DOI: 10.1002/cssc.201200524

Catalytic Synthesis of Hydroxymethyl-2-oxazolidinonesfrom Glycerol or Glycerol Carbonate and UreaAngela Dibenedetto,*[a] Francesco Nocito,[a] Antonella Angelini,[a] Imre Papai,[b]

Michele Aresta,[c] and Raffaella Mancuso[d]

Introduction

2-Oxazolidinones are important heterocyclic compounds thatfind applications in several fields of the pharmaceutical andchemical industry. In particular, they have long been used asantibiotics[1] or pesticides[2] and in drug[3] and fiber produc-tion.[4] In recent years, polymers containing 2-oxazolidinoneshave been developed for applications as foams, adhesives, andfibers.[5]

Industrially, 2-oxazolidinones are synthesized by phosgena-tion of the corresponding 1,2-amino alcohols[6, 7] (Scheme 1a);this is a non-ecofriendly route that calls for alternatives.

Only a few greener synthetic procedures have been pro-posed that avoid the use of toxic phosgene, including the re-action of 1,2-amino alcohols with CO2

[8] (Scheme 1b) or CO/O2

[9] (Scheme 1 c). Alternatively, 2-oxazolidinones can be syn-thesized from aziridines and carbon dioxide[10] or by treatingcyclic carbonates with 1,2-amino alcohol.[11–13] All such routesare affected by drawbacks such as 1) the formation of poly-

mers that generate a low yield of the target product, 2) unfav-orable reaction conditions (high pressure, high temperature),and 3) high costs.

Herein, we report the synthesis of hydroxymethyl-2-oxazoli-dinones by reacting glycerol carbonate and urea (Scheme 2) or

Oxazolidinones have been synthesized by reacting glycerol car-bonate or glycerol with urea in the presence of g-Zr phosphateas a catalyst. The conversion yield of the polyol or its carbon-ate depends on the temperature. Below 408 K the selectivity is100 % with a conversion of up to 25 %, whereas increasing thetemperature means that conversion yield grows, but the selec-tivity decreases, which makes the separation process more dif-ficult. Starting from glycerol carbonate, two isomers, 6 and 6’,are formed with a quasi 1:1 molar ratio because urea can

attack the carbonate moiety on both sides of the carboxylicCO moiety. From glycerol the formation of the 6’ isomer is pre-ferred: the ratio of 6’/6 is close to 7. The oxazolidinonesformed act as templates because they interact through hydro-gen bonding with glycerol. The intensity of the interaction de-pends on the 6 or 6’ isomer: DFT calculations showed that theenergy was 22.6 kcal mol�1 for 6-oxazolidinone and 25.7 kcalmol�1 for 6’-oxazolidinone.

Scheme 1. The “phosgene-based” route (a) to the synthesis of 2-oxazolidi-none and two alternative “phosgene-free” routes (b and c).

Scheme 2. Route for the synthesis of 2-oxazolidinones from glycerol carbon-ate.

[a] Prof. A. Dibenedetto, Dr. F. Nocito, Dr. A. AngeliniDepartment of Chemistry and CIRCCUniversity of Bari, Campus Universitariovia Orabona 4, 70126 Bari (Italy)Fax: (+ 39) 080-5443606E-mail : a.dibenedetto@chimica.uniba.it

[b] Dr. I. PapaiCIRCC, University of BariVia Celso Ulpiani 27, 70126 Bari (Italy)

[c] Prof. M. ArestaChemical Research CenterHungarian Academy of Sciences1525 Budapest, P.O.B. 17 (Hungary)

[d] Dr. R. MancusoDipartimento di ChimicaUniversit� della Calabria87036 Arcavacata di Rende (CS) (Italy)

� 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2013, 6, 345 – 352 345

CHEMSUSCHEMFULL PAPERS

glycerol and urea (Scheme 3). Glycerol carbonate is a nontoxicand easily biodegradable compound that can be prepared inhigh yields and at low cost under mild reaction conditions byglycerolysis of urea.[14] Urea can be considered as a reagent forindirect carbon dioxide utilization.

We have used g-Zr phosphate (g-ZrP)[14, 15] as a catalyst forboth processes and show that modulation of the acid/baseproperties of the catalyst plays a key role in the transformationof glycerol carbonate or glycerol into 2-oxazolidinone.

Such an approach represents, on the one hand, a new appli-cation of glycerol carbonate or glycerol and, on the otherhand, a new “green” route to oxazolidinones. Note that theonly byproducts are ammonia and carbon dioxide (Scheme 2),which can be collected and reprocessed to afford urea, andthus, ensuring good atom economy of the process, or ammo-nia and water (Scheme 3).

Results and Discussion

Glycerol carbonate and urea in the presence of uncalcined g-ZrP, at 20 Pa and 408 K, react to afford 2-oxazolidinones witha selectivity close to 100 %. In the absence of catalyst, the reac-tion does not take place under the same experimental condi-tions, indicating that what we have monitored is not a thermalreaction. Several catalysts were used without any pretreat-ment, each with a different ratio of acid to basic sites (Table 1).

The most active catalyst was g-ZrP in its form with strongacid sites.[14] The reaction was followed step by step and moni-tored by using multinuclear NMR spectroscopy. At 408 K, com-pounds 2 and 6 were identified, isolated, and fully character-ized by using spectroscopic andchromatographic techniques. Inour first attempts, using 10 % w/w of catalyst with respect toglycerol carbonate after 15 h(Scheme 4), a conversion of 14 %(subsequently improved to 40 %)of the starting carbonate intothe target product was achievedwith 100 % selectivity.

The slow step of the processunder the conditions reportedabove was the reaction of glyc-erol carbonate 1 with urea. Thelatter can attack glycerol carbon-ate on either side of the carbon-yl moiety (Scheme 4), affording 2(attack on the right side) or 2’

(attack on the left side). The two isomers can exist in twoforms: acyclic (2 o, 2 o’) or cyclic (2 c, 2 c’; Scheme 5).

The thermochemistry of the overall reaction, as well as ofthe assumed steps, has been examined computationally (fordetails, see the Experimental Section). The Gibbs free energiesobtained under standard and experimental conditions arelisted in Table 2. The data indicate that the formation of prod-uct isomers 6 and 6’ is highly exergonic and, of the three reac-tion steps (Scheme 4), the addition of urea to 1 is clearly ther-modynamically disfavored.

Scheme 3. Route for the synthesis of 2-oxazolidinone from glycerol.

Table 1. Activity of different catalysts in the synthesis of 2-oxazolidinones(decreasing order of acidity).[a]

Catalyst Acidity/basicity Yield of 6+6’[%]

g-ZrP 24.6 14ZrO2 9.0 6Nb2O5 6.8 9CeO2/Nb(20 %) 6.0 6ZnO 1.0 9CaO 0.0 4

[a] Reaction conditions: T = 408 K, t = 15 h, P = 20 Pa, glycerol carbonateand urea were used with equimolar ratio, catalyst 10 % w/w with respectto glycerol carbonate.

Scheme 4. Pathways for the synthesis of 2-oxazolidinones from glycerol carbonate and urea at 408 K.

Table 2. Computed Gibbs free energies.[a]

Reaction DG1

[kcal mol�1]DG2

[kcal mol�1]

1 + urea!2/2’ 1.3/�0.8 13.2/11.62/2’!5/5’+ CO2 �15.8/�15.9 �27.9/�293/3’!6/6’+ NH3 �4.8/�4.4 �18.5/�17.81 + urea!6/6’+ CO2 + NH3 �19.4/�21.2 �33.2/�35.1

[a] DG1 and DG2 refer to standard (T = 298 K, P = 1.01325 � 105 Pa) and ex-perimental (T = 408 K, P = 20 Pa) conditions. Data obtained for the twoisomeric forms of products and intermediates are given. See Scheme 5for product labeling.

� 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2013, 6, 345 – 352 346

CHEMSUSCHEMFULL PAPERS www.chemsuschem.org

Calculations showed that the cyclic form of species 2 and 2’(Scheme 5) were stabilized by 13 kcal mol�1 (1 kcal = 4.2 kJ)with respect to the corresponding open isomers. Most of thestabilization arises from the internal NH···O=C hydrogen bond,giving rise to six-membered ring structures, but OH···O=Cbond formation contributes notably to stabilization as well.The formation of the double hydrogen-bonded structures islikely to facilitate the decarboxy-lation process to afford species 5and 5’.

The formation of 2 and 2’(Scheme 4) was verified by 13Cand 1H NMR spectroscopy meas-urements. Figure 1 shows the13C NMR spectra of the com-pounds formed in the reactionof 1 with urea. Starting from Fig-ure 1 a (glycerol carbonate +

urea), after heating at 398 K for2 h, new signals appeared in ad-dition to those of the startingcompounds. Signals betweend= 157 and 158 ppm are attrib-uted to 2 and 2’, whereas signalsbetween d= 158 and 159 ppmare due to 5 and 5’. By heatingthe same mixture at T>408 K,such signals disappear and theunique signal of 2-oxazolidinoneappears at d= 159.5 ppm (Fig-ure 1 c). Indeed, we have shownthat two isomers 6 and 6’ areformed with a molar ratio equalto 1:1.

As an attempt to increase the reaction rate, the reactiontemperature was increased to 453 K. The conversion of glycer-ol carbonate significantly increased, affording 2-oxazolidinonesin 21 % yield after 3 h of reaction, but at the same time the se-lectivity decreased and new molecular products were formed.In addition, a rubbery mixture was also formed (most probablya polymeric material not yet well characterized) that was inac-tive and did not afford the target product. We have investigat-ed why the selectivity was lost and discovered that at 453 Ka completely different reaction mechanism operated becauseat such a temperature urea quickly decomposes into HNCOand NH3 (Scheme 6 a). Thus, at 453 K, ammonia formed upon

Scheme 5. Cyclization of opened species 2 o/2’o into species 2 c/2’c(o = open, c = cyclic).

Figure 1. 13C NMR spectra of the reaction mixture in three different phasesof the reaction: a) starting mixture, b) mixture after 2 h at 398 K, and c) mix-ture after 2 h at 418 K.

Scheme 6. Synthesis of 2-oxazolidinone from glycerol carbonate and urea at 453 K.

� 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2013, 6, 345 – 352 347

CHEMSUSCHEMFULL PAPERS www.chemsuschem.org

urea decomposition very rapidly reacts with glycerol carbonateto afford glycerol carbamates 3 and 3’ (Scheme 6 b).

Under these reaction conditions, glycerol carbamates 3 and3’ are unstable and can decompose in three different ways(Scheme 7): i) loss of isocyanic acid to afford glycerol ; ii) loss ofammonia to afford glycerol carbonate 1; and iii) loss of carbondioxide to form aminopropane diol (4/4’).

At 453 K, all such reactions may occur in thermally. In thepresence of uncalcined g-ZrP, glycerol carbamates 3 and 3’tend to lose carbon dioxide to afford 4 and 4’ (Scheme 6 c),which were characterized in situ. They were not isolated be-cause of their short lifetime. In fact, under these reaction con-ditions, they rapidly react with isocyanic acid to form 2,3-dihy-droxydiisopropylurea (5 and 5’; Scheme 6 d).

To support such a reaction pathway, 6-amino-1-hexanol (7),which presents the same structural features as 4 and 4’, wasreacted under a nitrogen flow with urea at 453 K; a tempera-ture at which urea affords ammonia and isocyanic acid. Ammo-nia was removed from the reaction medium and HNCO was al-lowed to react with 7, which has both the �OH and �NH2

functionalities like 4 ; this reaction demonstrated that an aminoalcohol and isocyanic acid reacted to afford a urea derivative(Scheme 8). The conversion of 7 was 89 %. Such a reactionmatches existing reports in the literature and confirms that thereaction in Scheme 8 is a method for the synthesis of substitut-

ed ureas.[16] Compound 8 was fully characterized. Differentfrom 5, which easily loses NH3 to afford cyclic species 6, com-pound 8 is not able to undergo cyclization due to the longerspacer between the�OH and�NH2 moieties.

In the last step of the synthesis reported in Scheme 6, com-pounds 5 and 5’ lose ammonia to afford 2-oxazolidinones 6and 6’. The reaction was carried out under vacuum. If ammo-nia was not removed, compound 5 reverted back to 4. Wehave ascertained that glycerol carbonate can be easily reactedwith ammonia with 100 % conversion into 3/3’ (Scheme 6 b) atroom temperature. In principle, NH3 can attack the carbonylfunctionality of the carbonate on both sides to afford carba-mates 3 and 3’.

We have also found that when glycerol carbonate is reactedwith NH3, by changing the reaction temperature, it is possibleto modulate the ratio of the two carbamates 3 and 3’. In fact,isomer 3’ is preferentially formed with respect to 3 at tempera-tures above 423 K: the selectivity towards 3’ tends to increasewith temperature. This is clearly demonstrated by analyzingthe reaction mixture by HPLC, which is able to separate theisomers. Figure 2 shows chromatograms of the mixture ob-

tained at 323 (a) and 423 K (b). However, working at a tempera-ture lower than 430 K, degradation of urea into ammonia andisocyanic acid was avoided, so that a high selectivity wasreached, but with a low conversion rate (14 %).

When increasing the temperature to 453 K, conversion ofthe carbonate increased significantly to 40–50 % and the yieldof oxazolidinone grew to 21 % with respect to 14 % at 423 K,but the selectivity for the 2-oxazolidinones decreased (from

100 to 40 % or less) due to theformation of several side prod-ucts, while the molar ratio of 6to 6’ remained 1:1.

Starting from glycerol, 2-oxa-zolidinones 6 and 6’ (Scheme 9)were obtained in the same yieldas that obtained when usingglycerol carbonate at high tem-perature (21 %), but with a molar

Scheme 7. Thermal degradation of glycerol carbamate.

Figure 2. HPLC signals of the two carbamates formed from glycerol carbon-ate and ammonia at a) 323 and b) 423 K.

Scheme 8. Reaction of 6-amino-1-hexanol (7) with isocyanic acid at 453 K.

� 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2013, 6, 345 – 352 348

CHEMSUSCHEMFULL PAPERS www.chemsuschem.org

ratio of 6’/6 = 7:1 The reaction pathway, investigated by usingFTIR and NMR spectroscopy, is shown in Scheme 9.

Under such conditions, glycerol reacts with one molecule ofisocyanic acid to afford 1-glycerol carbamate (3’), which contin-ues to react, following the route reported in Scheme 6, toafford 6’. In this case, the synthesis requires two molecules ofurea per molecule of glycerol. The good selectivity towardsspecies 6’ is due to the poor reactivity of the secondary �OHmoiety of glycerol in the first step of the reaction with respectto the primary�OH.

Trend for the conversion of glycerol or glycerol carbonateinto 2-oxazolidinones at 408 and 453 K

To highlight the different reactivity of glycerol and glycerol car-bonate with urea, the reaction was carried out at 408 and453 K and monitored at different intervals of time. The resultsobtained for the reaction at 453 K are shown in Figures 3 and4. The same reaction was monitored when starting from glyc-erol carbonate at 408 K (Figure 5).

In conclusion, when using glycerol carbonate at 408 K, theyield of 2-oxazolidinones (14 % by GC, reaction mixture) islower than that obtained when using glycerol carbonate orglycerol at 453 K (21 % by GC, reaction mixture). Nevertheless,

in the last case, the formation ofother products makes extractionof the oxazolidinones more diffi-cult and quite expensive. Asa matter of fact, extraction ofthe target product from the re-action mixture was close to100 % with glycerol carbonate at408 K and lower than 50–60 %with glycerol carbonate or glyc-erol at 453 K, resulting ina lower overall isolated yield(12 % with respect to 14 %).

Catalyst deactivation and improving the yield of 2-oxazolidi-nones

As reported in Figure 5, after 14 h of reaction at 408 K, the con-version of glycerol carbonate into 2-oxazolidinones reacheda maximum (14 %) that remained constant for longer reactiontimes. This fact pushed us to investigate whether the catalystwas deactivated during the reaction and what could causesuch deactivation. We investigated, in particular, if its acid–base properties were modified. We discovered that the acidityof the catalysts was strongly reduced during the catalytic run,most probably because of the presence of bases such as urea

Scheme 9. Synthetic route for the synthesis of 2-oxazolidinone 6’ from glycerol.

Figure 3. Influence of the reaction time on the synthesis of 2-oxazolidinonesfrom glycerol and urea at 453 K.

Figure 4. Influence of the reaction time on the synthesis of 2-oxazolidinonesfrom glycerol carbonate and urea at 453 K.

Figure 5. Influence of the reaction time on the conversion of glycerol car-bonate and urea into 2-oxazolidinones at 408 K.

� 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2013, 6, 345 – 352 349

CHEMSUSCHEMFULL PAPERS www.chemsuschem.org

or ammonia. The catalyst was recovered, washed with metha-nol, and after heating at 440 K showed the same catalytic ac-tivity as the starting material.

To improve the conversion yield and keep the selectivity ashigh as possible, we worked at 408 K and by adding the cata-lyst in two successive portions (0, 16 h) reached a total of 20 %with respect to glycerol carbonate.

The conversion yield increased from 14 to 18 % (isolated).Figure 6 shows the effect of the addition of fresh catalyst after

16 h: the production of oxazolidinone started again. When thereaction was carried out by loading the catalyst at 20 % w/wrate at the zero time point under the same reaction conditionsreported in Figure 5, 25 % oxazolidinone (isolated) was ob-tained at 408 K. At a catalyst loading of 25 %, the conversionwas close to 40 % with a selectivity of 100 %. These figures arequite interesting for potential applications. The catalyst wasquantitatively recovered and, after thermal treatment, reused:it showed the same activity.

Isolation of oxazolidinones

As reported in the Experimental Section, isolation of oxazolidi-none from the reaction mixture was made by column chroma-tography on silica using ethyl acetate/hexane (1:1). Whenusing a long column under flash chromatography conditions,the product isolated (mixture of 6/6’ as reported above) car-ried an amount of glycerol (either formed in the reaction orused as a reagent) in the range 8–11 %. The elimination ofsuch an amount of glycerol is impossible because it makesstrong interactions with the oxazolidinone molecules. DFT cal-culations have shown that the two isomers 6 and 6’ have dif-ferent interaction energies (Figure 7) with glycerol, but in anycase the energy is higher than 20 kcal mol�1, which justifies thedifficulty of glycerol removal.

This makes the isolation of pure oxazolidinone time consum-ing. We are working on implementing new techniques thatmay make this last step simpler.

Conclusions

The catalytic synthesis of oxazolidinones from glycerol or glyc-erol carbonate and urea is a clean and sustainable route tothese heterocycles, avoiding the use of toxic phosgene ormore expensive amino alcohols. The conversion yield and se-lectivity depend on the temperature and the amount of cata-lyst used. The heterogeneous catalyst used in this work, g-ZrP,was completely recovered and easily reactivated and recycled.Because both glycerol and glycerol carbonate are of biologicalorigin (glycerol is obtained from triglycerides and glycerol car-bonate is derived from it by reaction with urea in the Eurobior-ef Project), this route is an interesting alternative route to thephosgenation reaction for producing oxazolidinones and asa possible new utilization of bio-glycerol.

Experimental Section

General

All catalysts used were commercially available, except g-ZrP, whichwas prepared as reported in Ref. [14], and CeO2/Nb, which wasprepared as reported in Ref. [17].Note that for freshly prepared g-ZrP, the ratio nA/nB may dependon the ambient conditions under which samples are dried. It mayvary from 9 to 25, depending on the residual weak acid sites. Oncecalcined, all samples present the same nA/nB ratio: at 773 K it is4.15. Glycerol carbonate was synthesized as reported in Ref. [14]and purified by using a chromatographic column filled with silicagel and hexane/ethyl acetate = 1:1 as the eluent. Bio-glycerol wasan Arkema product and urea was purchased from Sigma Aldrich(RP).The IR spectra were obtained by using a spectrometer Shimadzu IRPrestige 21 instrument by placing the sample between KBr disks ifliquid, or dispersed in Nujol if solid. The reaction mixture was ana-lyzed by using a Thermo Focus GC gas chromatograph equippedwith a Varian Select Biodiesel for FAME 30 m � 0.32 mm capillarycolumn and a flame ionization detector. The GC–MS analyses wereconducted with a GC–MS Shimadzu QP5050 instrument equippedwith the same column as that used for the gas chromatograph.The 1H and 13C NMR spectra were recorded by using a Varian400 MHz or a Bruker 600 MHz spectrometer, as specified.

Reaction of glycerol carbonate and urea to afford 2-oxazoli-dinones

Method a: Glycerol carbonate (2.5 g, 21.4 mmol), powdered urea(1.28 g, 21.3 mmol), and catalyst (0.25 g, 10 % w/w with respect to

Figure 6. Effect of the addition of fresh catalyst to the reaction mixture atthe “zero growth” point on the yield of oxazolidinone.

Figure 7. The most stable forms of hydrogen-bonded complexes between 6or 6’ and glycerol. Calculated binding energies are given in kcal mol�1.

� 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2013, 6, 345 – 352 350

CHEMSUSCHEMFULL PAPERS www.chemsuschem.org

glycerol carbonate) were placed in a reactor connected toa vacuum system for ammonia and carbon dioxide removal(20 Pa). The mixture was heated at 408 K under stirring for 15 h. Atthe end of the reaction, the oil obtained was dissolved in the mini-mum amount of methanol and purified by column chromatogra-phy (silica gel 0.06–0.2 mm) with n-hexane/ethyl acetate= 1:1 v/vas the eluent. After removing the solvent under vacuum, a 1:1 mix-ture of 6 and 6’ was obtained that still contained 8–11 % glycerol.5-Hydroxymethyloxazolidin-2-one: 13C NMR (100 MHz, CDCl3,Me4Si): d= 41.5 (�CH2�), 62.4 (�CH2OH), 76.4 (�CH�), 159.5 ppm(�CO�) ; 1H NMR (600 MHz, CDCl3, Me4Si): d= 4.68 (�CH�), 3.72(�CH2O�), 3.59 (�CH2O�), 3.61 (�CH2�), 3.41 ppm (�CH2�).4-Hydroxymethyloxazolidin-2-one: 13C NMR (100 MHz, CDCl3,Me4Si): d= 53.2 (�CH�), 62.8 (�CH2OH), 66.4 (�CH2�), 159.5 ppm(�CO�) ; 1H NMR (600 MHz, CDCl3, Me4Si): d= 4.31 (�CH2�), 4.34(�CH2�), 4.27 (�CH�), 3.75 (�CH2O�), 3.71 ppm (�CH2O�) ; MS: m/z :31, 42, 44, 55, 58, 74, 99, 100, 116.

Method b: Glycerol carbonate (2.5 g, 21.4 mmol), powdered urea(1.28 g, 21.3 mmol), and catalyst (0.25 g, 10 % w/w with respect toglycerol carbonate) were placed in a reactor connected toa vacuum system for ammonia and carbon dioxide removal(20 Pa). The mixture was heated at 453 K under stirring for 3 h. Atthe end of the reaction, a highly viscous liquid was obtained. Itwas dissolved in the minimum amount of methanol and purifiedby column chromatography (silica gel 0.06–0.2 mm) with n-hexane/ethyl acetate = 1:1 v/v as the eluent. After removing sol-vent under vacuum, a 1:1 mixture of 6 and 6’ was obtained thatcontained 8–11 % glycerol.5-Hydroxymethyloxazolidin-2-one: 13C NMR (100 MHz, CDCl3,Me4Si): d= 41.5 (�CH2�), 62.4 (�CH2OH), 76.4 (�CH�), 159.5 ppm(�CO�) ; 1H NMR (600 MHz, CDCl3, Me4Si): d= 4.68 (�CH�), 3.72(�CH2O�), 3.59 (�CH2O�), 3.61 (�CH2�), 3.41 ppm (�CH2�).4-Hydroxymethyloxazolidin-2-one: 13C NMR (100 MHz, CDCl3,Me4Si): d= 53.2 (�CH�), 62.8 (�CH2OH), 66.4 (�CH2�), 159.5 ppm(�CO�) ; 1H NMR (600 MHz, CDCl3, Me4Si): d= 4.31 (�CH2�), 4.34(�CH2�), 4.27 (�CH�), 3.75 (�CH2O�), 3.71 ppm (�CH2O�) ; MS: m/z :31, 42, 44, 55, 58, 74, 99, 100, 116.

Method c: Glycerol carbonate (2.5 g, 21.4 mmol) and powderedurea (1.28 g, 21.3 mmol) were placed in a reactor connected toa vacuum system for ammonia and carbon dioxide removal(20 Pa). The catalyst (0.50 g or 20 % w/w) was added in portions tothe reaction mixture after 0, 10, and 16 h and the reaction was al-lowed to continue for 32 h at 408 K. The oil obtained was isolatedand purified as reported for Method a and analyzed by using NMRspectroscopy and GC-MS.5-Hydroxymethyloxazolidin-2-one: Yield 25 %; 13C NMR (100 MHz,CDCl3, Me4Si): d= 41.5 (�CH2�), 62.4 (�CH2OH), 76.4 (�CH�),159.5 ppm (�CO�) ; 1H NMR (600 MHz, CDCl3, Me4Si): d= 4.68(�CH�), 3.72 (�CH2O�), 3.59 (�CH2O�), 3.61 (�CH2�), 3.41 ppm (�CH2�).4-Hydroxymethyloxazolidin-2-one: 13C NMR (100 MHz, CDCl3,Me4Si): d= 53.2 (�CH�), 62.8 (�CH2OH), 66.4 (�CH2�), 159.5 ppm(�CO�) ; 1H NMR (600 MHz, CDCl3, Me4Si): d= 4.31 (�CH2�), 4.34(�CH2�), 4.27 (�CH�), 3.75 (�CH2O�), 3.71 ppm (�CH2O�) ; MS: m/z :31, 42, 44, 55, 58, 74, 99, 100, 116.

Reaction of glycerol and urea to afford 2-oxazolidinones

Glycerol (2 g, 21.7 mmol), urea (2.61 g, 43.5 mmol), and catalyst(0.2 g, 10 % w/w with respect to glycerol) were placed in a round-bottomed flask connected to a vacuum system for removing am-monia and carbon dioxide formed. The reaction was carried out at

453 K and stirred for up to 4 h. At the end of the reaction, the oilformed was cooled to room temperature, dissolved in a minimumamount of methanol, and purified as reported for Method a. A 1:7mixture of 6 and 6’ was obtained that contained 8–11 % glycerol.5-Hydroxymethyloxazolidin-2-one: Yield 15 %; 13C NMR (100 MHz,CDCl3, Me4Si): d= 41.5 (�CH2�), 62.4 (�CH2OH), 76.4 (�CH�),159.5 ppm (�CO�) ; 1H NMR (600 MHz, CDCl3, Me4Si): d= 4.68(�CH�), 3.72 (�CH2O�), 3.59 (�CH2O�), 3.61 (�CH2�), 3.41 ppm(�CH2�).4-Hydroxymethyloxazolidin-2-one: 13C NMR (100 MHz, CDCl3,Me4Si): d= 53.2 (�CH�), 62.8 (�CH2OH), 66.4 (�CH2�), 159.5 ppm(�CO�) ; 1H NMR (600 MHz, CDCl3, Me4Si): d= 4.31 (�CH2�), 4.34(�CH2�), 4.27 (�CH�), 3.75 (�CH2O�), 3.71 ppm (�CH2O�) ; MS: m/z :31, 42, 44, 55, 58, 74, 99, 100, 116.

Reaction of 6-amino-1-hexanol with urea to afford N-hydrox-yhexyl urea

6-Amino-1-butanol (2 g, 22.5 mmol) and urea (1.35 g, 22.5 mmol)were placed into a 10 mL Sovirel vessel. The reaction was carriedout under nitrogen flow to remove ammonia at 453 K for 3 h. Aftercooling to room temperature, the mixture (a highly viscous liquid)was dissolved in the minimum amount of methanol and purifiedchromatographically (eluent: hexane/ethyl acetate 1:1). After re-moving the solvent, a pale-yellow liquid was obtained and ana-lyzed by GC-MS and NMR spectroscopy. The conversion of aminoalcohol was 89 % with 81 % selectivity for N-hydroxyhexyl urea.13C NMR (100 MHz, CDCl3, Me4Si): d= 25.1(�CH2�), 26.1 (�CH2�),29.3 (�CH2�), 31.6 (�CH2�), 40.3 (�CH2�), 62.2 (�CH2�), 161.2 ppm(�CO�) ; MS: m/z : 44, 61, 73, 88, 116, 129, 142.

Reaction of glycerol carbonate with ammonia to afford glyc-erol carbamates

Glycerol carbonate (1 g, 8.5 mmol) was placed in a stainless-steelautoclave loaded with 5 bar of ammonia (10.2 mmol). The reactionwas carried out at 323 and 343 K. At the end of the reaction, thegas phase was removed and the liquid phase containing the twocarbamates was analyzed by HPLC (0.005 m H2SO4 as the mobilephase) and 13C NMR spectroscopy.1-Glyceryl carbamate: 13C NMR (100 MHz, CD3CN, Me4Si): d= 63.3(�CH2�), 66.4 (�CH2OH), 70.9 (�CHOH), 157.1 ppm (�CO�).2-Glyceryl carbamate: 13C NMR (100 MHz, CD3CN, Me4Si): d= 63.8(�CH2O�), 73.1 (�CH�), 157.1 ppm (�CO�).

Acknowledgements

The research leading to these results has received funding fromthe European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 241718 EuroBioRef.

Keywords: glycerol · homogeneous catalysis · hydrogenbonds · oxazolidinones · zirconium

[1] D. J. Diekema, R. N. Jones, Lancet 2001, 358, 1975 – 1982.[2] H.-L. Elbe, S. Bohm, D. Berg, H.-W. Dehne, S. Dutzmann, US Patent

5278179, 1992.[3] L. M. Ednie, M. R. Jacobs, P. C. Appelbaum, J. Antimicrob. Chemother.

2002, 50, 101 – 105.[4] D. H. Crater, R. D. Howells, R. M. Stern, J. A. Temperante, US Patent

5025052, 1991.

� 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2013, 6, 345 – 352 351

CHEMSUSCHEMFULL PAPERS www.chemsuschem.org

[5] R. Funk, V. Frenz, U. Riegel, M. Weismantel, F. Engelhardt, T. Daniel, USPatent 6503979, 2000.

[6] D. J. Ager, I. Prakash, D. R. Schaad, Chem. Rev. 1996, 96, 835 – 876.[7] a) J. R. Gage, D. A. Evans, Org. Synth. 1990, 68, 77 – 82; b) J. R. Gage,

D. A. Evans, Org. Synth. 1990, 68, 83 – 91.[8] B. Gabriele, R. Mancuso, G. Salerno, M. Costa, J. Org. Chem. 2003, 68,

601 – 604.[9] M. Feroci, A. Gennaro, A. Inesi, M. Orsini, P. Palombi, Tetrahedron Lett.

2002, 43, 5863 – 5865.[10] M. R. Banks, J. I. G. Cadgan, I. Gosney, P. K. G. Hodgson, D. E. Thomas, J.

Chem. Soc. Perkin Trans. 1 1991, 961 – 962.[11] L. F. Xiao, L. W. Xu, C. G. Xia, Green Chem. 2007, 9, 369 – 372.[12] M. Tu, R. J. Davis, J. Catal. 2001, 199, 85 – 91.

[13] S. R. Jagtap, Y. P. Patil, S.-I. Fujita, M. Arai, B. M. Bhanage, Appl. Catal. A2008, 341, 133 – 138.

[14] M. Aresta, A. Dibenedetto, F. Nocito, C. Ferragina, J. Catal. 2009, 268,106 – 114.

[15] S. Yamanaka, M. Tanaka, J. Inorg. Nucl. Chem 1979, 41, 45 – 48.[16] V. P. Kozyukov, E. V. Muzovskaya, V. F. Mironov, Zh. Obshch. Khim. 1980,

50, 1425.[17] A. Dibenedetto, M. Aresta, A. Angelini, J. Ethiraj, B. M. Aresta, Chem. Eur.

J. 2012, 18, 10324 – 10334.

Received: July 24, 2012Revised: September 5, 2012Published online on December 7, 2012

� 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2013, 6, 345 – 352 352

CHEMSUSCHEMFULL PAPERS www.chemsuschem.org