1
Fu anic bio uels p oduc ion om biomass using Cu-based he e ogeneous
ca alys s
Ne ea Via , Jesús M. Requies, Ion Agi e, Ai zibe I iondo, Ped o L. A ias
Chemical and En i onmen al Enginee ing Depa men . Enginee ing Facul y o Bilbao, Uni e si y
o Basque Coun y (UPV/EHU), Plaza Ingenie o To es Que edo 1 – 48013 Bilbao (Spain)
e-mail: ne ea. ia @ehu.eus, jesus. [email protected], ion.agi [email p o ec ed],
ai zibe .i [email p o ec ed], ped oluis.a [email protected]
Abs ac
Bio e ine ies can be a good al e na i e o co e he cu en needs o ene gy, such as
anspo a ion uels and also ine chemicals. 5-hyd oxyme hyl u u al (HMF) can be ob ained
di ec ly om biomass, and i is conside ed one o he mos p omising in e media e o he
syn hesis o bio uels. In his wo k he p oduc ion o 2,5-dime hyl u an (DMF) and 2,5-
dime hyl e ahyd o u an (DMTHF), wo pa icula ly a ac i e compounds o be used as bio uels
due o i s simila cha ac e is ics o gasoline, has been s udied. Fo his pu pose, a con inuous
ixed bed ca aly ic eac o was employed using HMF as aw ma e ial. Z -suppo ed Cu, Ce-Cu,
Ru-Cu and Ni-Cu ca alys s we e es ed. Mo eo e , he ca alys s ex u al p ope ies we e also
measu ed - be o e and a e he ac i i y ess - in o de o ge a deepe unde s anding o he
ela ionships among he measu ed DMF yields and he ca alys cha ac e is ics. Final DMF,
DMTHF and bu anol sepa a ion was also e alua ed.
Keywo ds
2,5-Dime hyl u an; 2,5-Dime hyl e ahyd o u an; 5-Hid oxyme hyl u u al; Hyd ogena ion;
Bime allic
Highligh s
x DMF & DMTHF p oduc ion, wo sui able bio uels, was s udied
x Ni a oids sin e iza ion o Cu and inc eases he p esence o mo e educed Cu species
x 15Ni15CuZ ca alys shows he highes DMF+DMTHF yield (53.2%)
x 15Ni15CuZ ca alys is s able a e 24 hou s on s eam o DMF (25% yield)
x Ce and Ru inco po a ion did no ha e a posi i e e ec in he DMF o DMTHF yield.
*Co esponding Au ho : Phone: +34-94 601 7297; e-mail: ne ea. ia @ehu.eus; Uni e si y o
he Basque Coun y (UPV/EHU). Enginee ing Facul y o Bilbao. Chemical and En i onmen al
Enginee ing Depa men , Plaza Ingenie o To es Que edo 1, 48013 Bilbao, Spain
This is he accep ed manusc ip o he a icle ha appea ed in inal o m in Ene gy 172 : 531-544 (2019), which has
been published in inal o m a h ps://doi.o g/10.1016/j.ene gy.2019.01.109. © 2019 Else ie unde CC BY-NC-ND
license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/)
2
1. INTRODUCTION
One o he mos se ious wo ld en i onmen al conce ns is ela ed o he clima e change due o
he g eenhouse e ec gene a ed by he an h opological CO2 emissions. Mos o hese emissions
come om ossil uels combus ion [1,2]. Ac ually, acco ding o he In e na ional Ene gy Agency
(IEA) [3–5], oil, coal and na u al gas ep esen he main ene gy esou ces (mo e han 80% o
p ima y ene gy) and hei usage is s ill inc easing [6,7]. The anspo a ion sec o is esponsible
o a ound 70 % and 20 % o he CO and CO2 emissions, espec i ely [3]. Mo eo e , he
geopoli ical ins abili y in he oil p oduce coun ies and he u u e deple ion o he a ailable
ese es [1,4,8,9] demand sus ainable al e na i es. Howe e , oil is no only he sou ce o uels
bu also o many chemicals o he p oduc ion o mul iple i ems like plas ics and clo hs [10].
Thus, he R&D sea ching o sus ainable al e na i es o oil and na u al gas has become one o
he main echnological challenges in he nea u u e.
Biomass seems o be a good candida e as a di ec sou ce o ene gy bu also as he aw ma e ial
o pla o m p oduc s o he p oduc ion o bio uels and ine chemicals [3,11]. I is clea ha he
bio uels p esen a high po en ial as enewable ene gy o anspo a ion. In ac , in 2014, a ound
90 billion li e s we e p oduced wo ldwide [12]. Al hough he i s gene a ion o bio uels
(biodiesel and bioe hanol) gene a e con o e sy due o hei compe i ion o ood c ops, he
second gene a ion can be p oduced using building blocks om lignocellulosic biomass [13–15].
Special in e es as hese building blocks me i he ones de i ed om suga s p oduced by
hyd olysis o he cellulosic and hemicellulosic ac ions o his biomass.
Among hose building blocks, 5-hyd oxyme hyl u u al (HMF) was selec ed as a op chemical in
2010 [16] and i is conside ed as an in e media e be ween biomass and inal bio uels (2,5-
dime hyl u an (DMF) and 2,5-dime hyl e ahyd o u an (DMTHF)) and biochemicals (e hyl
le ulina e (EL), 2,5- u andica boxilic acid (FDCA) o le ulinic acid (LA)) [3,8,11,17–19]. HMF can
be p oduced by dehyd a ion o C6 suga s like glucose, uc ose and suc ose [16]. Ac ually, AVA
Biomchem is al eady p epa ing i s comme cializa ion [20].
DMF and DMTHF u anic bio uels could be aluable subs i u es o anspo a ion ossil uels [21].
A la ge numbe o models ha e published abou he combus ion o u anic compounds, mainly
be o e 2000 [21]. This can be due o u anic bio uels being excellen oc ane boos e s [22].
Rega ding he DMF, some au ho s [2,8,23–28] conclude ha i could be a p omising addi i e o
oil-based uels bu aking in o conside a ion he p e ailing p essu e and empe a u e
condi ions. DMF, which shows a e y low solubili y in wa e [15,29], is epo ed as a high-quali y
bio uel o bioaddi i e, o con en ional gasoline [4,19]. I p esen s some in e es ing chemical
p ope ies as i s high oc ane numbe , highe han he one o con en ional gasoline [3] and high
ene gy densi y, simila o ha o gasoline and highe han ha o e hanol [8,15] (see Table 1).
One ad an age o i s usage is ha he la ge pa icles emission is dec eased conside ably,
al hough i is no able o educe signi ican ly he numbe o pa icles smalle han 10 nm [30].
Some au ho s epo i s good pe o mance as a uel on di ec injec ion spa k igni ion (DISI)
engines [2,8], wi hou any impo an modi ica ions o he engine, bu also in comp ession
engines [31]. Chen e al. [24,25] es ed diesel/DMF blends ollowing di e en mul i-injec ion
combus ion s a egies in diesel engines and cha ac e ized he co esponding emissions. They
obse ed ha igni ion-delay was p olonged and smoke dec eased when he EGR (exhaus gas
3
eci cula ion) a e inc eased. Liu e al. [26] compa ed he emissions o ou di e en diesel
mix u es wi h a 20% blending a io by olume (ce ane + iso-ce ane, n-hep ane, DMF and DMF+2-
e hylhexyl ni a e) a ying he EGR om 0% o 62%. They demons a ed ha he e is li le
di e ence in combus ion and emissions among he i s wo blends and pu e diesel. The lowe
ce ane numbe o he DMF20 blend is he uppe mos ac o o con ol he combus ion p ocess,
as ha ing a mo e p emixed combus ion educes soo o ma ion and emission. Howe e , i seems
ha diesel/DMF blends show highe NOx emissions when compa ed o pu e diesel usage [27,28].
Table 1 Compa ison o speci ica ions among con en ional and enewable uels [21,24–28,32]
Pa ame e s Diesel Gasoline Biodiesel DMF DMTHF E hanol n-
bu anol
Molecula o mula ϭϮоϮϱ ϰоϭϮ ϭϮоϮϰ C
6H8O C6H12O C2H5OH C4H9OH
Oc ane numbe ϵϬоϵϵ 119 82 108 96
Ce ane numbe ϰϬоϱϱ ϭϬоϭϱ ϰϳоϱϮ 9 17.23a 8 25
Oxygen con en (w %)
10 16.7 16 34.8 21.6
S oichiome ic ai / uel a io 14.3 14.7 12.5 10.79 11.73 9.02 11.21
Lowe hea ing alue (MJ/kg) 42.5 42.7 38.81 33.7 38.3 26.8 33.1
Densi y a 20 °C (g/cm3) 0.82 0.745 0.87 0.89 0.83 0.79 0.81
Viscosi y (mm2/s) a 40 °C 1.9–4.0 0.4–0.8b 4 0.65 0.47a 1.08 2.63
Flash poin (°C) 65–88 оϰϱоϯϴ 166 оϭ 27 8 35
La en hea ing (kJ/kg) a
25 °C ϮϳϬоϯϬϭ 380–500 200 333 348a 904 582
Boiling poin (°C) ϭϴϬоϯϳϬ 25–215 ϮϲϮоϯϱϵ ϵϮоϵϰ 90-92 78.4 117.7
Sa u a ion p essu e (kPa) a
38 °C 1.86 31.01 12.8a 15.1a 13.8 2.27
Au o-igni ion empe a u e
(°C) 246 420 363 285.85
434 385
a Speci ica ion es ima ed by Aspen Plus, b a 20 ºC
Acco ding o he las published indings in he open li e a u e, DMF can be p oduced om
selec i e hyd ogena ion o HMF [15] using non-noble and noble me al ca alys s [3,4]. This
con e sion in ol es di e en pa hways [3,15,33] in which dehyd a ion and hyd ogena ion
eac ions a e p edominan . Figu e 1 shows a simpli ied eac ion scheme o DMF p oduc ion om
biomass. The hyd ogenolysis o HMF o 5-me hyl u u yl alcohol (MFA) ia 5-me yl u u al (MF)
o ia 2,5-bishyd oxyme hyl u an (BHMF) a e he main ou es o DMF p oduc ion. A e wa ds,
he MFA hyd ogena ion p oduces DMF.
4
Figu e 1 P oduc ion ou e o DMF and DMTHF ia HMF om biomass (adap ed om [34,35]).
On he one hand, me al hyd ogena ion ca alys s a e necessa y o he hyd ogena ion o he
HMF. The e a e wo ypes o me al-based ca alys s, hose con aining noble me als such as Pd,
P , Ru, Rh and Au [3,15,29,36,37], and hose con aining ansi ion me als, such as Cu, Co, Ni and
Fe [36,38,39]. Some au ho s [17] ha e also s udied he combina ion o bo h ype o me als
ob aining a high selec i i y (o e 70 %) in he p oduc ion o DMF using Cu-Ru suppo ed on
ca bon a 220 ºC and 6.8 ba o H2 and using 1-bu anol as sol en [17]. Chidamba am e al. [40]
ob ained a selec i i y o 32 % o DMF wi h a 47 % HMF con e sion using a Pd/C ca alys and
ionic liquids a 120 ºC and 62 ba o H2. Zu e al. [4] epo ed a e y high yield o DMF (93.4 %)
using e ahyd o u an (THF) as sol en a 130 ºC, and 7 ba o H2 pa ial p essu e using a Ru/
Co3O4 ca alys . Ru pa icipa es in he hyd ogena ion while Co3O4 was he esponsible o he
HMF adso p ion and he b eakage o he C-O bonds. Goyal e al. [38] achie ed o al HMF
con e sion wi h a DMF selec i i y o e 98.7 % employing a Ni ca alys suppo ed on mesopo ous
ni ogen ich ca bon ma e ial and using liquid wa e as sol en a 200 ºC and 30 ba o H2
p essu e. I seems ha using Ni nanopa icles (below 5 nm) leaded o a high Ni dispe sion
associa ed o a high in e ac ion wi h he suppo . On he con a y, wi h highe nickel pa icle
sizes lowe DMF selec i i ies we e achie ed [38]. Howe e , all hese da a epo ed we e
de eloped using ba ch eac o s and he co esponding ca alys s abili y was no es ed. On he
o he hand, apa om he me als, he suppo also plays a key ole in DMF p oduc ion. The
p esence o acid suppo s could cause he loss o one o he adicals p esen in he HMF
molecule o deg ada ion o he DMF since acid si es, B øns ed o Lewis, a o he C-C bond
clea age. Thus, neu al-basic and weak acid suppo s a e sugges ed o HMF hyd ogenolysis o
DMF. This ype o neu al-basic suppo s seem o a oid C-C bond clea age o he HMF e ia y
ca bon [13]. Ca bon ma e ials [17,41,42], zi conia [13] and me al oxides such as Co3O4 [4] seem
o be good candida es o his pu pose.
Ano he in e es ing by-p oduc ha appea s is DMF p oduc ion compound is DMTHF. This
compound, oge he wi h DMF, is conside ed as a sui able liquid bio uel due o hei simila
HMF
BHMF
MF
MFA DMF
BHMTF DMTHF
MTHFA
Ring OpeningP oduc s
(ke ones)
Ring OpeningP oduc s
(polyols)
5
p ope ies (such as ene gy densi y, boiling poin , and wa e solubili y) o comme cial gasoline
and high oc ane numbe [35,43,44] (see Table 1). DMTHF can be ob ained by o e
hyd ogena ion o MFA (hyd ogena ion o DMF) bu also ia hyd ogena ion+hyd ogenolysis o
MFA being 5-me hyl e ahyd o u u yl alcohol (MTHFA) he in e media e compound. Howe e ,
his la e eac ion pa hway seems o be only ca alyzed by noble me als such as Pd [35].
Addi ionally, some o he byp oduc s, such as 1,6-hexanediol and 1,2,6-hexane iol, can also be
ob ained due o he u anic ing opening. Mo eo e , some au ho s epo he possibili y o
ge ing DMTHF by e he i ica ion o 2,5-hexanedione [45]. E he i ica ion eac ions a e ypically
ca alyzed by acid si es, sugges ing ha HD cycliza ion o DMTHF migh be acili a ed by Lewis
acidic si es on Ru in he o m o RuOx [30–32].
Al hough he p oduc ion o DMF is cu en ly in a e y ea ly s age, some au ho s like Kazi e al.
[46] ha e al eady ca ied ou a echno-economic assessmen o DMF and HMF p oduc ion using
uc ose as he aw ma e ial. The minimum selling p ices o HMF and DMF we e es ima ed as
1.3 $/L and 2 $/L, espec i ely. Howe e , his analysis was able no only o es ima e minimum
selling p ices bu also o iden i y di e en a eas o imp o emen . Thus, one o he mos
impo an conclusions is ha he ca alys pe o mance a he p esen le el o de elopmen does
no seem o be enough o p oduce comme cial DMF o be used as a uel. Mo eo e , in he
sensi i i y analysis hey pe o med, he DMF yield is he pa ame e ha a ec s he mos he
DMF minimum selling p ice. Howe e , he e a e also some o he impo an ques ions o deal
wi h such as he equi ed syne gies in a bio e ine y in o de o alo ize/u ilize by-p oduc s, o
he necessi y o in es iga e in enewable sou ces o H2.
Based on his backg ound, he p esen pape ocuses on he ca aly ic ans o ma ion o HMF
in o DMF and DMTHF, bu using a con inuous ixed bed eac o ins ead o ba ch eac o s. High
bio uel p oduc ion a es equi e he use o con inuous eac o s. Besides ha , eal ca alys
s abili y s udies a e also easie unde con inuous ope a ing condi ions. Fo his pu pose, he use
o inexpensi e bi unc ional ca alys s was s udied, analyzing he in luence o he suppo and
di e en me als on he DMF yields. Ca alys s s abili y o he mos p omising ca alys s was also
been s udied.
2. EXPERIMENTAL PROCEDURE
The expe imen al p ocedu e used o his s udy is shown in Figu e 2. Fi s , ca alys s and
ope a ing condi ions we e selec ed based on he li e a u e. A e his, a basic cha ac e iza ion
o he ca alys s was ca ied ou ollowed by he ca aly ic es s. The esul s ob ained om he
es s and he basic cha ac e iza ion a e hen used o es ablish wha ca alys s equi ed a deep
cha ac e iza ion and assess he possibili y o imp o ing ca alys s ac i i y and s abili y by
modi ying hem.
6
Figu e 2 Ou line ollowed o he ca aly ic s udy o he DMF and DMTHF p oduc ion om HMF.
2.1. DMF and DMTHF p oduc ion
Syn he ic HMF (Sigma-Ald ich, 99%) dilu ed in 1-bu anol was used as he eed o a con inuous
ixed-bed eac o in o de o p oduce DMF and DMTHF. 1-Bu anol p esen s sui able p ope ies
o be used as a g een sol en in DMF p oduc ion and i is one o he mos used sol en agen s
o his pu pose [17,29,36,47]. 1.5 w % o HMF was con inuously ed o he eac o by means o
a HPLC pump. 0.5 g o ca alys dilu ed wi h ine CSi (weigh ca alys s/CSI a io = 1:9) we e placed
in he eac o and a WHSV (w HMF/(w ca ·h)) o 0.15 h-1 was used. In o de o know he e ec
o he ca alys educ ion s ep in ca aly ic ac i i y, in some expe imen s he ca alys was in-si u
educed be o e he ac i i y es using H2 a 1 ba and 200 – 275 °C. A e his, he ac i i y es s
we e ca ied ou wi h esh p e- educed and no p e- educed esh ca alys s unde di e en
empe a u e and p essu e condi ions o 200 - 275 °C and 15 - 20 ba o H2, espec i ely. The
indica ed ope a ion condi ions we e selec ed based on he esul s ob ained in a p e ious wo k
[13].
In o de o epo he ac i i y es s esul s, he HMF con e sion (%) and DMF o DMTHF yields
(%) we e used de ined as ollows:
in ou
HMF HMF
in
HMF
N (mol/min)-N (mol/min)
Con e sion (%) = ·100
N(mol/min)
(1)
ou
DMF/DMTHF
in
DMF/DMTHF
N (mol/min)
Yield (%) = ·100
N (mol/min)
(2)
2.2. Analysis
Reac o eed and ou pu liquid s eams we e sampled and analyzed by o -line high-
pe o mance liquid ch oma og aphy (HPLC) and gas ch oma og aphy (GC). HMF con en o bo h
inle and ou le s eams was de e mined by HPLC 1260 In ini y equipped wi h HI-Plex H column
and an in a ed de ec o . An Agilen 6804 GC equipped wi h a lame ioniza ion de ec o (FID)
and using a Sup awax 280 capilla y column was used o analyze DMF, DMTHF and o he
byp oduc s p esen in he ou le s eam.
2.3. Ca alys p epa a ion
Di e en monome allic and bime allic ca alys s we e p epa ed by he incipien we ness
imp egna ion me hod. Fo his pu pose, Z O(NO3)2·xH2O (Sigma-Ald ich, 99%), Cu(NO3)2·2.5H2O
Ca alys s’
P epa a ion
Imp egna ion
Ca aly ic
Pe o mance
F esh Ca alys s’
Basic Cha ac e iza ion
F esh/Used Ca alys s’
Deep Cha ac e iza ion
N
2
physiso p ion; ICP-AES;
H
2
-TPR TPD; XRD; SEM; XPS
Inco po a ion o a second me al and suppo modi ie
7
(Al a Aesa , 98%), RuCl3·xH2O (Me ck, me allic base Ru 35-40%), CeN3O9·6H2O (Al a Aesa ,
99,5%) and Ni(NO3)2·6H2O (Sigma-Ald ich, 99.999%) eagen s we e used.
Cu suppo ed on Z O2 ca alys s we e p epa ed by a single s ep imp egna ion and we e designed
as 15CuZ , 30CuZ , 45CuZ , in which he nume ical ac o indica es he nominal coppe con en
as w %. Subsequen ly, he Ru-Cu bime allic ca alys s, designed as 1RuXCuZ (being X he nominal
weigh o Cu), we e p epa ed by imp egna ion o he p e iously calcina ed 15CuZ and 30CuZ
ca alys s wi h a u henium p ecu so solu ion. Cu suppo ed on CeZ was a e wa ds p epa ed
in wo sequen ial s eps, i s imp egna ing Ce on Z and hen Cu on CeZ , ha ing d ied and
calcina ed he CeZ be o e being imp egna ed he Cu. These ca alys s we e designed as
30CuNCeZ and 15CuNCeZ (we e N e e s o nominal weigh o Ce). Finally, bime allic
15Ni15CuZ was p epa ed in a single s ep co-imp egna ion. To unde s and he beha io o
bime allic ca alys , he monome allic 15NiZ ca alys was also p epa ed. A e imp egna ion and
d ying, he samples we e calcined in ai a 250 °C o 2 h.
2.4. Ca alys cha ac e iza ion
The ca alys s we e cha ac e ized by se e al physicochemical echniques such as N2
physiso p ion (BET me hod), Plasma A omic Emission Spec oscopy (ICP-AES), Tempe a u e
P og ammed Reduc ion (H2-TPR), ammonia Tempe a u e P og ammed Deso p ion (TPD), X- ay
Di ac ion (XRD), Scanning Elec on Mic oscope (SEM) and X- ay Pho oelec on Spec oscopy
(XPS).
2.4.1. Tex u al p ope ies
BET su ace a ea and po e cha ac e is ics o he calcined esh ca alys s we e e alua ed om
he N2 adso p ion–deso p ion iso he ms ob ained a -198 °C o e he whole ange o ela i e
p essu es. In he case o he Z -suppo ed Cu monome allic and bime allic ca alys s, his sample
was cha ac e ized using Au oso b®-1-C/TCD (Quan ach ome, USA) a e ou gassing solid
samples a 150 °C o 24 h.
2.4.2. Plasma a omic emission spec oscopy (IPC-AES)
Me al con en s we e de e mined by ICP-AES echnique, using a Pe kin-Elme Op ima 3300DV
appa a us, p e ious dissolu ion o he g ound samples in acid solu ions (37.5% HNO3, 37.5% HF
and 25% HCl, in olume).
2.4.3. Tempe a u e P og ammed Reduc ion wi h hyd ogen (H2-TPR)
An Au oChem II Ins umen (Mic ome i ics, USA) equipped wi h a TCD de ec o was used o
s udy he educibili y o he calcined ca alys s. The samples we e wa med up o 900 °C a a a e
o 10 °C/min passing 50 mL/min o educing gas (5% / H2 dilu ed in A ) h ough he sample.
2.4.4. Tempe a u e p og ammed deso p ion wi h ammonia (TPD)
The acidi y o he calcined ca alys s was measu ed by empe a u e p og ammed deso p ion o
NH3. Fo his pu pose, as well as in he H2-TPR echnique, an Au oChem II ins umen was used.
Fi s ly, samples we e lushed by He a 250 °C o 30 min and hen educed wi h a 5% H2/He
mix u e. A e wa ds he ca alys was cooled down o 40 °C and NH3 loading was main ained o
8
30 min. Be o e s a ing he deso p ion p ocess, physically abso bed NH3 was emo ed using He
a 85 °C un il no u he deso p ion was eco ded. Finally, he elease o chemically adso bed
NH3 was eco ded inc easing he empe a u e om 85 °C o 800 °C a a a e o 10 °C /min.
2.4.5. X- ay Di ac ion (XRD)
X- ay di ac ion pa e ns we e ob ained using a Sei e XRD 3000P di ac ome e , equipped
wi h a PW 2200 B agg-B en ano u/2u goniome e , ben g aphi e monoch oma o and au oma ic
sli , using Cu K adia ion (0.15418 nm) and 0.028° s eps o scanning. The Sche e equa ion was
used o calcula e he a e age pa icle size o he c ys alline species egis e ed o educed and
used ca alys s.
2.4.6. Scanning Elec on Mic oscope (SEM)
The mo phological appea ance o he esh educed and he used ca alys s was s udied by SEM
analysis in a JEOL JSM-6400 wi h W ilamen and a esolu ion o 3.5 nm. The esh educed and
he used samples we e p epa ed ia dispe sion in o isop opanol as sol en and placed on a
ca bon-coa ed coppe g id (300 Mesh) ollowed by d ying unde acuum.
2.4.7. X- ay pho oelec on spec oscopy (XPS)
This echnique was used o e alua e he su ace cha ac e is ics (oxida ion s a e o he species
o med, in e ac ions, a omic a ios, e c.) o he samples. The measu emen s we e ca ied ou
wi h a VG Escalab 200R spec ome e equipped wi h a hemisphe ical elec on analyze and an
Al K1 (h = 1486.6 eV) 120W X- ay sou ce. The powde ed samples we e deposi ed on a s ainless
s eel sample holde , placed in he p e- ea men chambe and degassed a 300 °C. The
spec ome e base p essu e was ypically 10о9 o (0.0133 ba ). The spec a we e collec ed a a
pass ene gy o 20 eV, which is ypical o high esolu ion condi ions. Bo h esh educed and spen
ca alys s we e analyzed wi h his echnique.
3. RESULTS AND DISCUSSION
3.1. Ca alys s cha ac e iza ion esul s
3.1.1. Chemical and ex u al p ope ies
Acco ding wi h he N2 adso p ion-deso p ion esul s shown in Table 2, he calcined Z p esen s
high BET a ea and po e olume.The p esence o Ce on he Z suppo educes a li le bi he
su ace a ea and he po e olume due o he Ce deposi ion on he po es. Ne e heless, a e he
inc emen o Ce on he ca alys om 5 o 20 %, he BET a ea was p ac ically he same (due o
he equipmen expe imen al e o is 3 m2/g) and also po e olume. Howe e , he p esence o
a highe amoun o ce ia dec eased he po e diame e . P obably he inco po a ion o mo e Ce
a ou s he o ma ion o la ge ce ia pa icles.
The addi ion o Cu on o he Z and CeZ mesopo ous suppo s causes a dec ease in he speci ic
su ace a ea o he ca alys s possibly due o he p esence o CuO c ys alli es in he po es, limi ing
he adso p ion o N2 du ing he adso p ion-deso p ion expe imen [48]. I is ema kable ha his
educ ion endency o he su ace a ea is mo e impo an o low Cu con en s han o high Cu
con en s, excep o he 15Cu20CeZ ca alys . This beha io was also obse ed in he ca alys s
wi h 15 w % o Cu like 1Ru15CuZ , 15Ni15CuZ and 15Cu5CeZ . I seems ha he p esence o
9
high ce ia loadings modi ied he way in which Cu was inco po a ed on he ca aly ic su ace. In
he case o bime allic ca alys s, he p esence o a second me al educes he su ace a ea and he
po e olume. Rega ding he 15NiZ ca alys , he nickel inco po a ion educed d as ically he
su ace a ea. As well as Cu, he me allic nickel clus e s blocked he su ace, esul ing in a low
dispe sion o he me al [49]. The 15RuZ ca alys p esen s simila beha io han he
monome allic ca alys as he Ru inco po a ion dec eased bo h he BET su ace and he po e
olume.
Table 2 Nomencla u e and ex u al p ope ies o calcined suppo s and Cu, Ru, RuCu and NiCu
ca alys s.
G oup Ca alys s BET
(m2/g)
Po e Volume
(cm3/g)
A e age po e
diame e (nm)
Suppo s
Z 137.6 0.267 7.2
5CeZ 112.3 0.228 7.9
20CeZ 114.1 0.211 3.7
Cu, Ru and RuCu
ca alys s
15CuZ 35.1 0.094 10.7
30CuZ 81.8 0.267 13.1
45CuZ 52.0 0.043 4.2
1RuZ 89.3 0.401 18.0
1Ru15CuZ 22.0 0.068 12.1
1Ru30CuZ 49.0 0.062 4.9
CeZ -suppo ed Cu
ca alys s
15Cu5CeZ 40.8 0.073 7.2
15Cu20CeZ 110.3 0.204 7.2
30Cu5CeZ 60.1 0.099 6.4
Ni and NiCu
ca alys s
15NiZ 5.0 0.033 25.3
15Ni15CuZ 18.1 0.034 7.7
The eal me al con en s o he ca alys s we e de e mined by ICP-AES and he esul s a e
p esen ed in Table 3. As i can be obse ed, in gene al, he eal alues we e lowe han he
nominal ones o bo h monome allic and bime allic ca alys s. This di e ence was specially high
when ce ia was deposi ed on he zi conia suppo . These big di e ences in he ce ia-doped
suppo s also we e measu ed, as expec ed, in he co esponding ca alys s (CuCeZ ). The
di e ence o he eal amoun o ce ia inco po a ed on he CeZ suppo s is so low ha his can
explain he small di e ence be ween he BET su ace a eas o ce ia doped and non-doped
zi conia suppo s. P obably he su ace o he Z O2 is no e y eac i e wi h he ce ium, i.e., he
Z O2-Ce in e ac ion is e y weak, and he e o e he Ce does no inco po a e on Z O2 su ace
e icien ly.
Table 3 Real me al load o he ca alys s de e mined by ICP-AES.
G oup Ca alys s Cu (%) Ru (%) Ni (%) Ce (%)
Suppo s 5CeZ
20CeZ
0.4
2.4
Cu, Ru and RuCu
ca alys s
15CuZ 13.0
30CuZ 17.0
45CuZ 28.4
1Ru15CuZ 10.9 0.26
1Ru30CuZ 20.6 0.50
16
3.1.5. Mo phological cha ac e is ics
The SEM measu emen s we e ca ied ou o esh educed and used ca alys . As i can be
obse ed in Figu e 8, Cu is well dispe sed on Z , in bo h esh and used ca alys s, and i appea s
in pa icles smalle han 1000 nm. Z pa icles show a smo he su ace and appea as bigge
pa icles, wi h a ound 0.1-0.5 mm. In case o he 30Cu20CeZ and 15Cu20CeZ ca alys s, EDX
measu emen s ce i y he Ce p esence only in he 30Cu20CeZ ca alys bu no in all he s udied
poin s (see Figu e S2). Mo eo e , i is no possible o dis inguish he Ce om Cu in he SEM
pic u es. These esul s a e in good ag eemen wi h ICP-AES and XRD, whe e he ce ium con en
on he ca aly ic su ace was e y low.
Conce ning he used ca alys s, EDX analysis ce i ied he p esence o coke pa icles on he
ca alys su ace. Cu s ill emained well dispe sed wi h a simila pa icle size han he one in he
esh ca alys . Z pa icles also showed a simila pa icle size excep in he case o he 30CuZ
ca alys , whe e he deposi ed coke co e s Z and Cu pa icles and i is no possible o de e mine
hei pa icle sizes. The p esence o coke was also de ec ed by XPS (see Table 7). As well as wi h
he esh ca alys s, i is no possible o dis inguish Ce om Cu in he case o he 30Cu20CeZ and
15Cu20CeZ ca alys s. Ne e heless, con a y o he esh ca alys s, no Ce could be de ec ed by
EDX in any o he s udied a eas.
15CuZ ( esh) 15CuZ (used)
30CuZ ( esh) 30CuZ (used)
45CuZ ( esh) 45CuZ (used)
17
15Cu20CeZ ( esh) 15Cu20CeZ (used)
30Cu20CeZ ( esh) 30Cu20CeZ (used)
Figu e 8. SEM measu emen s o esh and used ca alys s.
3.1.6. Su ace p ope ies
Table 7 summa izes he XPS esul s whe e su ace p ope ies, such as su ace a omic a ios and
Cu species a ios, a e p esen ed. The a io o me allic Cu o Cu ions (Cu0/Cu++Cu2+) was calcula ed
o iden i y i he ca alys s we e educed o oxidized du ing hei use in he s udied eac ion. Fo
his a io calcula ion, di e en ene gy bands we e employed. Cu 2p binding ene gys hows a
alue o 934 eV o Cu2+ [75]. Howe e , in his egion i is no possible o dis inguish be ween
Cu0 and Cu+, because bo h o hem exhibi he same ene gy binding o 932 eV [75]. The e o e Cu
LMM egion was hen analyzed, whe e Cu0and Cu+species p esen a peak a 335 eV and 337.5
eV [75–78], espec i ely.
As i can be obse ed, he Cu/Z su ace a omic a io o esh educed monome allic ca alys s
inc eased when highe Cu loadings we e inco po a ed. In gene al, he p esence o o he co-
me als like Ru and Ce inc eased he Cu/Z a io imp o ing he Cu p esence on ca alys ’s su ace.
The addi ion o a low amoun o a noble me al, o ins ance Ru, lead also o a highe me al
dispe sion and me allic su ace a ea, and high deg ee o educ ion, as he TPR p o iles sugges ,
a o ing he o ma ion o me allic Cu.
18
The Ni inco po a ion imp o ed he esh educed Cu/Z a io when coma ed o he
monome allic one. The global e ec is lowe han he one associa ed o Ru and Ce inco po a ion.
The inco po a ion o a noble me al o a non-noble monome allic ca alys gene a es an in ima e
con ac and a syne gy be ween bo h me als, as he XRD da a co obo a es, esul ing in a be e
su ace exposu e, educ ion capaci y, me al dispe sion and me al su ace a ea [79,80]. In he
case o he 1RuCuZ ca alys s, his was no ul illed. P obably, he coke deposi ion on big me al
pa icles p o oked he educ ion o di ac ion pa e ns in ensi y o used ca alys s, esul ing in
lowe a e age pa icle size when compa ed wi h he da a de e mined o he educed ones.
Ne e heless, a e he ac i i y es s, all he ca alys s, he monome allic ones like he bime allic
ones, showed a lowe Cu/Z a io, which is an indica ion o some sin e ing aking place. In
addi ion, he 15Ni15CuZ ca alys also showed some Ni sin e ing. These esul s a e in good
ag eemen wi h he XRD esul s. Mo eo e , he C/Z a io o he used 15Ni15CuZ ca alys was
e y high i i is compa ed wi h he es o he 15Cu monome allic and bime allic ca alys s.
P obably he p esence o he Ni a o ed he C-H bond clea age, inc easing he coke o ma ion.
Some ca alys s showed a lowe Cu0/(Cu++Cu2+) a io a e eac ion, meaning ha hey ha e
su e ed some ac i e phase oxida ion du ing he eac ion, bu in gene al he a ia ion o his
a io is small. Only he 30Cu5CeZ and 30Cu20CuZ su e ed some signi ican oxida ion
phenomena..
Table 7 XPS esul s o Z -suppo ed Cu and RuCu, and XCuNCeZ ca alys s.
G oup Ca alys Cu/Z
Ƶ϶
(Cu+нƵϸЀͿ C/Z Ce/Z Ru/Z Ni/Z
Cu
ca alys s
15CuZ F . ed. 0.24 0.73 1.04
Used 0.12 0.70 10.61
30CuZ F . ed. 0.49 0.43 1.74
Used 0.32 0.60 17.78
45CuZ F . ed. 1.36 0.31 5.38
Used 0.22 0.54 144.77
CeZ -
suppo e
d Cu
ca alys s
15Cu5CeZ F . ed. 0.51 0.48 3.95 0.0042
Used 0.16 0.53 16.47 0.0022
15Cu20CeZ F . ed. 0.30 0.58 1.71 0.0027
Used 0.28 0.54 4.51 0.0009
30Cu5CeZ F . ed. 0.42 0.43 2.76
0.0041
Used 0.33 0.37 12.83
0.0019
30Cu20CeZ F . ed. 0.70 0.61 3.61
0.0025
Used 0.23 0.45 22.77
0.0015
Ru and
RuCu
ca alys s
1RuZ F . ed. 1.21 0.08
Used 14.60 0.85
1Ru15CuZ F . ed. 1.31 0.55 7.10 0.44
Used 0.24 0.63 22.61 1.51
1Ru30CuZ F . ed. 11.23 0.30 13.55 1.03
Used 1.10 0.33 7.72 0.49
NiCu
ca alys s 15Ni15CuZ F . ed. 0.54 3.95 0.43
Used 0.25 73.65 0.10
19
In gene al, he use o ca alys s in he HMF hyd ogenolysis eac iona p o oked he o ma ion o
g ea quan i ies o coke, excep in he case o he 1Ru30CuZ and 15Cu20CeZ ca alys s. In his
sense, he usage o Ce clea ly limi ed coke p oduc ion [81]. Thus al hough he amoun o Ce on
he ca aly ic su ace was e y small, a he highes Ce/Z a ios he bes coke esis ance abili y
esul s we e ob ained. The measu ed Ce/Z a ios on he XCuNCeZ ca aly s’ su ace we e
smalle han he expec ed ones, bu hey a e in a good ag eemen wi h he ICP-AES, XRD and
SEM esul s [81]. In he case o he 1Ru30CuZ he high dispe sion eached on he ca alys
su ace dis a o ed he coke p oduc ion, as small me al pa icles do no a o coke p oduc ion.
3.2. Ac i i y esul s
An ini ial ca aly ic sc eening was ca ied ou in o de o se he mos op imal ope a ing
condi ions, a ying he empe a u e, H2 pa ial p essu e and he need o no o ca alys
p e ea men (see Supplemen a y Ma e ial). Thus, he ac i i y expe imen s we e ca ied ou a
he ollowing ope a ing condi ions: T = 275 °C, P = 15 ba o H2, no p e- ea men ( educ ion) o
he ca alys s and a WHSV o 0.15 h-1
3.2.1. Me al and suppo in luence on ca aly ic ac i i y
Di e en ca alys s we e es ed in o de o know he in luence o me al and suppo on he
ca aly ic ac i i y and selec i i y. A e his s udy, he mos p omising ca alys was selec ed o a
long es . The ac i i y esul s o he es ed ca alys s a e summa ized in Table 8. This able
p esen s he maximum yield o DMF, DMTHF and hei summa ion, emphasizing in b acke s he
ime on s eam whe e hese maximum yields we e ob ained. Mo eo e , he yield a 6 h on
s eam is also de ailed. In all he cases, comple e con e sion o HMF was eached. Addi ionally,
a g aph o he eac ion e olu ion o each ca alys wi h he yields o bo h p oduc s appea s in
Figu e S1.
Table 8 Ac i i y esul s o di e en ca alys s a 275 ºC, 15 ba o H2 and space eloci y o 0.15
h-1 .
Ca alys DMF yield (%) DMTHF yield (%) DMF + DMTHF yield (%)
Max. 6h Max. 6h Max. 6h
15CuZ 25.8 (6h) 25.8 7.9 (2h) 1.1 26.9 (6h) 26.9
30CuZ 24.1 (6h) 24.1 7.1 (2h) 2.1 27.2 (4h) 26.2
45CuZ 14.5 (3h) 3.5 2.6 (3h) 0.3 17.1 (3h) 3.8
1Ru15CuZ 23.5 (3h) 23.1 8.1 (2h) 1.1 29.7 (3h) 24.2
1Ru30CuZ 22.2 (4h) 21.6 5.4 (3h) 3.7 27.0 (4h) 25.3
1RuZ 0.8 (2h) 0.4 0.1 (1h) 0.0 0.8 (2h) 0.4
15Cu5CeZ 24.0 (6h) 24.0 7.1 (2h) 2.5 27.3 (5h) 26.5
15Cu20CeZ 14.7 (5h) 14.2 7.0 (3h) 5.8 21.6 (5h) 20.0
30Cu5CeZ 21.9 (6h) 21.9 5.6 (3h) 4.8 26.7 (6h) 26.7
30Cu20CeZ 23.5 (6h) 23.5 10.3 (2h) 3.8 31.7 (4h) 27.3
15NiZ 21.9 (4h) 15.2 0.0 0.0 21.9 (4h) 15.2
15Ni15CuZ 15.1 (6h) 15.1 40.2 (4h) 32.9 53.5 (4h) 48.0
Rega ding DMF+DMTHF yield, he ob ained esul s o he 15CuZ and 30CuZ ca alys s a e e y
simila . In bo h cases, a o al yield o mo e han 25 % was achie ed and i was s able a e 6 h
20
on s eam (see Figu e S1). In con as , he 45CuZ ca alys eaches a maximum yield o 17 %
a e 3 h o eac ion and hen he ca alys su e s om deac i a ion.
The simila selec i i y esul s ob ained o he 15CuZ and 30CuZ ca alys s sugges ha he o al
amoun o Cu in he ca aly ic sys em o on he ca aly ic su ace is no a key ac o (see XPS esul s
in Table 7 and ICP-AES esul s in Table 3). I seems ha he Cu dispe sion is he one playing a key
ole in he DMF p oduc ion. The ca alys wi h he lowes a e age pa icle size (measu e by XRD,
see Table 5) also p esen ed he bes ac i i y and s abili y esul s. The acidi y is also an impo an
ac o , bu all Cu monome allic ca alys s showed a simila acidi y. Bu he acidi y o he ca alys s
played an impo an ole on he dec ease o DMF yield h ough he o ma ion o deg ada ion
p oduc s de i ed om an excess o C-C bond clea age a o ing also ca bon/coke pa icles
o ma ion. Fo ins ance, as i was expec ed, due o he simila acidi y, s eng h o acidic si es,
TPR educ ion p o iles and dispe sion o he Cu on ca aly ic su ace (XPS), he ac i i y beha io
o he 15CuZ and 30CuZ ca alys s did no di e oo much and compa able DMF and DMTHF
yields we e achie ed unde he s udied ope a ing condi ions. A 6 hou s on s eam, he 15CuZ
and 30CuZ ca alys s p esen ed a s able beha io al hough he Cu/Z a io in he used ca alys s
dec eased along he es , while he C/Z a io inc eased. P obably, o longe imes on s eam
some deac i a ion would ha e been de ec ed, as i can be obse ed o he 45CuZ ca alys . This
45CuZ ca alys p esen ed he lowes acidi y and a highe educibili y han 15CuZ and 30CuZ
ca alys s. This lowes acidi y is an indica ion o a low Cu-Z in e ac ion inc easing he Cu
educibili y, bu his weak me al-suppo in e ac ion a o ed he sin e ing p ocesses (Cu/Z a io
dec eases om 1.32 o 0.22), as i can be obse ed in he Table 7. The inc emen o he pa icle
size a o s he coke p oduc ion ( he C/Z a io o he 45CuZ is he highes one, 141.3). This
p oduc ion seems o ake place on he ca aly ic su ace o he 45CuZ [82].
In o de o imp o e he ca aly ic ac i i y o he mos s able ca alys s, 15CuZ and 30CuZ , Ru was
added using a second imp egna ion. The Ru is a widely used me al o he e ogeneous ca aly ic
hyd ogena ion [83]. To examine he e ec o Ru in he eac ion, he 1RuZ ca alys was es ed.
This ca alys p esen ed a o al con e sion o HMF bu e y low o negligible DMF o DMTHF
yields. This means ha he p esence o Ru implies ano he eac ion pa h (see Figu e 1) ha he
HMF hyd ogenolysis. The monome allic Ru ca alys b eaks p e e en ially he C-C bond ins ead
o he C-O bond, which esul s in deg ada ion p oduc s [84,85]. The 1RuZ ca alys was he
ca alys wi h he highes acidi y and wi h he s onges acid si es as TPD-NH3 da a co obo a ed.
This cha ac e is ic a o s he C-C bond clea age o he HMF molecule and he e o e, he
p esence o acid suppo s could cause he loss o one o he adicals p esen in he HMF
molecule o deg ada ion o he DMF [12]. This s ong acidi y could also explain he o al HMF
con e sion and he negligible DMF and DMTHF yield.
In Figu e S1 i can be obse ed ha he imp egna ion o Ru in CuZ ca alys s implied a highe
p oduc ion o DMTHF, bu he o al yield o DMF and DMTHF was no imp o ed. Mo eo e , a
dec ease o he o al yield a e 3 – 4 h ime on s eam could be de ec ed. Ne e heless, he
p esence o Ru imp o es he educibili y and he e o e, he Cu/Z a io on he ca aly ic su ace
was qui e highe han on he monome allic one. These ca alys s su e ed om high p oblems o
deac i a ion as i can be obse ed in he lowe Cu/Z a ios o he used ca alys s han he ones
measu ed o he esh educed ca alys s. P obably, he p esence o Ru weakens he Cu-Z
in e ac ion (see TPR esul s), a ou ing de Cu sin e ing (see XPS esul s). Due o he cha ac e is ic
21
o he RuCuZ ca alys , i can be expec ed highe DMF and DMTHF yields, bu a simila yields o
he ones co esoponding o he 15CuZ and 30CuZ ca alys s we e eached. This can be asc ibed
o he ac ha he u henium could no ca alysed he hyd ogenolysis eac ion as well as i was
es ed o he 1RuZ , whe e al hough a ull con e sion was eached, no DMF o DMTHF was
de ec ed. This means ha he e a e wo di e en main ou es ca alysed by Cu and Ru, and hey
compe e o he same molecule (HMF).
Acco ding o li e a u e [86,87], he mix u e o Z O2 and CeO2 has been ound o imp o e he
he mal s abili y, ca aly ic ac i i y and oxygen s o age capaci y. The la e e e s o he abili y o
deli e oxygen om he la ice o he gas phase o o solid (adso bed) ca bon. Based on he
gene al esul s, his e ec can be ela ed o a sligh imp o emen o s abili y o he ca alys s
suppo ed on Ce-Z compa ing wi h Z suppo ed ca alys s. To analyse i his a ec s o he
eac ion, Ce was added o he Z suppo .
As i can be obse ed in Table 8, he p esence o Ce on some ca alys s ha dly imp o ed he
selec i i y o DMF, bu in he case o he 30Cu5CeZ ca alys inc eased he s abili y o he 30CuZ
ca alys , educing he coke p oduc ion (C/Z used ca alys s in Table 7). Howe e , in he es o
XCuNCeZ ca alys s his e ec was negligible due o he low amoun o ce ium ha i was
inco po a ed o he Z s uc u e (see ICP-AES, XRD, SEM-EDX and XPS esul s). This low amoun
o imp egna ed Ce esul s on highe DMTHF yields (see Figu e S1).
The good beha iou egis e ed o he 15CuZ ca alys , when compa ed wi h he 15Cu5CeZ and
15Cu20CeZ ca alys s, could be due o wo easons: i) a be e me al dispe sion (see Table 5),
a ou ed by s onge me al-suppo in e ac ion obse ed by TPR, and ii) i s low acidi y (see Table
4), which educed coke o ma ion [88] (see Table 7). The expec ed e ec o he ce ium as
elimina o o solid adso bed ca bon did no ake place in he 15CuNCeZ ca alys s. In he case o
Z - and CeZ -suppo ed 30Cu ca alys s, he lowe DMF and DMTHF yield ob ained could be
ela ed o a lowe me al dispe sion and a highe acidi y, which a ou ed highe coke o ma ion.
Based on he li e a u e, Ni enables he clea age o C–O bond [89,90], so NiCu bime allic ca alys
was also p epa ed by he co-imp egna ion me hod. I can be obse ed (See Table 8) ha DMTHF
yield achie es a maximum o 40.2 % a e 4 h o eac ion, wi h a o al yield o DMF and DMTHF
o 53.5 %, p o iding he highes p oduc s yields. Acco ding o TPR and XRD cha ac e iza ion da a,
his ca alys p esen s a syne gic e ec o Ni and Cu h ough he o ma ion o a NiCu alloy species
(high esolu ion XRD). These ac s can be he explana ion o his highes yield, since he
li e a u e epo s ha NiCu alloy phases a e ex emely ac i e o hyd ogenolysis eac ions [91].
3.2.2. S abili y es o he 15Ni15CuZ ca alys
Since he 15Ni15CuZ ca alys showed he mos p omising esul s in e ms o DMF+DMTHF
yields, he s abili y o his ca alys was es ed measu ing i s ac i i y o longe imes on s eam.
Thus, as i can be obse ed in Figu e 9, his ca alys is s able o he p oduc ion o DMF, ob aining
a yield o 25% a e 24 h on s eam. Howe e , in he case o DMTHF, i s yield s a s o dec ease
a e achie ing a maximum yield o 40% a e 4 h on s eam. The s abili y o his ca alys could
be asc ibed o he syne gic e ec o Ni and Cu, which a oids any signi ican sin e ing e ec as i
can be obse ed analysing his ca alys da a de e mined by high esolu ion XRD. In addi ion o
his, he DMTHF yield dec eased a e 4 hou s on s eam, while he DMF yield inc eased a e
22
22 hou s on s eam, emaining almos cons an . I seems ha he 15Ni15CuZ ca alys los pa
o i s ini ial hyd ogena ion capaci y, and he e o e i s ac i i y o a he hyd ogena e DMF o
DMTHF dec eased along he ime on s eam. This loss o hyd ogena ion ac i i y could be ela ed
o he o nickel dispe sion dec ease, as XPS da a co obo a ed. I is well known ha Ni p esen s
a high hyd ogena ion capaci y, highe han he one o Cu. Tha is why he loss o nickel su ace
a ea could dec ease he ca alys hyd ogena ion capaci y. Mo eo e , he ec ys alliza ion o
NiCu obse ed in high esolu ion XRD esul s could also in luence his DMTHF and DMF
p oduc ion a ia ion.
Figu e 9. S abili y esul s o he 15Ni15CuZ ca alys ( DMF yield and DMTHF yield).
3.3. P elimina y pu i ica ion esul s
The bio uels p oduc ion no only include he de elopmen o e icien ca aly ic p ocesses o
con e biomass o biomass-de i ed building blocks in o hese bio uels. Fa he p ocessing is
equi ed o each he pu i y equi ed o he inal liquid uels o mula ions. In ou case, as all he
ac i i y expe imen s we e ca ied ou using n-bu anol as sol en and jus 1.5 w % o HMF as
eed, he p oduced DMF and DMTHF a e e y dilu ed in he sol en and a pu i ica ion ain is
equi ed. Fo ha pu pose, a dis illa ion p ocess a e he eac o ou le has been s udied by
means o p ocess simula ions using Aspen Plus. This simula ion has been simpli ied using DMF,
DMTHF and n-bu anol as eed in he p opo ions ob ained when ope a ing wi h he be
de eloped ca alys (15Ni15CuZ ). I can be concluded ha a qui e simple dis illa ion p ocess (20
equilib ium s ages, 3.8 o e lux a io and using 6.3 kW as eboile du y) is enough in o de o
ge a liquid uel composed by 21.4 mol% o DMF, 63.2 mol% o DMTHF and 15.0 mol% o n-
bu anol as he dis illa e. Addi ionally, n-bu anol liquid s eam o 99.99 mol% would be eco e ed
om he bo oms o be ecycled. I mus be indica ed ha bu anol is also a good bio uel i
p oduced by e men a ion. Howe e , i lowe n-bu anol concen a ion is equi ed in he uel,
he heigh o he owe as well as he eboile du y s a s inc easing conside ably. Table 9 shows
he uel p ope ies o he pu i ied uel wi h 15 mol% o n-bu anol. Mo eo e , i can be obse ed
2 4 6 8 10 22 24
0
10
20
30
40
50
60
DMF and DMTHF yield (%)
Time on s eam (h)
23
ha di e en DMF and DMTHF p oduc dis ibu ion in he uel does no a ec conside ably o
he uel p ope ies.
Table 9 Speci ica ions o h ee di e en DMF, DMTHF and n-bu anol composi ions
DMF mol% 0.21 0.64 0.43
DMTHF mol% 0.64 0.21 0.43
BUT mol% 0.15 0.15 0.15
Ce ane numbe 16.90 15.64 16.24
Oxygen con en (% weigh ) 16.8% 17.1% 16.9%
S oichiome ic ai / uel a io 11.43 10.82 11.13
Lowe hea ing alue (MJ/kg) 36.70 34.68 35.70
Densi y a 20 °C(g/cm3) 0.79 0.75 0.77
Viscosi y (mm2/s) a 40 °C 0.66 0.69 0.687
Boiling poin (°C) 95.26 96.48 95.86
4. CONCLUSIONS
DMF and DMTHF p esen good cha ac e is ics o hei aplica ion as high-quali y bio uels o
bio-addi i es o con en ional liquid anspo a ion uels. These compounds can be con inuously
p oduced om HMF using a ca aly ic ixed-bed eac o . Th ough an ex ensi e expe imen al wo k
ying o maximize s able DMF and DMTHF p oduc ion, he de eloped 15Ni15CuZ ca alys
showed a o al yield o DMF and DMTHF o 53.5 % a e 4 hou s on s eam, main aining only a
s able DMF yield o 25 % a e 24 hou s ope a ing a 15 ba , 275 ȗĂŶĚwi h no p e ea men
( educ ion) o he ca alys .
In o de o each comme cial p oduc ion some ca alys imp o emen s a e equi ed. F om his
wo k some u u e o ien a ions can be selec ed: i) zi conia is an adequa e suppo because o i s
high su ace a ea and limi ed acidi y, ii) he Cu dispe sion is a key pa ame e , he e o e i s
necessa y o imp o e he Cu dipe sion o inc ease he DMF yield and educe coke o ma ion, iii)
Ni addi ion gene a es he e y ac i e and s able NiCu solid alloy bu excess Ni a o s he
p oduc ion o o e -hyd ogena ion p oduc s such as DMTHFand i ) a noble me al inco po a ion
(Ru) o he zi conia modi ica ion using ce ia did no gene a e posi i e e ec s.
The bu hanol is an excellen sol en o his hyd ogenolysis p ocess ( om HMF o DMF) and i s
sepa a ion h ough dis illa ion is a compa a i ely cheap p ocess p o ided ha pa o he
bu hanol was also accep ed as a componen o he bio uel mix u e o be comme cialized.
Acknowledgemen s
This wo k was suppo ed by Uni e si y o he Basque Coun y (UPV/EHU), Spanish Minis y o
Economy and Inno a ion and Eu opean Union h ough he Eu opean Regional De elopmen
Fund (FEDER) (P ojec s: CTQ2015-64226-C3-2-R), and Basque Coun y Go e nmen (P ojec :
IT993-16). The au ho s also exp ess hei g a i ude o he Analysis Gene al Se ices (SGIke )
a ached o he UPV/EHU Uni e si y o i s assis ance wi h he SEM analysis, and o he Ped o J.
Mai eles (Uni e si y o Malaga) o i s assis ance wi h he XPS esul s.
24
REFERENCES
[1] Zhang M, Tong X, Ma R, Li Y. Ca aly ic ans o ma ion o ca bohyd a es in o 5-
hyd oxyme hyl u u al o e in phospha e in a wa e -con aining sys em. Ca al Today
2016;264:131–5. doi:10.1016/j.ca od.2015.06.031.
[2] Qian Y, Zhu L, Wang Y, Lu X. Recen p og ess in he de elopmen o bio uel 2,5-
dime hyl u an. Renew Sus ain Ene gy Re 2015;41:633–46.
doi:10.1016/j. se .2014.08.085.
[3] Nishimu a S, Ikeda N, Ebi ani K. Selec i e hyd ogena ion o biomass-de i ed 5-
hyd oxyme hyl u u al (HMF) o 2,5-dime hyl u an (DMF) unde a mosphe ic hyd ogen
p essu e o e ca bon suppo ed PdAu bime allic ca alys . Ca al Today 2014;232:89–98.
doi:10.1016/j.ca od.2013.10.012.
[4] Zu Y, Yang P, Wang J, Liu X, Ren J, Lu G, e al. E icien p oduc ion o he liquid uel 2,5-
dime hyl u an om 5-hyd oxyme hyl u u al o e Ru/Co3O4 ca alys . Appl Ca al B
En i on 2014;146:244–8. doi:10.1016/j.apca b.2013.04.026.
[5] In e na ional Ene gy Agency (IEA). Key wo ld ene gy s a is ics. Pa is: 2016.
[6] In e na ional Ene gy Agency. Wo ld Ene gy Ou look 2015. Pa is: 2015.
[7] In e na ional Ene gy Agency. Key Oil T ends. 2016.
[8] Yang P, Cui Q, Zu Y, Liu X, Lu G, Wang Y. Ca aly ic p oduc ion o 2,5-dime hyl u an om
5-hyd oxyme hyl u u al o e Ni/Co3O4 ca alys . Ca al Commun 2015;66:55–9.
doi:10.1016/j.ca com.2015.02.014.
[9] De S, Saha B, Luque R. Hyd odeoxygena ion p ocesses: Ad ances on ca aly ic
ans o ma ions o biomass-de i ed pla o m chemicals in o hyd oca bon uels.
Bio esou Technol 2015;178:108–18. doi:10.1016/j.bio ech.2014.09.065.
[10] G een Chemis y. F aunho e magazine 2.12. 2012.
[11] Rosa ella AA, Simeono SP, F ade RFM, A onso CAM. 5-Hyd oxyme hyl u u al (HMF) as
a building block pla o m: Biological p ope ies, syn hesis and syn he ic applica ions.
G een Chem 2011;13:754–93. doi:10.1039/C0GC00401D.
[12] U banchuk JM. Con ibu ion o he E hanol Indus y To he Economy o he Uni ed
S a es in 2014. 2015.
[13] I iondo A, Mendigu en A, Güemez MB, Requies J, Camb a JF. 2,5-DMF p oduc ion
h ough hyd ogena ion o eal and syn he ic 5-HMF o e ansi ion me al ca alys s
suppo ed on ca ie s wi h di e en na u e. Ca al Today 2017;279:286–95.
doi:10.1016/j.ca od.2016.02.019.
[14] Menon V, Rao M. T ends in biocon e sion o lignocellulose: Bio uels, pla o m
chemicals & bio e ine y concep . P og Ene gy Combus Sci 2012;38:522–50.
doi:10.1016/j.pecs.2012.02.002.
[15] Climen MJ, Co ma A, Ibo a S. Con e sion o biomass pla o m molecules in o uel
addi i es and liquid hyd oca bon uels. G een Chem 2014;16:516–47.
doi:10.1039/C3GC41492B.
[16] Choi S, Song CW, Shin JH, Lee SY. Bio e ine ies o he p oduc ion o op building block
chemicals and hei de i a i es. Me ab Eng 2015;28:223–39.
doi:10.1016/j.ymben.2014.12.007.
25
[17] Roman-Leshko Y, Ba e CJ, Liu ZY, Dumesic JA. P oduc ion o dime hyl u an o liquid
uels om biomass-de i ed ca bohyd a es. Na u e 2007;447:982–5.
[18] Zhao J, Zhou C, He C, Dai Y, Jia X, Yang Y. E icien dehyd a ion o uc ose o 5-
hyd oxyme hyl u u al o e sul ona ed ca bon sphe e solid acid ca alys s. Ca al Today
2016;264:123–30. doi:10.1016/j.ca od.2015.07.005.
[19] Zhang Y, Zhang J, Su D. 5-Hyd oxyme hyl u u al: Akey in e media e o e icien
biomass con e sion. J Ene gy Chem 2015;24:548–51.
doi:10.1016/j.jechem.2015.09.005.
[20] Kläusli T. AVA Biochem: comme cialising enewable pla o m chemical 5-HMF. G een
P ocess Syn h 2014;3:235–6. doi:10.1515/gps-2014-0029.
[21] T an LS, Si jean B, Glaude PA, Fou ne R, Ba in-Lecle c F. P og ess in de ailed kine ic
modeling o he combus ion o oxygena ed componen s o bio uels. Ene gy 2012;43:4–
18. doi:10.1016/j.ene gy.2011.11.013.
[22] B oca d JC, Ba onne F, O’Neal HE. Chemical kine ics o he oxida ion o me hyl e -
bu yl e he (MTBE). Combus Flame 1983;52:25–35. doi:10.1016/0010-2180(83)90118-
9.
[23] :ħǏĂŬ^njŝĚĂDŽƌħďƐŬŝD,ŝŐŚƉƌĞƐƐƵƌĞƉŚLJƐŝĐŽĐŚĞŵŝĐĂůƉƌŽƉĞƌƚŝĞƐŽĨϮ-
me hyl u an and 2,5-dime hyl u an – second gene a ion bio uels. Fuel 2016;184:334–
43. doi:10.1016/j. uel.2016.07.025.
[24] Chen G, Di L, Zhang Q, Zheng Z, Zhang W. E ec s o 2,5-dime hyl u an uel p ope ies
coupling wi h EGR (exhaus gas eci cula ion) on combus ion and emission
cha ac e is ics in common- ail diesel engines. Ene gy 2015;93:284–93.
doi:10.1016/j.ene gy.2015.09.066.
[25] Chen G, Shen Y, Zhang Q, Yao M, Zheng Z, Liu H. Expe imen al s udy on combus ion and
emission cha ac e is ics o a diesel engine ueled wi h 2,5-dime hyl u an-diesel, n-
bu anol-diesel and gasoline-diesel blends. Ene gy 2013;54:333–42.
doi:10.1016/j.ene gy.2013.02.069.
[26] Liu H, Xu J, Zheng Z, Li S, Yao M. E ec s o uel p ope ies on combus ion and emissions
unde bo h con en ional and low empe a u e combus ion mode ueling 2,5-
dime hyl u an/diesel blends. Ene gy 2013;62:215–23.
doi:10.1016/j.ene gy.2013.09.057.
[27] Zheng Z, Wang XF, Zhong X, Hu B, Liu H, Yao M. Expe imen al s udy on he combus ion
and emissions ueling biodiesel/n-bu anol, biodiesel/e hanol and biodiesel/2,5-
dime hyl u an on a diesel engine. Ene gy 2016;115:539–49.
doi:10.1016/j.ene gy.2016.09.054.
[28] Zheng Z, Xia M, Liu H, Wang X, Yao M. Expe imen al s udy on combus ion and emissions
o dual uel RCCI mode ueled wi h biodiesel/n-bu anol, biodiesel/2,5-dime hyl u an
and biodiesel/e hanol. Ene gy 2018;148:824–38. doi:10.1016/j.ene gy.2018.02.015.
[29] Shi J, Wang Y, Yu X, Du W, Hou Z. P oduc ion o 2,5-dime hyl u an om 5-
hyd oxyme hyl u u al o e educed g aphene oxides suppo ed P ca alys unde mild
condi ions. Fuel 2016;163:74–9. doi:10.1016/j. uel.2015.09.047.
[30] Con u so M, Si ignano M, Anna AD. E ec o 2,5-dime hyl u an doping on pa icle size
dis ibu ions measu ed in p emixed e hylene / ai lames. P oc Combus Ins 2016;In
p ess:1–8. doi:10.1016/j.p oci.2016.06.048.
