Catalytic routes towards acrylic acid

22 Pages·2017·1.91 MB·English

UvA-DARE (Digital Academic Repository)
Catalytic routes towards acrylic acid, adipic acid and epsilon-caprolactam
starting from biorenewables
Beerthuis, R.; Rothenberg, G.; Shiju, N.R.
DOI
10.1039/c4gc02076f
Publication date
2015
Document Version
Final published version
Published in
Green Chemistry
Link to publication
Citation for published version (APA):
Beerthuis, R., Rothenberg, G., & Shiju, N. R. (2015). Catalytic routes towards acrylic acid,
adipic acid and epsilon-caprolactam starting from biorenewables. Green Chemistry, 17(3),
1341-1361. https://doi.org/10.1039/c4gc02076f
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)
and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open
content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please
let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material
inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter
to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You
will be contacted as soon as possible.
UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
Download date:05 Jan 2023
Green Chemistry
View Article Online
CRITICAL REVIEW
View Journal | View Issue
8. Catalytic routes towards acrylic acid, adipic acid
1
6: ε †
2 and -caprolactam starting from biorenewables
0:
5 1 Citethis:GreenChem.,2015,17,
01 1341
2 RolfBeerthuis,GadiRothenbergandN.RaveendranShiju*
6/
0
9/
1
n The majority of bulk chemicals are derived from crude oil, but the move to biorenewable resources is
o
m gainingbothsocietalandcommercialinterest.Reviewingthistransition,wefirstsummarisethetypesof
a
erd today’s biomass sources and their economical relevance. Then, we assessthe biobased productions of
mst threeimportantbulkchemicals:acrylicacid,adipicacidandε-caprolactam.Thesearethekeymonomers
A
n forhigh-endpolymers(polyacrylates,nylon6.6andnylon6,respectively)andareallproducedgloballyin
a
eit v excessoftwomillionmetrictonsperyear.Thebiobasedroutesforeachtargetmoleculeareanalysedsep-
sit arately, comparing the conventional processes with their sustainable alternatives. Some processes have
ver alreadyreceivedextensivescientificattention.Other,morenovelroutesarealsobeingconsidered.Wefind
ni
U severalcommontrends:Forallthreecompounds,therearenocommercialmethodsfordirectconversion
y
b ofbiobasedfeedstocks.However,combinationsofbiotechnologicallyproducedplatformchemicalswith
d
de subsequentchemicalmodificationsareemergingandshowingpromisingresults.Wethendiscussseveral
a
o
nl distinctstrategiesforimplementingbiorenewableprocesses.Foreachbiotechnologicalandchemocataly-
w
o Received25thOctober2014, tic route, current efficiencies and limitations are presented, but we urge that these routes should be
D
4. Accepted18thNovember2014 assessedmainlyontheirpotentialandprospectsforfutureapplication.Today,biorenewableroutescannot
1
20 DOI:10.1039/c4gc02076f yetcompetewiththeirpetrochemicalequivalents.However,giventhatmostofthemarestillintheearly
ber www.rsc.org/greenchem stagesofdevelopment,weforeseetheircommercialimplementationinthenexttwodecades.
m
e
v
o
N
8
n 1 Van’tHoffInstituteforMolecularSciences,UniversityofAmsterdam,P.O.Box
o
d 94157,1090GDAmsterdam,TheNetherlands.E-mail:[email protected];
e
h http://hims.uva.nl/hcsc
s
bli †Electronic supplementary information (ESI) available: Summary of processes
u
P discussedinthisreview.SeeDOI:10.1039/c4gc02076f
Rolf Beerthuis has recently Gadi Rothenberg is Professor
obtained his MSc chemistry andChairofHeterogeneousCat-
diplomacumlaudeattheUniver- alysisandSustainableChemistry
sity of Amsterdam, focussing on at the Van’t Hoff Institute for
heterogeneous catalysis with a MolecularSciencesattheUniver-
minorinmolecularcomputational sity of Amsterdam, and teaches
chemistry. His thesis centered on catalysis, thermodynamics and
catalytic biomass conversion, scientific writing. He has pub-
namely the oxidation of lactic lished 150 papers in peer-
acid to acrylic acid. Currently, he reviewedjournalsanddiscovered
is investigating the conversion of two catalysts, for which he
glycerol to acrolein at the Lab received the Marie Curie Excel-
RolfBeerthuis of Advanced Materials at the GadiRothenberg lence Award in 2004 and the
Fudan University in Shanghai, Paul Rylander Award in 2006.
withProf.HualongXu.Hisresearchinterestliesincombiningknowl- He also co-founded the companies Sorbisense A/S, YellowDiesel
edgeofdifferentfieldstodevelopsustainablechemicalprocesses. BV,andPlanticsBV.In2007hewasvoted‘teacheroftheyear’by
thechemistrystudents.Hislatestinventionisa100%plant-based
cheapbiodegradableplastic.
Thisjournalis©TheRoyalSocietyofChemistry2015 GreenChem.,2015,17,1341–1361 | 1341
View Article Online
CriticalReview GreenChemistry
1. Introduction
Crude oil is currently the feedstock for manufacturing most
bulk and fine chemicals. This causes competition over the
available resources with the fuels for automotive and power
industry, creating fluctuating prices of chemical feedstocks
(Fig. 1).1,2 Combined with concerns over the environmental
impact of petrochemical processing, the chemical industry is
18. considering sustainable and more environmentally-friendly Fig.2 Overviews by weight percentage. Left: global fibre market, in
10:26: alternatives. The biorenewable production of many chemicals 2cl0a1s2se.sR,iginh2t:0c1o2.n7stituentsofsyntheticfibremarket,bygeneralpolymer
5 emits less greenhouse gases (GHGs) and employs more envir-
1
20 onmentally-friendlychemistry.3,4However,thetransitionfaces
6/
0 hightechnologicalandeconomicalbarriers.
n 19/ Here, we address this transition for three important bulk fibres, the largest part was embodied by polyesters and poly-
m o chemicals:acrylicacid,adipicacid,andε-caprolactam.Eachof olefins (Fig. 2, right), such as poly(ethylene terephthalate)
a (PET),polyethylene(PE)andpoly-propylene(PP).PETismade
d these is produced at over two million metric tons per annum
ster (Mtpa)withcurrentmarketpricesaround$1500perton. fromethyleneglycol(EG)andterephthalicacid(TPA).Though
m
A In 2012, more than 60% of all fibres produced worldwide biobased EG is commercially available, the non-availability of
van were synthetic materials6 (Fig. 2, left). Of these synthetic biobased TPA prevents the production of fully biorenewable
eit PET.7 Braskem produces 200000 tons of biobased PE in
sit Brazil,8 using ethylene obtained by dehydrating bioethanol.
er
niv However, biobased PE is currently still more expensive than
U
y petrobasedPE.Anemergingroutetowardsbiobasedpropylene
b
d is by producing ethanol and butane from sugars by fermenta-
e
ad tion,andthesubsequentdehydrationandmetathesisofethyl-
o
wnl ene and butene to propylene. However, developing process
o
D technologythatcaneconomicallycompetewithpetrobasedPP
14. isachallenge.9Thesemonomersareincorporatedintoagreat
0
er 2 many chemical and economical value chains. Their prices are
mb low:below$1500permetricton.10Conversely,thebulkchemi-
e
v cals that we will cover here are relatively expensive, ensuring
o
N
8 economicalmarginsforinnovativealternatives.
1
n Acrylicacidisusedformakingpolyacrylicacidandvarious
o
ed Fig.1 Annualaveragepricesforethylene,propyleneand1,3-butadiene acrylic esters, known for their superabsorbent properties and
blish in$perton.5 attractive properties in co-polymerization (Fig. 3). They are
u usedinarangeofsyntheticproducts,includingdiapers,plas-
P
tics, synthetic rubber, coatings and paint formulations.11
N. Raveendran Shiju obtained
his PhD in heterogeneous cataly-
sis from National Chemical Lab-
oratory,Pune.Hethenworkedat
the Royal Institution of Great
Britain, the Universityof Cincin-
nati and the University of Hud-
dersfield. In 2009 he accepted
the Assistant Professor position
at the University of Amsterdam.
Shiju now leads a team of PhD
students and post-docs, working
N.RaveendranShiju oncatalystdesignforgreentech-
nology. His research interests
range from materials science to reaction engineering. He teaches
greenandindustrialchemistry,functionalmaterialsandcatalytic Fig.3 Structures of acrylic acid, adipic acid, ε-caprolactam and their
reactionengineering. majorend-products.
1342 | GreenChem.,2015,17,1341–1361 Thisjournalis©TheRoyalSocietyofChemistry2015
View Article Online
GreenChemistry CriticalReview
Adipic acid and ε-caprolactam are used as monomers for 2. Implementing biorenewable
making nylon 6.6 and nylon 6, respectively (Fig. 3). These
chemicals
are the archetypes of polyamides, accounting for 85–90%
of the world nylon market. Polyamides are applied chiefly in
Therearevariousincentivesforapplyingbiorenewablesinthe
the fibre and textile industry and thus have competitive end-
chemical industry. Government regulations are putting
uses, yet dissimilar properties. In terms of performance,
pressure on chemical companies to make more environmen-
nylon6 hasbetter processabilityandresistancetowear,while
tally-friendly products. However, these companies can only
nylon 6.6 has better heat resistance and mechanical
18. properties.11 provide products that are commercially competitive. The dis-
6: cussion on using biomass for making chemicals is often
0:2 Over the last few decades, much research has gone into emotionally charged, giving the biobased industry the added
15 1 biorenewablechemicalsandchemicalbiomassutilization.12–20 value of the ‘bio’, ‘eco’, or ‘green’ label, which may make up
6/20 The growing interest in biorenewables focused mainlyon pro- foradditionalcostsforstartingupbiorenewableprocessesand
9/0 ducing platform chemicals, which can be applied in the syn- productswithanenvironmentally-friendlyimage.37
n 1 thesis of various compounds. It istherefore important to also Biorenewable chemicals are socially attractive. However,
m o reviewtheinfluenceof‘whitebiotechnology’orindustrialbior- their production will only be viable when it can compete
a
d efineries, on manufacturing bulk chemicals. For producing
mster acrylicacid,adipicacidandε-caprolactam,nocommercialbio- epcoosnsiobmleic–alblyiowmitahssthiespreetardoiblyasaevdaiolanbelse.,Tshtaisbliesfiunnsduapmpelyntaanlldy
A technological routes are currently employed. Emerging plat-
n (depending on type) can be cheap. What is more, biobased
eit va fvoiarbmlechreomutiecsa.lsHfroowmevbeiro,temchonstoloregsyemarcahypdreesaelnstwecitohnosmpeiccaiflilcy chemicalscanoftenbeproducedundermilderconditionsand
sit with less toxic reagents and waste, than the petrobased equi-
ver advancements,ratherthangivinganoverallview. valents,beingmore‘green’withlowerprocessingcosts.38
ni Toassessthecurrentprocessesandpossibleadvancements
U However, logistic considerations may determine the choice
y madewith biorenewable feedstocks, we first analyse the avail-
b of companies to produce their chemicals biobased or petro-
ded able biomass constituents and biobased feedstocks (Table 1). based. 1,3-Propanediol (1,3-PDO), for instance, is currently
a
o The benchmark prices are averaged across regions and qual-
nl manufactured via both pathways. At Shell, the hydroformyla-
w ities, giving a general impression of the availability of
o tion of ethylene oxide gives an intermediate, which is sub-
D
4. biobased feedstocks and their incorporation into chemical sequentlyhydrogenatedto1,3-PDO.Conversely,intheDuPont
201 valuechains. Tate & Lyle BioProducts process, 1,3-PDO is made from corn
ber Here, we will focus only on the technical analyses of the syrupusingmodifiedE.coli.DuPontclaimsthatthebiobased
m biorenewable routes and refrain from any economic analyses.
e process consumes 40% less energy and reduces GHG emis-
ov Full economical assessments33–36 are needed for reliable esti-
N sions by 20%, compared to the petrobased process. Despite
8 mations and conclusions. As rough economical estimations
1 this, there is no report on Shell adopting a biobased process.
d on areoftensubjective,wefeelthatthoseshouldbeavoided. Shellisthelargestproducerofethyleneoxide,with40%ofthe
he The combined results give a critical overview on the tran- globalproduction,atmultipleplantsworldwide.11Thoughthe
s
bli sition from petrobased to biorenewable productions of acrylic biobasedprocessisprovenviableandmoreeco-friendly,econo-
Pu acid, adipic acid and ε-caprolactam. To understand these
micsandlogisticsdominate.
developments, we will examine the biobased pathways, and
compare these to the petrochemical pathways. We use
examples of on-purpose reactions towards target molecules, 2.1. Platformchemicalsvs.chemicalmodificationof
focusingonthemostrecentandefficienttodate. biobasedfeedstock
Unlike crude oil, biomass is typically over-functionalized.
Thus, biobased feedstocks must be broken down to provide
Table1 Overviewofavailablebiomassfeedstocks basicchemical‘buildingblocks’orplatformchemicals.39Plat-
form chemicals offer the possibility of synthesizing various
Global Benchmark
end-products. However, biomass feedstocks may also be uti-
Biobased productiona pricea(U.S.$
feedstock Chemicalformula (Mtpa) perton) lized towards specific end-products with similar chemical
structures,byusingthealreadypresentfunctionalities.40
Starch21 Glucosepolymers 75 500
Glucose1,22 H–(CvO)–(CHOH) –H 175 500
5
Fructose23 H–(CHOH)4–(CvO)– 7 900 2.2. Top-downvs.bottom-up
(CH OH)
2
Ethanol2,24 CH3CH2OH 65 750 Someexistingchemicalprocessesmaybereplacedbycompeti-
Virginoils1,25 Varioustriglycerides 155 1100
Glycerol26–28 (CH OH)–(CHOH)– 1 850 tivebiorenewableprocessesto producethesame end-product.
2
(CH OH) The production of ethanol, for example, relies both on
2
Lysine29,30 C6H14N2O2 0.85 1900 microbial fermentation of sugars and hydration of ethylene.
Glutamicacid29,31,32 C H NO 1.6 1300
5 9 4 Process economics compete depending on feedstock prices.
aWorldwideproductionandpriceindexesfor2012. Suchapproachestobiorenewabilitycanbeseenas‘top-down’.
Thisjournalis©TheRoyalSocietyofChemistry2015 GreenChem.,2015,17,1341–1361 | 1343
View Article Online
CriticalReview GreenChemistry
However, biobased chemicals can also compete on a func- In 2012, around 4.5 Mt of acrylic acid was produced world-
tional basis. Biobased feedstock and platform chemicals may wide, with a growing demand of 4% per year. The current
offer novel compoundsthat cannot be made on acommercial marketpriceis$1600–$1800pertonforlow-gradeand$1900–
scale by petrochemical processes. These new market products $2200 per ton for glacial-grade. The Asia-Pacific consumption
mayofferaddedfunctionality,suchasbiodegradabilityorlow/ is about 46%, U.S. 27% and Western Europe 21%. Its major
no toxicity. One example of such a ‘bottom-up’ approach is producers are BASF, Dow and Arkema, but several other com-
replacing polyethylene terephthalate (PET) with biodegradable paniesalsoinvestinbiobasedprocesses.46
polyethylene furanoate (PEF) made of 2,5-furandicarboxylic
18. acid (FDCA) derived from hemicellulose, for making ‘green’
0:26: bottles.TheforerunnerinthisfieldisAvantiumTechnologies, ð3Þ
5 1 whichpartneredwithCoca-ColaintheYXYproject.Avantium’s
1
20 40 tpa pilot plant is scheduled to open in 2014 in the
06/ Netherlands.41 Mostoftheacrylicacidproductiontodayfollowsatwo-step
n 19/ Another example is polylactic acid (PLA), produced by energy-intensive gas-phase process.11,47 Herein, propylene, a
o side-product of the ethylene and gasoline production, is first
m NatureWorks under the product name Ingeo. This is the first
a oxidized to acrolein using a Bi/Mo–O catalyst at 320 °C. Then
d biopolyester made on an industrial scale (140 ktpa). Commer-
ster cial application relies on added functionalities of the novel thereactionmixtureisdirectlyconvertedintoacrylicacidina
Am polymer. High efficiencyenables competitive economics, with secondreactor,usingaBi/V–Ocatalystat280°C(eqn(3)).
n
va every2.5kgcorn(15%moisture)yielding1.0kgPLA.42
eit Avery recent development from our group isthe invention
ersit of Glycix – a thermoset resin made from glycerol and citric 3.2. Alternativebiorenewableprocesses
v
ni acid, that is now being commercialized in the Netherlands.43 Here, we will discuss the most recent and noticeable alterna-
U
y Inthiscase,theaddedvalueofthebiorenewablepolymerlies tive routes towards acrylic acid. Some advanced processes
b
ed in its biodegradability and strong adhesive properties, that includeconvertingglycerol,butalsousingplatformchemicals
d
oa enable the formation of superior composites.44 This polymer that are already produced on a large scale, such as lactic
nl
w isnowbeingcommercializedbyPlanticsB.V. acid and acrylonitrile. We will also review novel routes, using
o
D
014. 2.3. Newbiorenewableroutesvs.intersectingexisting eamnder2g-iancgetpolxaytpforrompiocnhiecmaicciadls. sFuigc.h4asgi3v-ehsydarnoxoyvperrovpieiowniocfatchide
2
er chemicalvaluechains conventional petrobased routes in grey, and the alternative
b
em Most novel routes cannot compete with existing technologies routes based on biorenewable platform chemicals in light
v
No because those are highly optimized. Instead of direct compe- blue.
18 tition, parts of existing processes may be adapted. As such, 3.2.1. Production of biorenewable propylene. Several com-
on biobased intermediates may support established routes. This panies are investing in the biobased production of propylene.
d
he combines proven and optimized routes with biorenewable Global Bioenergies, for example, produces isobutene from
blis feedstocks.However,manyexisting‘green’alternativesareready glucose and is looking to expand its process to propylene.48
u
P tobeexploitedwhenenvironmentalrestrictionsdemandit.45 Another pathway to biopropylene is through converting
3. Acrylic acid
3.1. Introduction
Acrylic acid is a versatile monomer and intermediate, with
major end-uses as acrylic esters for superabsorbent polymers
(55%)andplasticsandsyntheticrubber(30%).Theremainder
is used in the manufacture of coatings, paint formulations,
andleatherfinishing(eqn(1)and(2)).
Fig.4 Productionroutestoacrylicacid,showingbiobasedfeedstocks
(green), biobased platform chemicals (light blue), and existing petro-
basedroutes(grey).
1344 | GreenChem.,2015,17,1341–1361 Thisjournalis©TheRoyalSocietyofChemistry2015
View Article Online
GreenChemistry CriticalReview
bioethanol. Iwamoto et al. reported this route, using a scan- Purac, Galactic, and several Asian companies.54,55 The major
dium-loadedIn O catalystat500°C,giving60%yield.49 end-usein2012wastheproductionofPLAatnearly200ktpa.
2 3
3.2.2. Glycerol to acrylic acid. Today, glycerol is mainly Bacterial routesto lactic acidaccount for >90%of all lactic
produced as a biodiesel by-product from the trans-esterifica- acid production, using Lactobacillus acidophilus and Strepto-
tion of triglycerides to fatty acid methyl ester (FAME). This coccusthermophilusbacteria(eqn(7)).Generally,starchisused
process co-generates approximately 10% glycerol by weight.26 asafeedstockandyieldsaregreaterthan90%.
Itscurrentglobalproductionisaround1Mtpa,withamarket Lactic acid may also be synthesized chemically from other
price of around $850 per ton.27,28 For converting glycerol to biobased feedstocks, such as glycerol or hexoses via triose
18. acrylicacid,bothdirectconversionbyasinglecatalyst,aswell derivatives.ArecentexampleisgivenbyChaudharietal.,56by
6:
0:2 as combinations of multiple catalysts are known. The latter reacting glycerol in the presence of Cu2O and 1.5 equivalents
1
15 mayutilizeone-potprocessesorconsecutivereactorbeds. of NaOH,in H2O under14bar N2 at 200 °C. Within6 h, 95%
20 In 2012, Chieregato et al.50 showed a robust V–W–Nb- conversionisreached,withaselectivityof80%andprovenre-
6/
0 based catalytic system, composed either mainly of vanadium usabilityofthecatalyst(eqn(8)).
9/
n 1 or niobium. Complete conversion was observed with 34%
o
m acrylicacidyieldand17%acroleinco-productformation.After
a
d 100 h on stream, the acrylic acid yield was reduced from 34%
er
st to 31%, while acrolein formation rose from 17% to 21%,
m
A retaining 51% overall combined yield of acrylic acid and acro-
n
va lein(eqn(4)).
eit
sit
er
v
ni
U
y
b
d
e
d
a
o
nl
w
o
D
4.
01 Withincreasedresearchintoutilizingcheaperfeedstockssuch
2
er as molasses and whey waste-streams or crude lignocellulose,
b
m the production of lactic acid is expected to grow.57 Foracom-
e
v
o prehensiveoverviewofthepositionoflacticacid,seeDusselier
N
8 etal.58
1
on The dehydration of lactic acid to acrylic acid proceeds by
d
e abstracting a hydroxyl group and proton, giving the vinyl
h
s
bli double bond. The reaction proceeds via a carbocation at the
Pu In 2011, Witsuthammakul et al.51 described a process using a carbonyl α-position. This means that decarboxylation ensues
single reactor with two consecutive reactor beds. First, the readily.Athightemperatures,thisreactionsuffersfromlactide
complete conversion of glycerol with 81% selectivity to acro- formation and decomposition to acetaldehyde, CO and water.
lein was recorded over a ZSM-5 reactor bed at 300 °C. Sub- Furthermore, inhibiting oligomerization is important for
sequently, a V–Mo–O/SiO catalyst bed afforded 48% maintaininghighselectivity.11
2
conversion with 98% selectivity. The combined catalytic Experiments in supercritical or near-critical water showed
systemgave38%overallyield(eqn(5)). that adding H SO increased lactide and acetaldehyde for-
2 4
Another patent, from Dubois and co-workers at Arkema,52 mation, while NaOH increased selectivity to acrylic acid.59
describedtheconversionofglyceroltoacrylicacidusingatwo- Moreover, adding Na HPO increased acrylic acid yield from
2 4
bed oxydehydration reaction, in the presence of molecular 35% to over 58%.60 Experiments at high temperature (450 °C)
oxygen.Optimalresultswerefoundforthefirstbedwith91% and pressure (400–1000 bar) showed that the latter promotes
ZrO –9% WO and for the second bed with a multi-metallic bothconversionandselectivity.61
2 3
catalyst53 (Mo V Sr W Cu O ). Full conversion and 75% The highest yield was reported by Ghantani and co-
12 4.8 0.5 2.4 2.2 x
overall yield were obtained at 280 °C. These results seem workers,62,63whoobtainedfullconversionand78%yield,con-
impressive, yet catalyst stability and re-use were not disclosed vertinglacticacid(25wt%feed)overacalciumpyro-phosphate
(eqn(6)). catalystat375°C,withaWHSVof3(eqn(9)).Adetailedover-
3.2.3. Lactic acid to acrylic acid. In 2012, the global pro- view of this reaction is published elsewhere.64 However,
duction of lactic acid was estimated at 300–400 ktpa, with an the yield is lower with feeds containing high concentrations
existing capacity of over 500 ktpa. The current market prices of lactic acid. For commercial application, this has to be
range from $1300 per ton (50% purity) to $1600 per ton (88% improved. Moreover, acrylic acid yield should be high at high
purity). Its major producers are NatureWorks LLC & Cargill, spacevelocities.
Thisjournalis©TheRoyalSocietyofChemistry2015 GreenChem.,2015,17,1341–1361 | 1345
View Article Online
CriticalReview GreenChemistry
ð13Þ
ð9Þ
A sustainable alternative for the production of acrolein
startsfromglycerol(eqn(13)).Thedehydrationofglycerolcan
Anotherinterestingroutetoacrylicacidcomesfromacetoxy- be done in the gas phase, the liquid phase and the (near)
lation of lactic acid towards 2-acetoxypropionic acid (2-APA) supercritical phase,67 using either homogeneous or hetero-
8. and its subsequent pyrolysis. Currently, there are no commer- geneous catalysts.68,69 Recently, Liu and co-workers obtained
1
26: cial processes using 2-APA. For this, the traditional acetic high yields using rare earth metal-pyrophosphates. Their best
0:
1 anhydride route is unsuitable because lactic acid is mostly result, 96% conversion with 83% selectivity, was attained at
5
201 available in aqueous solution. To overcome this, inexpensive pH 6, using an Nd4(P2O7)3 catalyst calcined at 400 °C.70 Pre-
6/ acetic acid may be used also as a solvent. The conversion of viously, we reported the dehydration of glycerol to acrolein
0
on 19/ flaucrtiiccaacciidditnoy2ie-AldPsAowvaesrr9e0p%or(teeqdnb(y1L0)i)lg.6a5etal.,usingconc.sul- soevleerctNivbit2yO5d/SepiOe2ndcaotanlysthtse, snhioobwiiunmg tlhoaatditnhge acnodnvecraslicoinnatainond
m
a temperature.71 Elsewhere, de Oliveira and co-workers72
d
ster investigated liquid-phase glycerol dehydration using various
m
A zeolite catalysts. They found that the catalytic activity was not
an directlycorrelatedtotheSi/Alratio.However,thecatalyststruc-
v
eit ture and porosity, and the strength of acid sites were
sit determiningfactors.Usingamordenitecatalyst,theyobtained
er
niv 92% conversion and full selectivity after 10 h at 250 °C
U
y (eqn(14)).
b
d
e
d
a
o
nl
w
o
D
4.
1
0
2
er Frucheyetal.claimthat2-APAmaybeproducedquantitatively
b
m from lactide and acetic acid, using nickel acetate, nickel
e
ov nitrate and phenothiazine at 250 °C (eqn (11)).66 Lactide is a
N
8 common by-product in reactions with lactic acid. Its valorisa-
1
on tion is crucial for cost-effective processes. Under certain con-
d
e ditions, this cyclic dimer shows enhanced activity over the
h
blis monomer to acrylic acid.66 On-purpose dimerization is typi-
u
P cally done in two steps. First, monomer condensation is The use of heteropolyacids (HPAs) was intensively researched
achieved by removing water at temperatures above 200 °C. in the last decade.73–75 Haider et al.76 reported the use of
Then,thedimeriscyclizedthermally,orbyacidcatalysis. a CsSiW O /Al O catalyst in a continuous flow reactor
12 40 2 3
(eqn (15)). They obtained full conversion and 96% selectivity
towards acrolein, after 3 h at 250 °C. HPAs can offer higher
Brønsted acidity than mineral acids, but suffer significant
ð12Þ
limitationsduetocatalystinstability.Severalrecentreviewson
thistopichavebeenpublished.26,77,78
Various catalysts are known for converting acrolein to
It was suggested that 2-APA readily undergoes pyrolysis at acrylic acid. Here, we focus on the popular Mo–V–O and Mo–
around 95%yield.47,66Thisreactionismoreselectivethanthe V–M–O(M=W,Cu,Nb,Te)typematerials.
directdehydrationoflacticacid,asitdoesnotinvolveacarbo- Asearlyasin1967,Kitaharaetal.presentedtheconversion
cation(eqn(12)). ofacroleintoacrylicacidusingV–Mo–Ocatalysts,synthesized
3.2.4. Acrolein to acrylic acid. Currently, acrolein is used from MoO , V O , Al O precursors in the respective ratio of
3 2 5 2 3
asaprecursorforarangeofderivatives,suchasacrylates,acrylo- 8:1:0.4at17.8%(byweight)supportedonspongyaluminum.
nitrile and acrylamide. It is usually not isolated, but used as Using O and steam at 200 °C, they attained 97% conversion
2
anintermediateandreactedtothedesiredend-products.Most with 86% selectivity to acrylic acid.79,80 In 1974, Tichý et al.
ofthecurrentcommercialprocessesdependongas-phaseoxi- improved the efficiency using an Mo–V–O catalyst supported
dationofpropylene.Theseprocessesgenerallyattainonly20% on SiO aerosil (30% by weight), with an Mo:V ratio of 5:1,
2
conversion and 70–85% selectivity and depend on intensive in the presence of molecular oxygen and steam at 180 °C.
propylenerecycling. The complete conversion of acrolein was observed with 96%
1346 | GreenChem.,2015,17,1341–1361 Thisjournalis©TheRoyalSocietyofChemistry2015
View Article Online
GreenChemistry CriticalReview
selectivity towards acrylic acid. However, little is known about by reacting 0.2 M 3-HPA solution with HCl (35%) at pH 2, at
its catalyst stability and re-use.81 The reaction mechanisms, roomtemperaturein1h(eqn(20)).84
kinetics,andtheeffectsofpromotersarereviewedelsewhere.82 3.2.6. 3-Hydroxypropionic acid (3-HP) to acrylic acid.
Another potential platform chemical is 3-hydroxypropionic
acid(3-HP),the β-isomeroflacticacidandthecarboxylic acid
ð16Þ derivative of 3-HPA. Many fermentation routes can produce
this compound (eqn (21)).88 Current yields from glucose are
too low for industrial application at high concentrations;
8.
1 although coupled fermentation with co-reactions may
10:26: yielRdesc,enutlsyi,ngAokainanMdoc–oV-–wWor–kCeurs–Oachciaetvaeldysthigshupapcorrytleicdacoind overcome this problem.89 The biobased production of 3-HP is
5 currentlynotcommercialized.However,inJuly2013,aconsor-
201 α-alumina in a fixed bed reactor. They obtained 98% conver- tium of BASF, Cargill and Novozymes successfully demon-
6/ sion of acrolein and 90% yield of acrylic acid at 280 °C
0 strated 3-HP production on a pilot scale. In September 2014,
n 19/ (eqn(16)).83 the same consortium announced the successful conversion of
m o 3.2.5. 3-Hydroxypropionaldehyde (3-HPA) to acrylic acid. 3-HP to glacial acrylic acid and superabsorbent polymers.90
a Another viable route to acrylic acid starts from 3-hydroxy-
d Moreover, this process was selected for a further scale-up.
ster propionaldehyde(3-HPA).Currently,twocommercialprocesses In 2013, another consortium of OPX Biotechnologies and
m
A produce 3-HPA as an intermediate for 1,3-PDO.11 In the Dow Chemical, announced the successful fermentation of
van Degussa process, propylene is transformed into acrolein, 3-HP in 3 thousand litre (kl) capacity en route to biobased
eit whichishydratedto3-HPA(eqn(17)).Furtherreductionyields acrylic acid. The consortium is now scaling up the process to
ersit 1,3-PDO at 43% overall yield, but product separation is costly. 20–50kl.91
v
ni In contrast, the Shell process relies on ethylene oxidation to
U
y ethylene oxide, its hydroformylation to 3-HPA under 150 bar
b
d and subsequent reduction to 1,3-PDO at 80% overall yield.
e
ad However, the efficiencies for the intermediate steps are not
o
wnl given(eqn(18)).
o
D
4.
1
0
2
er
b
m
e
v
o
8 N A different biobased approach to 3-HP is via fermentation of
1
n glycerol(eqn(22)).Recently,Kimetal.showeddirectbiotrans-
o
ed formation using Klebsiella pneumoniae. Conversion is 100%,
blish but 3-HP selectivity is only 11% mol mol−1. The main by-pro-
Pu ductsare1,3-PDO(47%)andaceticacid(18%).92
The dehydration of 3-HP to acrylic acid shows high yields
for various conditions and catalysts. A recent example was
patented by Ciba Specialty Chemicals.93 The best results
were obtained for a 20% aqueous solution over SiO yielding
2
97%, and 60–80% aqueous solutions over high surface area
Theenzymaticconversionofglycerolto3-HPAwasreportedin
2008, with yields up to 98% mol mol−1. The biorenewable γ-alumina, also yielding 97–98%. Reactions proceeded at
250 °C, with the complete conversion of 3-HP (eqn (23)).
route outperforms petrochemical routes,84 but is not yet com-
The difference in selectivity between lactic acid and 3-HP is
mercialized(eqn(19)).Detailsontheenzymaticproductionof
attributedtotheeliminationmechanisms.
3-HPAcanbefoundelsewhere.85
Theoxidationof3-HPAtoacrylicacidisaninterestingbio-
basedalternative,butnodirect (bio)chemicaltransformations
areknownatpresent.85–87 ð23Þ
ð20Þ
3.2.7. Acrylonitrile to acrylic acid. Acrylonitrile is a highly
desired bulk chemical and a potential biorenewable platform
chemical. In 2012, its production was around 6.0 Mtpa, with
Conversely,3-HPAmaybeconvertedwithhighefficiencyto a market price of $1600–$2000 per ton.94 Currently, it is
acrolein. In 2008, Toraya et al. reported 97% yield of acrolein produced predominantly by the SOHIO process. Herein,
Thisjournalis©TheRoyalSocietyofChemistry2015 GreenChem.,2015,17,1341–1361 | 1347
View Article Online
CriticalReview GreenChemistry
propeneisconvertedovera[Bi–Mo–O]catalyst,inthepresence 3.3. Acrylicacid–summaryandanalysis
of air and ammonia, at 400–500 °C. The direct conversion
givesover70%yield.7,95–97 Thepetrochemicalsynthesisofacrylicaciddependsonproces-
The direct ammoxidation of glycerol to acrylonitrile has sing propylene. The price of propylene has fluctuated greatly
only seen few publications. The most noticeable came from inrecentyears(risingabove$1300perton).Substitutingpetro-
Bañaresandco-workers98in2008.TheyusedaV–Sb–Nb/Al O based propylene with its biobased equivalent provides a bio-
2 3
catalyst, reaching 83% conversion and 58% selectivity, at renewable pathway to acrylic acid. This approach preserves
400°C.Thesamegroupalsoreportedasolvent-freemicrowave existingproductionprocessesandallowstheindustrytoadapt
8.
1 irradiationreactionat100°C,giving47%conversionwith80% more easily to biorenewability. Propylene may be produced
6:
0:2 selectivity within 1 h. Although the activity is modest, these fromethanol(around$750perton)at60%yield.Byimproving
5 1 conditions are mild, solvent-free, and use inexpensive bio- the efficiency, this route may soon become commercially
1
20 basedfeedstocks(eqn(24)).99 competitive.
6/
0 To obtain platform chemicals via fermentation, starch and
9/
n 1 glucose are typically observed as microbial feedstocks. These
o
m are cheap feedstocks (around $500 per ton) and thus provide
da large economic margins towards acrylic acid ($1600–2200
er
st perton).
m
A The efficient production of 3-hydroxypropionic acid from
n
va glucose is emerging rapidly, and commercialization is
eit envisioned in the coming years. Moreover, the dehydration of
sit
er 3-hydroxypropionic acid gives near quantitative yield. With at
v
ni least two important industrial consortia showing promising
U
y results,thisrouteseemstobecommerciallyviable.
b
ed Acrylonitrile hydrolysis to acrylic acid was demonstrated at
d
oa highefficiency(over90%),inbothchemocatalyticandbiotech-
nl
w nological processes. Converting glutamic acid shows full con-
o
4. D version, but suffers from selectivity issues (12% overall).
01 Moreover,thecurrentglutamicacidfeedstockprice(ca.$1300
2
er perton)makesthisroutefarfromeconomicallyviable.
b
m Glycerol is an attractive biobased feedstock for producing
e
v
o Recently,LeNôtreetal.showedthatacrylonitrilecanbemade acrylic acid. As a by-product from the biodiesel industry, its
N
8 fromglutamicacidintwosteps.Glutamicacidisreadilyavail- price(around$850perton)isexpectedtolowerinthecoming
1
on able from biomass and is an industrial waste-product (e.g. years. Its continuous reaction with acrolein shows high yield
d
e from bioethanol production). However, most glutamic acid is (96%yield).Thesubsequentacroleinconversiontoacrylicacid
h
s
bli currently produced by fermentation using Corynebacterium occurswith90%yield.Inthecombinedprocess75%yieldwas
u
P glutamicum.31Thefirststepinconvertingglutamicacidtoacrylo- obtained. This provides an economically viable pathway, but
nitrileisitsoxidativedecarboxylationto3-cyanopropanoicacid has not yet been commercially applied. Another pathway to
(70%isolatedyieldin1h).Thesecondstepisthedecarbonyla- acrolein is via the biocatalytic production of 3-hydroxypropio-
tion/eliminationreaction,yielding17%ofacrylonitrilein18h naldehyde from glycerol (98% yield). Its subsequent conver-
(eqn(25)).100Eveninthepresenceofthehydroquinonestabil- sion produces acrolein at 97% yield. The theoretical acrylic
izer, reactant degradation and product polymerization are acid yield is 86%, in three steps. However, the combined
thoughttocauseanoveralllowyield. processisnotyetreported.Mostofthestudiesonglycerolcon-
The hydrolysis of acrylonitrile to acrylic acid is one of the version are done with a refined feed. Additional studies need
conventional routes to acrylic acid, adopted by Mitsubishi to be performed on the catalytic performance and stability,
Petrochemical, Asahi Chemical and others. However, reaction when crude glycerol is used asthe feed. In general, crude gly-
with H SO gives stoichiometric NH HSO waste. The more cerol contains light solvents (water, methanol, and/or
2 4 4 4
recent Mitsui Toatsu process uses only water for conversion ethanol), fatty acid methyl esters, free fatty acids and ash.
over a B O -based catalyst. Specific details on reaction con- Since biodiesel production methods vary significantly, the
2 3
ditionsandyields arenot given, butcompleteconversionand compositionofcrudeglycerolalsovarieswidely.
ca.90%selectivityisexpected.11,101 Comparedtoglycerol,lacticacidismoreexpensive(around
Thefirstreportsofthebiotransformationofacrylonitrileto $1600 per ton(88% purity). However, bacterialroutesto lactic
acrylicacidcamein2010,usingRhodococcusruberbacteria.102 acid show high yields (around 90%). We expect that the
Under optimal conditions, using purified nitrilase, 92% mol expanded production and improved biotechnology will lower
mol−1 yield was achieved. Continued research is being per- lactic acid prices in the coming years. The dehydration of
formed towards its optimization and scale-up conditions. lacticacidshows selectivityissuesduetotheinstabilityofthe
Sincethen,variousbiotransformationshavebeenreported.103 intermediate. A possibility to overcome this problem is by
1348 | GreenChem.,2015,17,1341–1361 Thisjournalis©TheRoyalSocietyofChemistry2015
View Article Online
GreenChemistry CriticalReview
using derivative chemicals, such as 2-acetoxypropionic acid. waste. Noyori and co-workers109 developed in 1989 a halide-
This route is still limited to homogeneous catalysis and lacks freebiphasicprocessforthedirectoxidationofcyclohexeneto
processingconditions.Nevertheless,itisworthstudying,since crystalline adipic acid, using a phase-transfer catalyst in the
the pyrolysis of 2-acetoxypropionic acid is reported to lead to presence of 30% aqueous H O . This gave adipic acid at 90%
2 2
acrylicacidefficiently. yield,albeitafter8h.
8. 4. Adipic acid ð28Þ
1
6:
2
0: 4.1. Introduction
1
15 Adipic acid is mainly used for the manufacture of nylon Freitag et al.110 then improved this biphasic system using
0
6/2 6.6 (eqn (26)). Its polycondensation with hexamethylenedi- an Na2WO4 catalyst and microwave radiation, reducing the
0
9/ amine(HMDA)towardsnylon6.6 accountsforaround 85%of reactiontime to90min with68% yield(eqn(28)).Comparing
1
on all the adipic acid produced, with the remainder is used for theroutes,thedirectoxidationsaremoreeco-friendly,butsub-
am polyurethanesandadipicesters.11 strate prices and technical challenges still limit their
d
er implementation.
st
m
A
n 4.2. Alternativebiorenewableprocesses
a
v
eit Here, the most recent and noticeable biorenewable routes
sit towards adipic acid will be discussed. Some advanced routes
niver ð26Þ include pathways via muconic acid, glucaric acid and
U
y 5-hydroxymethylfurfural, all obtained from sugars. We also
b
d include the conversion of levulinic acid and 1,4-butanediol.
e
ad Fig. 5 summarizes both the conventional petrobased routes
o
nl towards adipic acid in grey, and the alternative biorenewable
w
Do routesinlightblue.
014. withIna2g0r1o2w,itnhgedpermodauncdtioonf 3o–f5%adippeircyaecaidr.wThaseacruorurenndt2m.3arMkett, 4.2.1. Production of biorenewable KA-oil. Converting
ber 2 price is $1500–$1700 per ton, and its major producers are lbiigonriennetowapbhleenoplasthanwdaythteon tKoAc-oyicll.o40hexSaenveornael iaspapnroiancthereesstifnogr
m Invista, DuPont, Rhodia, Ascend and BASF.104 Commercial
ve ‘cracking’ lignin are being pursued, such as hydrogenation,
No interest in biorenewable routes to adipic acid is found in
18 the plans of both major and start-up chemical companies i.e. hydrolysis and thermal cracking, to yield a mixture of substi-
n tuted phenols, (Scheme 1) which can be converted by de-
o BioAmber, Ronnavia, Genomatica, DSM, Celexion and
d
he Verdezyne.
s
bli In 2012, more than 90% of the global adipic acid pro-
u
P ductionreliedonthenitricacidoxidationofcyclohexanolora
mixture of cyclohexanol–cyclohexanone (KA-oil), all derived
from petrobased benzene (eqn (27)).11,105 This process gener-
atesnitrousoxidewaste.Consequently,developinglesspollut-
ing,more‘green’routeshasbecomeanimportantmatterand
has already seen large improvements. Here we outline the
mostrelevantcurrentroutes.Acomprehensiveoverviewispub-
lishedelsewhere.106
ð27Þ
Fig.5 Outline of the production routes to adipic acid, showing bio-
In 1975, an alternative route107,108 to adipic acid used the basedfeedstocks(green),biobasedplatformchemicals(lightblue),and
hydrocarboxylation of 1,3-butadiene, giving no nitrous oxide existingpetrobasedroutes(grey).
Thisjournalis©TheRoyalSocietyofChemistry2015 GreenChem.,2015,17,1341–1361 | 1349