ma e ials
A icle
Ca bon Dioxide Hyd ogena ion by Means o
Plasmonic Resonance Ac i a ion in Silica
Ae ogel Media
Se gio Muñoz 1, Alexande Na a e e 2,*, Ángel Ma ín1, Roland Di meye 2
and M. JoséCoce o 1
1BioEcoUVa, Bioeconomy Resea ch Ins i u e, High P essu e P ocesses G oup, Depa men o Chemical
Enginee ing and En i onmen al Technology, Uni e si y o Valladolid, C/P ado de la Magdalena s/n,
47011 Valladolid, Spain; [email p o ec ed] (S.M.); [email p o ec ed] (Á.M.); mjcoce [email p o ec ed] (M.J.C.)
2Ins i u e o Mic o P ocess Enginee ing, Ka ls uhe Ins i u e o Technology (KIT),
He mann- on-Helmhol z-Pla z 1, 76344 Eggens ein-Leopoldsha en, Ge many; [email p o ec ed]
*Co espondence: alexande [email p o ec ed]; Tel.: +49-721-608-26653
Recei ed: 29 Sep embe 2018; Accep ed: 25 Oc obe 2018; Published: 30 Oc obe 2018


Abs ac :
Su ace Plasmon Resonance can be used o ac i a e zinc oxide/coppe ca alys s in
o de o pe o m he ca bon dioxide hyd ogena ion eac ion by means o ligh ene gy, a oiding
high- empe a u e p ocesses. The syn hesis and imp egna ion me hods ha e been designed o ill
glass mic o eac o s wi h ZnO/Cu nanopa icles suppo ed on anspa en silica ae ogels o maximize
he ligh abso bed by he ca alys . A LED de ice su ounding he glass mic o eac o s p o ided whi e
ligh o ac i a e he ca alys homogeneously h oughou he eac o . Tempe a u e, p essu e, amoun o
ca alys and gases low we e s udied as possible a iables o enhance he p ocess ying o maximize
CO
2
con e sion a es, achie ing he bes esul s wo king a high p essu es. The use o anspa en
SiO
2
Ae ogels as suppo s o pho oca aly ic gas phase eac ions e en unde high-p essu e condi ions
is demons a ed.
Keywo ds: CO2hyd ogena ion; pho oca alysis; su ace plasmon esonance; sola uels
1. In oduc ion
Ca bon dioxide con e sion in o use ul p oduc s is a ac ing much a en ion in he las yea s as a
way o educe he g eenhouse e ec and i s d ama ic consequences o he plane [
1
,
2
]. Among he
op ions o pe o m his CO
2
con e sion, pho oca aly ic p ocesses using sola ene gy ep esen an
in e es ing way o ans o m CO2due o he use o a enewable ene gy. CO2can be ans o med in o
me hanol, which is a chemical commodi y and can s o e he ene gy om he sun [
3
,
4
]. T ans o ming
CO
2
in o me hanol using sola ene gy gene a es, he e o e, wo bene i s a he same ime: he educ ion
o CO2emissions and he s o age o ene gy om enewable sou ces.
Pho oca aly ic CO
2
ans o ma ion usually shows low con e sions because he e iciency o he
p ocess is limi ed [
5
]. One o he easons is ela ed o he low amoun o ligh ene gy ha mos o he
ca alys s abso b in he isible o ul a iole band. Su ace Plasmon Resonance (SPR) is a phenomenon
exhibi ed by some me allic nanos uc u es, which esul s in a highe ligh ene gy abso bance ha
can be con olled wi h he shape and size o he nanopa icles. One a ea whe e he SPR e ec has
ound applica ion is in biosensing [
6
,
7
]. Howe e , in he las ew yea s, i s applica ion has ex ended
o pho oca alysis [
8
], which is he main opic o his pape . Commonly, gold and sil e nanopa icles
ha e been s udied o enhance SPR e ec [
9
,
10
], al hough o he me als such as coppe also p oduce
his e ec [
11
]. Coppe has been widely used o ca bon dioxide hyd ogena ion in indus y combined
wi h he use o zinc oxide in o de o p oduce me hanol wi h high selec i i y [
12
]. Fo his eason,
Ma e ials 2018,11, 2134; doi:10.3390/ma11112134 www.mdpi.com/jou nal/ma e ials
Ma e ials 2018,11, 2134 2 o 13
a combina ion o bo h ma e ials Cu/ZnO could be also used o pe o m ca bon dioxide hyd ogena ion
by means o sola ene gy using SPR as a way o enhance he e iciency o he p ocess and o maximize
he CO2 ans o ma ion.
SPR e ec depends on coppe nanopa icles size, whose a ia ion can modi y he abso p ion
wa eleng h o isual ligh [
13
]. Ca bon dioxide hyd ogena ion is pe o med in he con ac a ea
be ween zinc oxide and coppe nanopa icles [
14
], gi ing a c ucial impo ance o maximize his con ac
a ea wi h he syn hesis me hod chosen. Fo his eason, he syn hesis me hod mus p oduce disc e e
coppe nanopa icles whose inal size can be con olled, and hen hese coppe nanopa icles mus be
deposi ed on o zinc oxide nano ods a oiding agglome a ion.
Ca bon dioxide and hyd ogen can p oduce me hanol di ec ly in a one-s ep eac ion. Howe e , he
con e sion is inc eased when he ans o ma ion is ca ied ou in wo s eps [
15
]. Fi s , ca bon dioxide
is ans o med in ca bon monoxide (endo he mic), and a e his CO again wi h hyd ogen p oduce
me hanol (exo he mic).
In his wo k, glass mic ochannels we e selec ed o pe o m he ca bon dioxide hyd ogena ion
because hey p o ided g ea homogenei y in he ligh dis ibu ion h ough he eac o [
16
].
Glass mic o eac o s a oided ligh abso p ion in e e ences, allowing o s udy he in luence o he
ma e ial p ope ies and eac ion condi ions on CO2con e sion.
In o de o ix he ca alys inside he mic ochannels, a suppo was equi ed. This suppo mus
p o ide high su ace a eas in o de o load he bime allic ca alys and, a he same ime, allow ha
enough ligh is ansmi ed o he ca alys o ac i a e i . Silica ae ogels a e mic o-mesopo ous ma e ials
wi h high su ace a eas (400–1500 m
2
/g) and high isible anspa ency [
17
], p ope ies ha placed
hem as he bes op ion o be used as a suppo in his wo k.
The e o e, a bime allic ca alys ZnO/Cu suppo ed in silica ae ogels can be ac i a ed wi h isual
ligh and be used o pe o m ca bon dioxide hyd ogena ion in glass mic ochannels. In a p e ious
wo k, he iabili y o he idea was es ed wi h good esul s in a se up desc ibed in [
18
], opening he
possibili y o imp o e he p ocess op imizing he eac ion pa ame e s. Wi h he pu pose o inc easing
CO
2
con e sion a e, a comple e s udy o he in luence o some eac ion a iables ( empe a u e,
p essu e, ca alys amoun and eac an s p opo ion) has been pe o med in his wo k. Wi h his wo k,
we aim o open new oppo uni ies o he in eg a ion o plasmonic pho oca alys s in u u e scalable
eac ion sys ems o p oduce use ul compounds as such me hanol using sola ene gy.
2. Ma e ials and Me hods
2.1. Reagen s
The chemicals used du ing his s age we e: Zinc ace a e dihyd a e (>98%), oleylamine (70%),
e ame hyl o hosilica e (98%), ammonia (30%), ie hylene glycol (99%) we e pu chased om
Sigma-Ald ich (S . Louis, MO, USA). E hylene glycol (99.5%) (Me ck, Kenilwo h, NJ, USA).
Coppe ace a e monohyd a e (99.9%) was pu chased om Al a Aesa (Ha e hill, MA, USA). Me hanol
(99.8%) was pu chased om Pan eac (Ba celona, Spain).
2.2. Syn hesis o Plasmonic Ca alys
Following he me hod p oposed by Tan e al. (2013) [
19
], zinc oxide nano ods we e de eloped
i s ly, and hen coppe nanopa icles we e deposi ed on o in a second s ep, con olling he inal size o
he nanopa icles and he wa eleng h whe e isual ligh was abso bed.
Oleylamine/zinc ace a e a io was he mos impo an pa ame e ha de ined he inal shape o
zinc oxide nanopa icles. In o de o p oduce nano ods, 3 mmol o zinc ace a e we e mixed wi h 1.3 mL
o oleylamine in a h ee-necked lask. A e degassing he mix u e a 80
◦
C o 45 min in a acuum
a mosphe e, a ni ogen s eam was in oduced in he sys em and he empe a u e was inc eased o
220
◦
C o 15 min. A e his, he inal p oduc ob ained was a whi e pas e s uck o he bo om o he
lask. 10 mL o e hanol we e used wi h s i ing o emo e he solid p oduc , and hen he mix u e
Ma e ials 2018,11, 2134 3 o 13
was cen i uged (cen i uge Kubo a 5100, Tokyo, Japan) a 5000 pm in o de o isola e he p ecipi a e.
The solid p oduc was washed again wice wi h 5 mL o e hanol o ensu e a pe ec emo al o he
eac an s. ZnO nano ods p epa ed we e edispe sed in 20 mL o ie hylene glycol o wo hou s by
sonica ion, and hen s i ed unde oom condi ions o e nigh .
A e ha , a dual glycol sys em was p epa ed o deposi coppe oxide nanopa icles on o he
ZnO nano ods. Coppe ace a e (1.5 mmol) was added in a second po o 4 mL o e hylene glycol,
using sonica ion o 1h o c ea e a homogeneous solu ion. E hylene glycol (2 mL) was also added o
he zinc oxide essel, and he solu ion was ans e ed o a h ee-necked lask again o be degassed a
ambien empe a u e o 10 min. Then, he coppe ace a e solu ion was placed in an addi ion unnel
connec ed o he h ee-necked lask wi h he coppe oxide, which was hea ed a 190
◦
C unde ni ogen
pu ging. Wi h he aim o a oiding agglome a ion, coppe ace a e solu ion was added slowly d op by
d op du ing 10 min. A e his, he solu ion was cooled, washed and cen i uged h ee imes wi h
isop opanol a 5000 pm. The mass o he solid ob ained was hen egis e ed.
2.3. Syn hesis and Imp egna ion o Silica Ae ogels
The silica gels we e syn hesized ollowing he me hod used by Sanz-Mo al e al. (2014) [
20
].
The p ecu so s used we e e ame hyl o hosilica e (TMOS), me hanol, wa e and ammonia in a
1:2.3:3.84:0.012 mola a io. The amoun o ammonia was he con olling pa ame e o he gela ion
ime. Depending on he ca alys load, which also a ec ed he gela ion ime, mola a io TMOS:
ammonia was a ied om 1:0.012 o 1:0.12 in o de o achie e he igh gela ion ime. We adjus ed
he amoun o silica p ecu so added in acco dance o he equi ed ca aly ic load (7.0%, 10.0%, 13.0%)
based on he mass o Cu/ZnO ca alys ob ained in he p e ious s ep.
Silica ae ogels also ep esen ed a bene i o subsequen imp egna ion in he glass mic ochannels.
Ae ogels we e p oduced h ough he sol-gel ou e, mixing all he p ecu so s in a liquid s a e unde
s i ing o gene a e he hyd ogel. These p ecu so s included me hanol, which was used o suspend
he ca alys nanopa icles by sonica ion. A e ha , me hanol was mixed wi h he o he p ecu so s
gene a ing he silica ne , while e aining he Cu/ZnO ca alys inside. This silica s uc u e was in a
liquid o m o a ew seconds, which allowed he imp egna ion in glass mic o eac o s by using jus a
sy inge. Then, he gela ion p ocess ook place di ec ly inside he mic o eac o s ob aining a hyd ogel
well a ached o he walls o he mic ochannels and a oiding p oblems o in oducing a solid ca alys in
a na ow mic ochannel. In his way, gela ion ime became he key ac o o he ca alys imp egna ion
inside he mic o eac o s. The injec ion wi h he sy inge in a liquid s a e equi ed gela ion imes no
oo sho . Howe e , i he gela ion p ocess we e oo slow, he ca alys would p ecipi a e and he inal
s uc u e would no be homogeneous.
Fi s , me hanol wi h he ca alys nanopa icles was mixed wi h TMOS in a small glass and s i ed
o a ew minu es. A he same ime, a second glass beake wi h wa e and ammonia was also s i ed
and co e ed o a oid ammonia e apo a ion. Then, bo h solu ions we e mixed and he sy inge was
used o ill he mic o eac o . A e a ew seconds, he liquid u ned in o he hyd ogel wi h he ca alys
inside showing good adhesion and homogenei y along he en i e leng h o he glass mic o eac o
(15 cm).
2.4. Ae ogels Supe c i ical D ying
Be o e ca ying ou he d ying wi h supe c i ical CO
2
, an in e media e s ep was equi ed o
aging he hyd ogels [21]. In his s age, he mic o eac o s we e placed in a essel illed wi h me hanol
and closed o one week. The aim was o emo e he wa e om he s uc u e p oducing alcogels,
which we e d ied la e wi h supe c i ical CO
2
. In o de o ensu e no wa e emained in he s uc u e,
me hanol was enewed wice du ing he week.
A e ha , he alcogels we e eady o be d ied wi h supe c i ical CO
2
. The expe imen al plan o
ae ogels d ying (Figu e 1) had an o en o hea he sys em a 45
◦
C, and CO
2
was pumped a 110 ba
Ma e ials 2018,11, 2134 4 o 13
abo e he supe c i ical poin [
20
]. Inside he o en, he e was a high-p essu e essel ha was illed
wi h me hanol be o e mic ochannels we e placed inside.
Ma e ials 2018, 11, x FOR PEER REVIEW 4 o 13
comple e sys em o 45 min. As me hanol has no a high solubili y in supe c i ical CO2, ou d ying
cycles we e pe o med, enewing CO2 be ween each cycle o ensu e a comple e d ying o me hanol.
The inal p oduc s we e he ae ogels co ec ly o med and wi h g ea adhe ence o he mic ochannel
walls.
Cooling
O en
High
P essu e
Vessel
Ven
CO
2
Pump
PI
PI
TI
Figu e 1. Silica ae ogels d ying plan .
2.5. Expe imen al Reac ion Plan
A e mic o eac o s we e p epa ed o pe o m he eac ion, hey we e ca ied o he expe imen al
se up whe e ca bon dioxide hyd ogena ion was pe o med. This plan (Figu e 2) was used in
p e ious wo k o es he eac o concep wi h good esul s [18].
H2
CO2
Mic oGC
FIC
FIC O en
LED
Ven
PIC
TIC
Figu e 2. Ca bon dioxide hyd ogena ion expe imen al plan .
The expe imen al se up was designed o be able o con ol he eac ion pa ame e s ha can ha e
an in luence on he ca bon dioxide ans o ma ion, such as p essu e, empe a u e and gas inle lows.
The plan had wo low mass me e /con olle s (EL-Flow F-200, B onkho s , The Ne he lands) o H2
and CO2 wi h anges up o 1 mL/min. The empe a u e was con olled by placing he mic o eac o s
inside a gas ch oma og aphy o en (Agilen 7890, San a Cla a, CA, USA) and he p essu e was
con olled by a p essu e me e /con olle (EL-P ess se ies, B onkho s , The Ne he lands) loca ed a
he exi o he o en.
The mic o eac o consis ed o a 0.5 mm ID glass capilla y wi h an ex e nal diame e o 5 mm
(Scho Du an, Mainz, Ge many). The mic o eac o was su ounded by a se o LEDs (Supe b igh ,
inspi ed LED), which p o ided a nominal powe o 9780 W/m2 o whi e ligh in he mos
homogeneous way o illumina e co ec ly he en i e eac o [18]. The gas ou le s eam was measu ed
wi h a Mic o Gas Ch oma og aph (CP-4900, Va ian, Palo Al o, CA, USA) equipped wi h wo
columns: a po aplo 10 m and a 5A molsie e.
Figu e 1. Silica ae ogels d ying plan .
When he high-p essu e essel was closed, he uppe al e was sligh ly opened o in oduce he
supe c i ical ca bon dioxide and o inc ease slowly he p essu e (i he e we e a sudden change
o p essu e inside he essel, he ae ogels would be b oken). A e he p essu e in he essel
eached 110 ba , he lowe al e was also opened in o de ha ca bon dioxide s a ed o ci cula e
h ough he comple e sys em o 45 min. As me hanol has no a high solubili y in supe c i ical CO
2
,
ou d ying cycles we e pe o med, enewing CO
2
be ween each cycle o ensu e a comple e d ying
o me hanol. The inal p oduc s we e he ae ogels co ec ly o med and wi h g ea adhe ence o he
mic ochannel walls.
2.5. Expe imen al Reac ion Plan
A e mic o eac o s we e p epa ed o pe o m he eac ion, hey we e ca ied o he expe imen al
se up whe e ca bon dioxide hyd ogena ion was pe o med. This plan (Figu e 2) was used in p e ious
wo k o es he eac o concep wi h good esul s [18].
Ma e ials 2018, 11, x FOR PEER REVIEW 4 o 13
comple e sys em o 45 min. As me hanol has no a high solubili y in supe c i ical CO2, ou d ying
cycles we e pe o med, enewing CO2 be ween each cycle o ensu e a comple e d ying o me hanol.
The inal p oduc s we e he ae ogels co ec ly o med and wi h g ea adhe ence o he mic ochannel
walls.
Cooling
O en
High
P essu e
Vessel
Ven
CO
2
Pump
PI
PI
TI
Figu e 1. Silica ae ogels d ying plan .
2.5. Expe imen al Reac ion Plan
A e mic o eac o s we e p epa ed o pe o m he eac ion, hey we e ca ied o he expe imen al
se up whe e ca bon dioxide hyd ogena ion was pe o med. This plan (Figu e 2) was used in
p e ious wo k o es he eac o concep wi h good esul s [18].
H2
CO2
Mic oGC
FIC
FIC O en
LED
Ven
PIC
TIC
Figu e 2. Ca bon dioxide hyd ogena ion expe imen al plan .
The expe imen al se up was designed o be able o con ol he eac ion pa ame e s ha can ha e
an in luence on he ca bon dioxide ans o ma ion, such as p essu e, empe a u e and gas inle lows.
The plan had wo low mass me e /con olle s (EL-Flow F-200, B onkho s , The Ne he lands) o H2
and CO2 wi h anges up o 1 mL/min. The empe a u e was con olled by placing he mic o eac o s
inside a gas ch oma og aphy o en (Agilen 7890, San a Cla a, CA, USA) and he p essu e was
con olled by a p essu e me e /con olle (EL-P ess se ies, B onkho s , The Ne he lands) loca ed a
he exi o he o en.
The mic o eac o consis ed o a 0.5 mm ID glass capilla y wi h an ex e nal diame e o 5 mm
(Scho Du an, Mainz, Ge many). The mic o eac o was su ounded by a se o LEDs (Supe b igh ,
inspi ed LED), which p o ided a nominal powe o 9780 W/m2 o whi e ligh in he mos
homogeneous way o illumina e co ec ly he en i e eac o [18]. The gas ou le s eam was measu ed
wi h a Mic o Gas Ch oma og aph (CP-4900, Va ian, Palo Al o, CA, USA) equipped wi h wo
columns: a po aplo 10 m and a 5A molsie e.
Figu e 2. Ca bon dioxide hyd ogena ion expe imen al plan .
The expe imen al se up was designed o be able o con ol he eac ion pa ame e s ha can
ha e an in luence on he ca bon dioxide ans o ma ion, such as p essu e, empe a u e and gas inle
lows. The plan had wo low mass me e /con olle s (EL-Flow F-200, B onkho s , The Ne he lands)
o H
2
and CO
2
wi h anges up o 1 mL/min. The empe a u e was con olled by placing he
mic o eac o s inside a gas ch oma og aphy o en (Agilen 7890, San a Cla a, CA, USA) and he
Ma e ials 2018,11, 2134 5 o 13
p essu e was con olled by a p essu e me e /con olle (EL-P ess se ies, B onkho s , The Ne he lands)
loca ed a he exi o he o en.
The mic o eac o consis ed o a 0.5 mm ID glass capilla y wi h an ex e nal diame e o 5 mm
(Scho Du an, Mainz, Ge many). The mic o eac o was su ounded by a se o LEDs (Supe b igh ,
inspi ed LED), which p o ided a nominal powe o 9780 W/m
2
o whi e ligh in he mos homogeneous
way o illumina e co ec ly he en i e eac o [
18
]. The gas ou le s eam was measu ed wi h a Mic o
Gas Ch oma og aph (CP-4900, Va ian, Palo Al o, CA, USA) equipped wi h wo columns: a po aplo
10 m and a 5A molsie e.
2.6. Ca alys Cha ac e iza ion
Ligh abso bance es s o he ZnO/Cu ca alys we e pe o med using a UV-Vis Spec opho ome e
(Shimadzu UV 2550, Shimadzu, Kyo o, Japan). T ansmission Elec on Mic oscopy was ca ied ou wi h
a JEOL JEM-2100F UHR (JEOL, Tokyo, Japan) o ca alys nanopa icles, and wi h a JEOL JEM-FS2200
HRP (JEOL, Tokyo, Japan) o silica nanocomposi es. The chemical s uc u e o he ae ogel was s udied
by Fou ie T ans o m In a ed Spec oscopy (FTIR model TENSOR om B uke , Bille ica, MA, USA).
Accele a ed Su ace A ea and Po osime y Sys ems (ASAP 2020 and 2420 om Mic ome i ics, A lan a,
GA, USA) we e used o calcula e he speci ic su ace a ea and o de e mine he ni ogen iso he mal
adso p ion—deso p ion cu e.
3. Resul s and Discussion
3.1. Ca alys Cha ac e iza ion
T ansmission Elec on Mic oscopy was used o check he co ec o ma ion o ZnO nano ods, and
o p o e ha coppe nanopa icles we e deposi ed on o nano ods su ace (Figu e 3) be o e loading he
ca alys in he ae ogel.
Ma e ials 2018, 11, x FOR PEER REVIEW 5 o 13
2.6. Ca alys Cha ac e iza ion
Ligh abso bance es s o he ZnO/Cu ca alys we e pe o med using a UV-Vis
Spec opho ome e (Shimadzu UV 2550, Shimadzu, Kyo o, Japan). T ansmission Elec on
Mic oscopy was ca ied ou wi h a JEOL JEM-2100F UHR (JEOL, Tokyo, Japan) o ca alys
nanopa icles, and wi h a JEOL JEM-FS2200 HRP (JEOL, Tokyo, Japan) o silica nanocomposi es. The
chemical s uc u e o he ae ogel was s udied by Fou ie T ans o m In a ed Spec oscopy (FTIR
model TENSOR om B uke , Bille ica, MA, USA). Accele a ed Su ace A ea and Po osime y
Sys ems (ASAP 2020 and 2420 om Mic ome i ics, A lan a, GA, USA) we e used o calcula e he
speci ic su ace a ea and o de e mine he ni ogen iso he mal adso p ion—deso p ion cu e.
3. Resul s and Discussion
3.1. Ca alys Cha ac e iza ion
T ansmission Elec on Mic oscopy was used o check he co ec o ma ion o ZnO nano ods,
and o p o e ha coppe nanopa icles we e deposi ed on o nano ods su ace (Figu e 3) be o e
loading he ca alys in he ae ogel.
Figu e 3. TEM images o ZnO nano ods (le ) and coppe nanopa icles deposi ed on o ZnO nano ods
( igh ).
These TEM images con i med ha ZnO nano ods we e co ec ly c ea ed and disc e e coppe
nanopa icles we e o med on he ZnO su ace wi hou agglome a ion as i was expec ed. Mo eo e ,
he imp egna ion o he ca alys inside he silica ne was also s udied in o de o ensu e ca alys was
p ope ly loaded (Figu e 4).
Figu e 3. TEM images o ZnO nano ods (le ) and coppe nanopa icles deposi ed on o ZnO nano ods ( igh ).
These TEM images con i med ha ZnO nano ods we e co ec ly c ea ed and disc e e coppe
nanopa icles we e o med on he ZnO su ace wi hou agglome a ion as i was expec ed. Mo eo e ,
he imp egna ion o he ca alys inside he silica ne was also s udied in o de o ensu e ca alys was
p ope ly loaded (Figu e 4).
The dis ibu ion o silica, zinc and coppe in he s uc u e was ob ained by using da k ield
echnique selec ing he e lec ions o each elemen . Silica was he main componen and appea ed
h oughou he s uc u e, while zinc oxide nanopa icles we e ound all oge he inside he silica
s uc u e. Mos o he coppe nanopa icles we e loca ed in he zinc oxide a ea, indica ing ha hey
we e co ec ly deposi ed on o ZnO nano ods. Howe e , an impo an amoun o coppe nanopa icles
we e ound away om zinc oxide nano ods gi en ha hey we e no co ec ly deposi ed, a ec ing he

Ma e ials 2018,11, 2134 6 o 13
e iciency o he p ocess. This means ha he Cu/ZnO ca alys nanos uc u e was a ec ed du ing i s
in oduc ion in o he ae ogel ne . This also sugges s he possibili y o maldis ibu ion o he ca alys
inside he ae ogel.
Ma e ials 2018, 11, x FOR PEER REVIEW 6 o 13
Figu e 4. TEM image o a silica ae ogel wi h ca alys and da k ield o iden i y silica (Si), zinc (Zn)
and coppe (Cu) dis ibu ion.
The dis ibu ion o silica, zinc and coppe in he s uc u e was ob ained by using da k ield
echnique selec ing he e lec ions o each elemen . Silica was he main componen and appea ed
h oughou he s uc u e, while zinc oxide nanopa icles we e ound all oge he inside he silica
s uc u e. Mos o he coppe nanopa icles we e loca ed in he zinc oxide a ea, indica ing ha hey
we e co ec ly deposi ed on o ZnO nano ods. Howe e , an impo an amoun o coppe
nanopa icles we e ound away om zinc oxide nano ods gi en ha hey we e no co ec ly
deposi ed, a ec ing he e iciency o he p ocess. This means ha he Cu/ZnO ca alys nanos uc u e
was a ec ed du ing i s in oduc ion in o he ae ogel ne . This also sugges s he possibili y o
maldis ibu ion o he ca alys inside he ae ogel.
Ligh abso p ion o his bime allic ca alys was es ed by UV-Vis Spec opho ome y (Figu e 5)
in o de o p o e bo h ma e ials we e abso bing ligh in he isible band (coppe ) and ul a iole
band (zinc oxide). ZnO nano ods exhibi ed a peak a 370 nm, and Cu nanopa icles induced a peak
a ound 600 nm, which co esponded o he abso p ion o he ange be ween o ange and ed colou
[22]. The b oad de ini ion o a peak was due o he a iable size o nanopa icles, mo ing in a na ow
ange o sizes bu no enough o de ine a clea peak.
Si
Cu
Z
n
Figu e 4.
TEM image o a silica ae ogel wi h ca alys and da k ield o iden i y silica (Si), zinc (Zn) and
coppe (Cu) dis ibu ion.
Ligh abso p ion o his bime allic ca alys was es ed by UV-Vis Spec opho ome y (Figu e 5)
in o de o p o e bo h ma e ials we e abso bing ligh in he isible band (coppe ) and ul a iole
band (zinc oxide). ZnO nano ods exhibi ed a peak a 370 nm, and Cu nanopa icles induced a peak
a ound 600 nm, which co esponded o he abso p ion o he ange be ween o ange and ed colou [
22
].
The b oad de ini ion o a peak was due o he a iable size o nanopa icles, mo ing in a na ow ange
o sizes bu no enough o de ine a clea peak.
FTIR analysis was used o s udy he in luence o ca alys loading on he s uc u e o he ae ogel.
Wi h his pu pose, silica ae ogel wi hou ca alys was i s analysed, in o de o ob ain he IR spec a o
he silica ma e ial and iden i y all i s chemical bonds. A e ha , wo di e en amoun s o ca alys
load (7.0 and 10.0% in weigh ) we e analysed o s udy i hei addi ion a ec ed he silica s uc u e o
he nanopa icles we e loca ed inside he po ous s uc u e wi hou c ea ing new bonds (Figu e 6).
The main peaks o he silica s uc u e we e iden i ied in he silica ae ogel spec a. The mos
in ense and b oad peak was ound a 1050 cm
−1
, wi h a shoulde a a ound 1200 cm
−1
due o Si-O-Si
asymme ic s e ching ib a ions [
23
]. The symme ic s e ching ib a ions o Si-O-Si we e obse ed a
800 cm
−1
. When he ca alys was imp egna ed in he ae ogel, no signi ican peaks appea ed on he
spec a, and he silica s uc u e peaks we e s ill co ec ly o med, concluding ha ca alys loads did
no a ec he chemical s uc u e o he ae ogel.
Ma e ials 2018,11, 2134 7 o 13
Ma e ials 2018, 11, x FOR PEER REVIEW 7 o 13
Figu e 5. UV-Vis abso p ion spec a o ZnO/Cu ca alys .
FTIR analysis was used o s udy he in luence o ca alys loading on he s uc u e o he ae ogel.
Wi h his pu pose, silica ae ogel wi hou ca alys was i s analysed, in o de o ob ain he IR spec a
o he silica ma e ial and iden i y all i s chemical bonds. A e ha , wo di e en amoun s o ca alys
load (7.0 and 10.0% in weigh ) we e analysed o s udy i hei addi ion a ec ed he silica s uc u e o
he nanopa icles we e loca ed inside he po ous s uc u e wi hou c ea ing new bonds (Figu e 6).
Figu e 6. FTIR spec a o silica ae ogel wi hou and wi h ca alys load.
The main peaks o he silica s uc u e we e iden i ied in he silica ae ogel spec a. The mos
in ense and b oad peak was ound a 1050 cm−1, wi h a shoulde a a ound 1200 cm−1 due o Si-O-Si
asymme ic s e ching ib a ions [23]. The symme ic s e ching ib a ions o Si-O-Si we e obse ed
a 800 cm−1. When he ca alys was imp egna ed in he ae ogel, no signi ican peaks appea ed on he
spec a, and he silica s uc u e peaks we e s ill co ec ly o med, concluding ha ca alys loads did
no a ec he chemical s uc u e o he ae ogel.
Same samples we e analysed o de e mine he su ace a ea and po e olume depending on
ca alys load. Bo h p ope ies we e calcula ed om N2 iso he m adso p ion—deso p ion cu e, ha
showed o all he samples a ype IV iso he m cu e ha is usual o mesopo ous silica ae ogels [24]
(Figu e 7). Almos no di e ences we e ound when 7.0% and 10.0% ca alys we e loaded in he
ae ogel.
10.0% Ca alys
7.0% Ca alys
Silica ae ogel
Figu e 5. UV-Vis abso p ion spec a o ZnO/Cu ca alys .
Ma e ials 2018, 11, x FOR PEER REVIEW 7 o 13
Figu e 5. UV-Vis abso p ion spec a o ZnO/Cu ca alys .
FTIR analysis was used o s udy he in luence o ca alys loading on he s uc u e o he ae ogel.
Wi h his pu pose, silica ae ogel wi hou ca alys was i s analysed, in o de o ob ain he IR spec a
o he silica ma e ial and iden i y all i s chemical bonds. A e ha , wo di e en amoun s o ca alys
load (7.0 and 10.0% in weigh ) we e analysed o s udy i hei addi ion a ec ed he silica s uc u e o
he nanopa icles we e loca ed inside he po ous s uc u e wi hou c ea ing new bonds (Figu e 6).
Figu e 6. FTIR spec a o silica ae ogel wi hou and wi h ca alys load.
The main peaks o he silica s uc u e we e iden i ied in he silica ae ogel spec a. The mos
in ense and b oad peak was ound a 1050 cm−1, wi h a shoulde a a ound 1200 cm−1 due o Si-O-Si
asymme ic s e ching ib a ions [23]. The symme ic s e ching ib a ions o Si-O-Si we e obse ed
a 800 cm−1. When he ca alys was imp egna ed in he ae ogel, no signi ican peaks appea ed on he
spec a, and he silica s uc u e peaks we e s ill co ec ly o med, concluding ha ca alys loads did
no a ec he chemical s uc u e o he ae ogel.
Same samples we e analysed o de e mine he su ace a ea and po e olume depending on
ca alys load. Bo h p ope ies we e calcula ed om N2 iso he m adso p ion—deso p ion cu e, ha
showed o all he samples a ype IV iso he m cu e ha is usual o mesopo ous silica ae ogels [24]
(Figu e 7). Almos no di e ences we e ound when 7.0% and 10.0% ca alys we e loaded in he
ae ogel.
10.0% Ca alys
7.0% Ca alys
Silica ae ogel
Figu e 6. FTIR spec a o silica ae ogel wi hou and wi h ca alys load.
Same samples we e analysed o de e mine he su ace a ea and po e olume depending on ca alys
load. Bo h p ope ies we e calcula ed om N
2
iso he m adso p ion—deso p ion cu e, ha showed
o all he samples a ype IV iso he m cu e ha is usual o mesopo ous silica ae ogels [
24
] (Figu e 7).
Almos no di e ences we e ound when 7.0% and 10.0% ca alys we e loaded in he ae ogel.
The BET su ace a ea calcula ed was 845.3 m
2
/g o silica ae ogels, and i was sligh ly educed o
843.4 and 842.7 m
2
/g o 7.0 and 10.0% ca alys load, espec i ely. This indica ed ha he p esence
o ca alys did no ha e a big in luence on he ex u al p ope ies o he silica ae ogel. Howe e , he
inclusion o he ca alys in he po ous s uc u e had an in luence on he po e olume. Fo silica ae ogels
he BJH po e olume calcula ed was 2.91 cm
3
/g, being educed o 2.46 and 2.29 cm
3
/g o 7 and 10%
ca alys load, espec i ely. Loading mo e ca alys led o lowe po e olume.
I is in e es ing ha al hough he po e olume is educed he su ace a ea emains cons an ,
in spi e o he inc ease o ca aly ic ma e ial loaded. This e ec could be owed o he al eady obse ed
maldis ibu ion (e.g., agglome a ion) o he ca alys inside he ae ogel s uc u e which inc eases a
highe loads. This could, in e ec , lea e mos o he su ace a ea ee. This sugges s ha mo e s udies
a e necessa y o he imp o emen o he dispe sion o he ma e ial and i s cha ac e iza ion.
Ma e ials 2018,11, 2134 8 o 13
Ma e ials 2018, 11, x FOR PEER REVIEW 8 o 13
Figu e 7. Iso he m adso p ion—deso p ion cu e o silica ae ogels.
The BET su ace a ea calcula ed was 845.3 m2/g o silica ae ogels, and i was sligh ly educed o
843.4 and 842.7 m2/g o 7.0 and 10.0% ca alys load, espec i ely. This indica ed ha he p esence o
ca alys did no ha e a big in luence on he ex u al p ope ies o he silica ae ogel. Howe e , he
inclusion o he ca alys in he po ous s uc u e had an in luence on he po e olume. Fo silica
ae ogels he BJH po e olume calcula ed was 2.91 cm3/g, being educed o 2.46 and 2.29 cm3/g o 7
and 10% ca alys load, espec i ely. Loading mo e ca alys led o lowe po e olume.
I is in e es ing ha al hough he po e olume is educed he su ace a ea emains cons an , in
spi e o he inc ease o ca aly ic ma e ial loaded. This e ec could be owed o he al eady obse ed
maldis ibu ion (e.g., agglome a ion) o he ca alys inside he ae ogel s uc u e which inc eases a
highe loads. This could, in e ec , lea e mos o he su ace a ea ee. This sugges s ha mo e s udies
a e necessa y o he imp o emen o he dispe sion o he ma e ial and i s cha ac e iza ion.
3.2. Ca bon Dioxide Hyd ogena ion
A e he p oo o concep p e iously done wo king a 20 ba and 50 and 70 °C [18], a comple e
s udy o he in luence o di e en pa ame e s has been done. The a iables whose in luence has been
s udied we e he p ehea ing empe a u e o gases, he p essu e, he gas inle mass lows and he
amoun o ca alys . The esul s a e p esen ed as a unc ion o he ime elapsed since he low (and
illumina ion) was s a ed.
3.2.1. In luence o Tempe a u e
The i s pa ame e s udied was he empe a u e o gases p ehea ing. Visual ligh was he main
sou ce o ene gy o he ca bon dioxide hyd ogena ion, al hough gases empe a u e could also ha e
an in luence on he p ocess.
In o de o analyze his a iable, se e al expe imen s we e pe o med. Fi s o all, an ini ial
expe imen wo king wi hou connec ing he LEDs was ca ied ou in o de o analyze i he eac ion
could be pe o med only wi h empe a u e. Wo king always a 20 ba , 10.0% ca alys loaded and
lows o 1 mL/min o hyd ogen and 0.33 mL/min o ca bon dioxide, expe imen s a 50, 100 and 150
°C we e done and he conclusion was he same in all he cases: no con e sion o ca bon dioxide was
p oduced.
A e his, he same expe imen s we e pe o med u ning on he ligh s. In his case, ca bon
dioxide s a ed o be ans o med (Figu e 8).
Figu e 8 showed ha p ehea ing o gases had almos no in luence on he CO2 con e sion a e
and he esul s we e almos cons an . This caused p oximi y be ween he da a poin s, and hus e o
ba s we e excluded. Howe e , he gases low con ol sys em did no allow o a oid p ehea ing. Flow
0
500
1000
1500
2000
0 0.2 0.4 0.6 0.8 1
Quan i y Adso bed (cm³/g STP)
Rela i e P essu e (P/Po)
Adso p ion Deso p ion
Figu e 7. Iso he m adso p ion—deso p ion cu e o silica ae ogels.
3.2. Ca bon Dioxide Hyd ogena ion
A e he p oo o concep p e iously done wo king a 20 ba and 50 and 70
◦
C [
18
], a comple e
s udy o he in luence o di e en pa ame e s has been done. The a iables whose in luence has
been s udied we e he p ehea ing empe a u e o gases, he p essu e, he gas inle mass lows and
he amoun o ca alys . The esul s a e p esen ed as a unc ion o he ime elapsed since he low
(and illumina ion) was s a ed.
3.2.1. In luence o Tempe a u e
The i s pa ame e s udied was he empe a u e o gases p ehea ing. Visual ligh was he main
sou ce o ene gy o he ca bon dioxide hyd ogena ion, al hough gases empe a u e could also ha e
an in luence on he p ocess.
In o de o analyze his a iable, se e al expe imen s we e pe o med. Fi s o all, an ini ial
expe imen wo king wi hou connec ing he LEDs was ca ied ou in o de o analyze i he eac ion
could be pe o med only wi h empe a u e. Wo king always a 20 ba , 10.0% ca alys loaded and
lows o 1 mL/min o hyd ogen and 0.33 mL/min o ca bon dioxide, expe imen s a 50, 100 and
150
◦
C we e done and he conclusion was he same in all he cases: no con e sion o ca bon dioxide
was p oduced.
A e his, he same expe imen s we e pe o med u ning on he ligh s. In his case, ca bon
dioxide s a ed o be ans o med (Figu e 8).
Figu e 8showed ha p ehea ing o gases had almos no in luence on he CO
2
con e sion a e
and he esul s we e almos cons an . This caused p oximi y be ween he da a poin s, and hus e o
ba s we e excluded. Howe e , he gases low con ol sys em did no allow o a oid p ehea ing. Flow
con olle s could no s abilize pe ec ly lows when he gases we e cold, and hey wo ked much be e
wi h a small hea ing. Fo his eason, i was decided o wo k always a 50
◦
C o he nex expe imen s.
Ma e ials 2018,11, 2134 9 o 13
Ma e ials 2018, 11, x FOR PEER REVIEW 9 o 13
con olle s could no s abilize pe ec ly lows when he gases we e cold, and hey wo ked much be e
wi h a small hea ing. Fo his eason, i was decided o wo k always a 50 °C o he nex expe imen s.
Figu e 8. In luence o p ehea ing empe a u e on he eac ion.
3.2.2. In luence o Ca alys Amoun
The nex a iable s udied was he amoun o ca alys . Va ia ion o he amoun inside he ae ogel
was eally easy, because he ca alys was di ec ly dilu ed in he me hanol needed as a p ecu so o
he silica ae ogel. Howe e , 13.0% ca alys loaded in he ae ogel was he maximum amoun loaded,
due o gela ion p oblems o highe loads. These highe loads equi ed longe gela ion imes, which
p omo ed he p ecipi a ion o he ca alys , and as a esul , he ca alys was no well dis ibu ed
h oughou he silica ne . Because o his, expe imen s wi h 7.0, 10.0 and 13.0% ca alys deposi ed
we e pe o med, wo king always a 20 ba , 50 °C and lows o 1ml/min o hyd ogen and 0.33 mL/min
o ca bon dioxide (Figu e 9).
Figu e 9. In luence o ca alys load on he eac ion a 20 ba (7.0% ■, 10.0% •, 13.0% ▲). Da a poin s
ep esen he a e age o 3 expe imen s, wi h e o ba s co esponding o a 95% con idence in e al.
Figu e 9 showed ha CO2 con e sion a es we e highe when he amoun o ca alys dec eased.
Using mo e ca alys did no help o imp o e he p ocess, a aining he opposi e e ec . Al hough
me allic nanopa icles did no p ecipi a e, hey could be agglome a ed inside he silica ne .
Agglome a ion p ocess en ailed a educ ion in he a ea used o ca bon dioxide hyd ogena ion, and
CO2 Con e sion a e (mmol/gca min)
CO2 Con e sion a e (mmol/gca min)
Figu e 8. In luence o p ehea ing empe a u e on he eac ion.
3.2.2. In luence o Ca alys Amoun
The nex a iable s udied was he amoun o ca alys . Va ia ion o he amoun inside he ae ogel
was eally easy, because he ca alys was di ec ly dilu ed in he me hanol needed as a p ecu so o he
silica ae ogel. Howe e , 13.0% ca alys loaded in he ae ogel was he maximum amoun loaded, due o
gela ion p oblems o highe loads. These highe loads equi ed longe gela ion imes, which p omo ed
he p ecipi a ion o he ca alys , and as a esul , he ca alys was no well dis ibu ed h oughou he
silica ne . Because o his, expe imen s wi h 7.0, 10.0 and 13.0% ca alys deposi ed we e pe o med,
wo king always a 20 ba , 50
◦
C and lows o 1ml/min o hyd ogen and 0.33 mL/min o ca bon
dioxide (Figu e 9).
Ma e ials 2018, 11, x FOR PEER REVIEW 9 o 13
con olle s could no s abilize pe ec ly lows when he gases we e cold, and hey wo ked much be e
wi h a small hea ing. Fo his eason, i was decided o wo k always a 50 °C o he nex expe imen s.
Figu e 8. In luence o p ehea ing empe a u e on he eac ion.
3.2.2. In luence o Ca alys Amoun
The nex a iable s udied was he amoun o ca alys . Va ia ion o he amoun inside he ae ogel
was eally easy, because he ca alys was di ec ly dilu ed in he me hanol needed as a p ecu so o
he silica ae ogel. Howe e , 13.0% ca alys loaded in he ae ogel was he maximum amoun loaded,
due o gela ion p oblems o highe loads. These highe loads equi ed longe gela ion imes, which
p omo ed he p ecipi a ion o he ca alys , and as a esul , he ca alys was no well dis ibu ed
h oughou he silica ne . Because o his, expe imen s wi h 7.0, 10.0 and 13.0% ca alys deposi ed
we e pe o med, wo king always a 20 ba , 50 °C and lows o 1ml/min o hyd ogen and 0.33 mL/min
o ca bon dioxide (Figu e 9).
Figu e 9. In luence o ca alys load on he eac ion a 20 ba (7.0% ■, 10.0% •, 13.0% ▲). Da a poin s
ep esen he a e age o 3 expe imen s, wi h e o ba s co esponding o a 95% con idence in e al.
Figu e 9 showed ha CO2 con e sion a es we e highe when he amoun o ca alys dec eased.
Using mo e ca alys did no help o imp o e he p ocess, a aining he opposi e e ec . Al hough
me allic nanopa icles did no p ecipi a e, hey could be agglome a ed inside he silica ne .
Agglome a ion p ocess en ailed a educ ion in he a ea used o ca bon dioxide hyd ogena ion, and
CO2 Con e sion a e (mmol/gca min)
CO2 Con e sion a e (mmol/gca min)
Figu e 9.
In luence o ca alys load on he eac ion a 20 ba (7.0%

, 10.0%
•
, 13.0%
N
). Da a poin s
ep esen he a e age o 3 expe imen s, wi h e o ba s co esponding o a 95% con idence in e al.
Figu e 9showed ha CO
2
con e sion a es we e highe when he amoun o ca alys
dec eased. Using mo e ca alys did no help o imp o e he p ocess, a aining he opposi e e ec .
Al hough me allic nanopa icles did no p ecipi a e, hey could be agglome a ed inside he silica
ne . Agglome a ion p ocess en ailed a educ ion in he a ea used o ca bon dioxide hyd ogena ion,
and his could lead o a dec ease in he CO
2
con e sion a e. In he same way, loading mo e ca alys