; Meng, L.; Sun, R.; Xin, F.; Ren, J. New Unde s andings o he
Rela ionship and Ini ial Fo ma ion Mechanism o Pseudo–lignin, Humins, and Acid–Induced Hyd o he mal
Ca bon. J. Ag ic. Food Chem. 2018,66, 11981–11989. [C ossRe ] [PubMed]
51.
Van Zand oo , I.; Wang, Y.; Ras end a, C.B.; an Eck, E.R.H.; B uijnincx, P.C.A.; Hee es, H.J.;
Weckhuysen, B.M. Fo ma ion, Molecula S uc u e, and Mo phology o Humins in Biomass Con e sion:
In luence o Feeds ock and P ocessing Condi ions. ChemSusChem
2013
,6, 1745–1758. [C ossRe ] [PubMed]
52.
Sume skii, I.V.; K u o , S.M.; Za ubin, M.Y. Humin-like subs ances o med unde he condi ions o indus ial
hyd olysis o wood. Russ. J. Appl. Chem. 2010,83, 320–327. [C ossRe ]
53. Ba holomew, C.H. Mechanisms o ca alys deac i a ion. Appl. Ca al. A Gen. 2001,212, 17–60. [C ossRe ]
54.
Dias, A.S.; Pillinge , M.; Valen e, A.A. Dehyd a ion o xylose in o u u al o e mic o–mesopo ous sul onic
acid ca alys s. J. Ca al. 2005,229, 414–423. [C ossRe ]
55.
Jeong, G.H.; Kim, E.G.; Kim, S.B.; Pa k, E.D.; Kim, S.W. Fab ica ion o sul onic acid modi ied mesopo ous
silica shells and hei ca aly ic pe o mance wi h dehyd a ion eac ion o d–xylose in o u u al. Mic opo .
Mesopo . Ma e . 2011,144, 134–139. [C ossRe ]
56.
B unaue , S.; Emme , P.H.; Telle , E. Adso p ion o Gases in Mul imolecula Laye s. J. Am. Chem. Soc.
1938
,
60, 309–319. [C ossRe ]
Ca alys s 2020,10, 821 20 o 20
57.
Sea on, N.A.; Wal on, J.P.R.B.; Qui ke, N. A new analysis me hod o he de e mina ion o he po e size
dis ibu ion o po ous ca bons om ni ogen adso p ion measu emen s. Ca bon
1989
,27, 853–861. [C ossRe ]
58.
U.S. EPA. Me hod 3051A (SW–846): Mic owa e Assis ed Acid Diges ion o Sedimen s, Sludges, and Oils; U.S. EPA:
Washing on, DC, USA, 2007.
59.
Boehm, H.P. Chemical Iden i ica ion o Su ace G oups. In Ad ances in Ca alysis; Eley, D.D., Pines, H.,
Weisz, P.B., Eds.; Else ie Academic P ess: San Diego, CA, USA, 1966; Volume 16, pp. 179–274. [C ossRe ]
60.
Boehm, H.P. Some aspec s o he su ace chemis y o ca bon blacks and o he ca bons. Ca bon
1994
,32,
759–769. [C ossRe ]
61.
Schönhe , J.; Buchheim, J.R.; Scholz, P.; Adelhelm, P. Boehm i a ion e isi ed (pa i): P ac ical aspec s o
achie ing a high p ecision in quan i ying oxygen-con aining su ace g oups on ca bon ma e ials. Ca bon
2018,4, 21. [C ossRe ]
62.
Goe zen, S.L.; The iaul , K.D.; Oickle, A.M.; Ta asuk, A.C.; And eas, H.A. S anda diza ion o he Boehm
i a ion. Pa , I. CO2expulsion and endpoin de e mina ion. Ca bon 2010,48, 1252. [C ossRe ]
63.
Oickle, A.M.; Goe zen, S.L.; Hoppe , K.R.; Abdalla, Y.O.; And eas, H.A. S anda diza ion o he Boehm
i a ion: Pa II. Me hod o agi a ion, e ec o il e ing and dilu e i an . Ca bon
2010
,48, 3313–3322.
[C ossRe ]
©
2020 by he au ho s. Licensee MDPI, Basel, Swi ze land. This a icle is an open access
a icle dis ibu ed unde he e ms and condi ions o he C ea i e Commons A ibu ion
(CC BY) license (h p://c ea i ecommons.o g/licenses/by/4.0/).