Applied Catalysis A: General 392 (2011) 19–27 Contents lists available at ScienceDirect Applied Catalysis A: General journa l homepage: www.e lsev ier .com Compa l to comme an Arturo Romeroa,∗, Aurora Santosa, Daniel Escrigb, Ernesto Sim a Departamento de Ingeniería Química, Facultad de Químicas, Universidad Complutense de Madrid, Ciudad b UBE Corporation Europa S.A., Polígono Industrial El Serrallo s/n, 12100 Castellón, Spain a r t i c l e i n f o Article history: Received 27 July 2010 Received in revised form 4 October 2010 Accepted 18 October 2010 Available online 12 November 2010 Keywords: Alumina Catalyst Chromium Copper Cyclohexanol Cyclohexene Cyclohexanone Dehydrogenation Phenol Zinc a b s t r a c t Catalytic dehydrogenation of cyclohexanol tinuous fixed bed reactor under atmospher have been checked. Effect of temperature (i studied. The catalytic activity has been eva secondary reactions of dehydration and de quantified by GC/MS. Catalysts have been characterized byX-ra nia and BET surface area measurement. H the operating conditions, concerning the that catalysts with alumina and chromium main impurity obtained. For a given cyclohexanone yield the impurities from dehydrogenation reac- tions showed similar trends for the three catalysts tested. Phenol was the main impurity obtained by dehydrogenation. © 2010 Elsevier B.V. All rights reserved. Nomenc dp �p �L SBET Vp WHSV YONE Yj YH2O YH2 YH2 imp SH2 imp 1. Introdu Catalytic anone is an ∗ Correspon E-mail add 0926-860X/$ – doi:10.1016/j. lature particle diameter (mm) particle density (g cm−3) bed density (g cm−3) BET surface area (m2 g−1) pore volume (cm3 g−1) weight hourly space velocity (h−1) cyclohexanone percentage yield (%) impurity j percentage yield (%) water percentage yield (%) total yield of hydrogen (%) hydrogen yield from impurities (%) selectivity to hydrogen from impurities ction dehydrogenationof cyclohexanol toproducecyclohex- important industrial process especially in producing ding author. Tel.: +34 913944171; fax: +34 913944171. ress:
[email protected] (A. Romero). �-caprolactam, main raw material in the manufacture of nylon- 6. As a polyamide fiber raw material must increasingly fulfill meet higher purity requirements [1]. The impurities can come from the products formed in the transformation stages of the reagents as dehydrogenation of cyclohexanol, which is a critical process, where is necessary to minimize the impurities that affect seriously the later stages. From an industrial point of view, the heterogeneous catalytic gas-phase dehydrogenation at atmo- spheric pressure is severely restricted by highly endothermic reaction (�H=65kJ/mol) and thermodynamic equilibrium [2], and also includes a complex consequent-parallel reactions, where cyclohexanone selectivity decreases because of an increase of the impurities yields [3,4]. There are two methods for dehydrogenation of cyclohexanol, at low temperature, from 200 to 300 ◦C, and at high temperature, from 350 to 450 ◦C. Copper oxide based catalyst is usually used at low temperature [4–18]. Metals such as Zn, Cr, Fe, Ni, alkali met- als, alkaline earth metals, and thermally stable metal oxides (Al, Si, and Ti) are added. Chromia acts as a structural promoter because it increases the BET surface area and also inhibits the sintering of copper particles [19]. Zinc calciumoxidehas beenused at high tem- perature. Copper catalysts arenotusedathigh temperature toavoid sintering of the copper [20]. In the recent years, more attention has been paid in literature to these low temperature catalysts compared to high temperature. see front matter © 2010 Elsevier B.V. All rights reserved. apcata.2010.10.036 rative dehydrogenation of cyclohexano rcial copper catalysts: Catalytic activity / locate /apcata cyclohexanone with d impurities formed óna Universitaria, 28040 Madrid, Spain to cyclohexanone has been carried out on phase gas in a con- ic pressure. Copper chromite and copper zinc oxide catalysts n the range 250–290 ◦C) and spatial time in reactor have been luated in terms of cyclohexanone yields and impurities from hydrogenation of cyclohexanol have also been identified and y diffraction, temperature programmeddesorption of ammo- igh activity was confirmed by copper-based catalysts under size and dispersion of the copper specie. It was also found exhibit higher dehydration capacity, being cyclohexene the 20 A. Romero et al. / Applied Catalysis A: General 392 (2011) 19–27 ol deh Cesar et al and a com tested a Cu [14] examin La2O3–Cr2O Cu–ZnO/SiO lyst, Cu–Cr2 catalyst. Many re theprepara selectivity t lysts, to inc In these wo yield data a few results be found. C from cycloh identified b The pres of copper c genation o times in re cyclohexan drogenation quantified b were also d 2. Experim 2.1. Chemic Cyclohex 29135), be 0), phenol 29255), 2- and 1,4-ben or standard Copper chr Cu-0203 (t copper zinc vided by Sü talyt alytic ase w xed-b l dia cat pher tme en a expe tem f the re re one ratur cyclo The ect t h hig .1 to Fig. 1. Experimental setup for cyclohexan . [9] used a bimetallic catalyst adding Co to Cu/SiO2 mercial Cu/SiO2 catalyst. Fridman and Davydov [10] /Mg, Cu/Zn and Cu/Zn/Al catalysts. Siva Kumar et al. ed a Cu/ZnO based catalysts promoted with Cr2O3 and 3 as double promoter. Ji et al. [15] used a Cu/SiO2 and 2 catalysts.Nagaraja et al. [16,17] testedaCu/MgOcata- O3/MgOpromoted catalyst anda commercial Cu-1800P searchers have analyzed the influence of the support, tionmethodandcopper loadingonboth theactivity and o cyclohexanone of different copper-containing cata- rease the conversion of cyclohexanol to cyclohexanone. rks, both cyclohexanol conversion and cyclohexanone re well documented. However, to our best knowledge, about the impurities and their corresponding yields can yclohexene from cyclohexanol dehydration and phenol exanol dehydrogenation are in general the impurities y authors [4,7,15]. ent study was carried out to evaluate the performance hromite and copper zinc oxide catalysts for dehydro- f cyclohexanol at different temperatures and spatial 2.2. Ca Cat gas ph flow fi interna of each glass s pretrea hydrog of the The start o peratu flushed tempe 5wt.% 22 ◦C). not aff throug from 0 actor. The catalytic activity was studied in terms of one yield. Moreover, the main dehydration and dehy- impurities from cyclohexanol were identified and y GC/MS. Results obtained for activity and selectivity iscuss attending the catalyst properties. ental als and catalysts anol (Sigma–Aldrich, 105899), cyclohexanone (Fluka, nzene (Fluka, 12540), cyclohexene (Aldrich, 24,099- (Riedel-de Haën, 33517), 2-cyclohexen-1-one (Fluka, cyclohexylidene-cyclohexanone (Alfa-Aesar, L09798) zodioxan (Aldrich 179000) have been used as reactants s. Three commercial catalysts have been employed. omite catalysts, Cu-1230 (crushed, 1.7× 4.7mm) and ablets, 3.1×3mm), were supplied by Engelhard, and oxide catalyst, T-2130 (tablets, 3×3mm), was pro- d-Chemie. effluent fro phasewere by chroma ations cond steady stat this period analyzed an tions.Weh transport re velocity. M 2.3. Catalys BET surf adsorption Beckman C sample was XRD pat using mono 45kV and 0.04◦ with ydrogenation. ic activity dehydrogenation of cyclohexanol to cyclohexanone on as carried out at atmospheric pressure in a continuous ed reactor made of a stainless-steel tube with 0.85 cm meter and 25 cm length. The bed was filled with 10g alyst. The bed volume was completed with nonporous es, inert glass wool and stainless-steel wire mesh. As nt the catalysts were reduced with 95% nitrogen 5% t 180 ◦C for 18h (GHSV=1100h−1). A detailed scheme rimental setup is given in Fig. 1. perature reactions were 250 and 290 ◦C. Before the reaction the catalyst was stabilized with N2 at tem- action. Once the reaction is finished, the catalyst was hour in N2 flow and later was also cooled at room e in nitrogen atmosphere. Cyclohexanol was fed with hexanone to avoid the cyclohexanol solidification (mp addition of cyclohexanone in the raw material does he results of impurities obtained. Feed was pumped h precision pump. The liquid flow rate was changed 1mLmin−1 (WHSV from 0.43 to 5.80h−1). The vapour m the reactor was cooled at 20 ◦C and liquid and gas separated and collected, and liquid phasewas analyzed tography. The catalysts were run for 6h under oper- itions for each experiment. After 2h of reaction, the e was achieved. No changes in catalyst activity during of time were observed. The steady state samples were d these values were used for all subsequent calcula- ave confirmed the absence of external and internalmass sistances by changing particle diameter and superficial oreover, pressure drop in fixed bed was negligible. ts characterization ace area and pore volume were determined using N2 method at liquid nitrogen temperature (77K) on a oulter SA3100 Analyzer. Before each measurement, the degassed at 563K for 60min. terns were recorded on a Philips X’Pert diffractometer, chromated Cu K� radiation (�=1.5418 A˚), operating at 40mA. The measurements were recorded in steps of a count time of 1 seg. in the 2 Theta range of 5–90◦. A. Romero et al. / Applied Catalysis A: General 392 (2011) 19–27 21 The acidity was determined by NH3-TPD. Before the adsorption of ammonia, the sampleswere treatedunderheliumat500 ◦C (from 25 to 500 ◦C in 20min.) for 1h. The samples were then cooled at 100 ◦C in He flow, and then treated with a NH3 flow for 5min at 100 ◦C. The for 1h at 30 10 ◦C/min a from Shima 2.4. Analyti Cyclohex (HP 6890 G tion were a both analy 30m×0.25 was used as Standard from their available a 2-cyclohexy were assi cyclohexan 3. Results 3.1. Catalys Table 1 reduced cat BET surface In C2 and C chromite an The NH3 curves for r these curve temperatur can be class strong (350 is reported erate acid s acid sites sh C2 and C3 c The XRD C2 and C3 a mated from reduced cat C1 cataly lysts show t crystalline p correspond CuO is in th and Cu0 spe appear. Som high (44wt in XRD are containing copper spec [14,21]. In t distinguish C2 cataly atmosphere of Cu0 appe distinguish observed th Ta b le 1 Ph ys ic o- ch em ic al p ro p er ti es of re d u ce d ca ta ly st s. C om m er ci al ca ta ly st C om p os it io n , w t. % C u ,w t. % Ph ys ic al an d ch em ic al p ro p er ti es d p m m � p g cm −3 � L g cm −3 S B ET ,m 2 g− 1 V P ,c m 3 g− 1 A ci d it y, � m ol N H 3 g− 1 Ph as es X R D C u cr ys ta ll it e si ze ,n m W ea k M od er at e St ro n g To ta l R ed u ce d R ed u ce d C 1 C u -1 23 0 En ge lh ar d C u O ,1 5 C u C r 2 O 4 ,4 4 B aC rO 4 ,1 2 A l 2 O 3 ,2 9 24 1. 6 1. 50 1. 17 12 1. 8 0. 30 0 38 20 6 74 31 8 C u C u 2 O 15 .5 8. 8* C 2 C u -0 20 3 En ge lh ar d C u O ,7 2 C u C r 2 O 4 ,2 6 C gr ap h it e, 2 65 3. 17 3. 33 2. 20 13 .5 0. 04 0 4 14 2 20 C u C u C r 2 O 4 C gr ap h it e 27 .2 C 3 T- 21 30 Sü d C h em ie C u O ,3 3 Zn O ,6 6 C gr ap h it e, 1 26 3 2. 30 1. 76 43 .4 0. 22 3 3 27 0 30 C u Zn O C gr ap h it e 9 * C or re sp on d in g to C u 2 O p h as e. physisorbed ammonia was eliminated by flowing He 0 ◦C. The NH3-TPD was run between 100 and 500 ◦C at nd followed by an on-line gas chromatograph, GC-15A dzu, provided with a thermal conductivity detector. cal methods anol and cyclohexanone were analyzed by GC/FID C-FID). Impurities of the cyclohexanol dehydrogena- nalyzed by GC/MS (HP 6890N GC MSD 5975B). For sis a HP-INNOWAX 19091N- 133 (crosslinked PEG) mm∅I ×0.25�m column were used. 1,4-benzodioxan ISTD for calibration. s used for quantitative analysis were calibrated commercial products. The impurities identified not s commercial products (2-cyclohexyl-cyclohexanone, lidene-cyclohexanol and 2-cyclohexyl-cyclohexanol) gned to the response of 2-cyclohexylidene- one. and discussion ts characterization summarizes the physico-chemical properties of the alysts. As can be observed, C1 catalyst presents major area due to the presence of alumina in its composition. 3 catalysts the BET surface area are provided by copper d zinc oxide, respectively [19]. -TPD profiles and the deconvolution of the NH3-TPD educed catalysts are shown in Fig. 2. The area under s is the total amount of NH3 desorbed over the range of e. Based on the desorption temperature, the acid sites ified as weak (150–250 ◦C), medium (250–350 ◦C) and –450 ◦C) [27]. The acidity depending on their strengths in Table 1. C1 catalyst presents highest amount of mod- ites and a significant amount of strong acid sites. These ow C1 catalyst as the most acidic, also due to alumina. atalysts do not exhibit acidity. patterns of the calcined and reduced catalysts of C1, re shown in Fig. 3. The copper crystallite size was esti- Debye–Scherrer equation from the XRD patterns of the alysts. These data are also summarizes in Table 1. st. The XRD results of the calcined and reduced cata- hat C1 is an amorphous catalyst where the intensity of hases is very low. The calcined catalyst presents peaks ing to chromate and chromite. In the calcined catalyst e amorphous phase. After reduction the peaks of Cu+ cies appear and the chromate and chromite ones dis- e authors have observed that when the Cr content is .% in the formof copper chromite for C1), copper species found as copper (I) oxide and metallic copper, and Cr species cannot be seen by XRD, which suggests that the ies are highly dispersed and exists in amorphous phase he reduced catalyst, Cu2O and Cu0 crystallites size are ed. The size of Cu2O is minor that Cu0. st. The XRD patterns indicate that after reduction in an of hydrogen, the peaks of CuO disappear and the peaks ar. In the calcined catalyst the CuCr2O4 specie is hardly ed and is maintained in reduced catalyst. Some authors at copper chromite is not reduced in an atmosphere 22 A. Romero et al. / Applied Catalysis A: General 392 (2011) 19–27 0,0 0,0 0,0 0,003 Real data Weak Moderate D e s o rb e d N H 3 ( m m o l/ g ·º C ) C1 reduced 0,0 0,0 0,0 0,0 D e s o rb e d N H 3 ( m m o l/ g ·º C ) 0,0 0,0 0,0 0,0 D e s o rb e d N H 3 ( m m o l/ g ·º C ) Fig. 2. NH3-TP of C1, C2 and C of hydrogen and Cu2O c Cu content dispersion. C3 cataly indicate tha oxide has n CuO specie dispersed in crystallite s order of ma Cu0 crystall than C2, ar catalyst. 3.2. Catalyt The mai hexanol to catalysts h impurities ( molecular s impurities a CXCXOL, CXECXONEandCXECXOL) anddehydrogenation reactions (CXENONE, PHOH and BZN). Benzene has been associated to both processes, as it may be obtained as the result of dehydration and further dehydrogenation of cyclohexanol. In order to confirm the ment e com incid nd th puri exyl- hex exyli e im NIST m th para puri e yie FON cata s cy y, W he c t bo espe 00 01 02 Strong Sum data 00 01 02 03 Real data Weak Moderate Strong Sum data C2 reduced 03 Real data Weak C3 reduced assign of thes The co dards a The im cycloh 2-cyclo cycloh to thes library Fro after se and im centag YONE = The Fig. 4 a velocit seen, t yield a 290 ◦C 500400300200100 00 01 02 Moderate Sum data Temperature (ºC) D patterns and deconvoluted NH3-TPD curves for reduced catalysts 3. [19,22]. Cu0 crystallite size of C2 is a higher than Cu0 rystallite sizes in C1. Minor BET surface area and major (65wt.%) suggests that C2 catalyst presents lower Cu st. The XRD data of the calcined and reduced catalyst t CuO is transformed into Cu0. The XRD profile of zinc ot been affected by reduction. The low intensity of the diffraction peaks may suggests that the copper is highly the zinc oxide phase [23,24]. For C3 catalyst, the Cu0 ize is smaller than that in C1 and C2 but in the same gnitude as Cu2O of the C1 catalyst. Presence of smaller ite size, major BET surface area and minor Cu content e a clear indication of higher copper dispersion in C3 ic activity n impurities generated in dehydrogenation of cyclo- cyclohexanone at 250 and 290 ◦C with C1, C2 and C3 ave been identified and quantified by GC/MS. These acronyms, formula, weight molecular, CAS number and tructure) are given in Table 2. It is noted that these re obtained from dehydration (CXEN, BZN, CXCXONA, major catal tallite size Cu–Zn cata It is also tion approa does not fu which star undesirable culated equ to cyclohex 250 ◦C and The perc of molar flo the reactor Yj = Fj FOLo × It was foun quite adequ rities obtain Results o genation re 3.3. Impuri Water is anol to cyc In this last 2-cyclohex cyclohexan are formed Molar fl reactor rati from dehyd YH2O = FH2 FOL +YC s of the identified compounds the MS library standards pounds have been analyzed and used for calibration. ence of retention time and spectrum between the stan- e runproducts allows validating the assignments done. ties identified not available as commercial products (2- cyclohexanone, 2-cyclohexylidene-cyclohexanol and yl-cyclohexanol) were assigned to the response of 2- dene-cyclohexanone. Spectrum of the peaks associated purities match spectrum of these compounds on the MS 5.0 with a quality higher than 95%. e composition of liquid samples at the collection tank, ting hydrogen, the molar flows of cyclohexanone, FONE, ties, Fj, are determined, as molh−1. Cyclohexanone per- ld, YONE, is defined as: E − FONEo FOLo × 100 (1) lytic activity for C1, C2 and C3 catalysts is presented in clohexanone percentage yield vs. weight hourly space HSV, at both temperatures 250 and 290 ◦C. As can be atalyst C3 presents higher cyclohexanone percentage th temperatures tested, being even better behavior at cially at low weight hourly space velocity (WHSV). The ytic activity of C3 could be ascribed to smaller Cu crys- and higher dispersion, as reported several authors for lysts [6,14,16,18]. observed at these low values of WHSV that as the reac- ches at equilibrium, cyclohexanone percentage yield rther increase, and even shows a maximum beyond ts down the cyclohexanone formed, suggesting that reactions from cyclohexanone are taking place. Cal- ilibrium conversion for cyclohexanol dehydrogenation anone, at the conditions employed, were about 70% at 80% at 290 ◦C. entage yield for an impurity j is calculated as the ratio w of impurity j to the molar flow of cyclohexanol fed to , and is given as percentage: 100, j /= ONE (2) d that the mass balance of cyclohexanol reacted fits ately with the cyclohexanone formed and the impu- ed. btained for impurities from dehydration and dehydro- actions are detailed below. ties from dehydration reactions directly formed by both dehydration from cyclohex- lohexene and condensation among six carbon cycles. way, 2-cyclohexylidene-cyclohexanone (CXECXONE), yl-cyclohexanone (CXCXONE), 2-cyclohexylidene- ol (CXECXOL) and2-cyclohexyl-cyclohexanol (CXCXOL) . ow of water to molar flow of cyclohexanol fed to the o, denominated molar water yield, YH2O, is calculated ration impurities by stoichiometry, as follows: O o × 100 ∼= YCXEN + YCXCXOL + YCXCXONE XECXOL + YCXECXONE + YBZN (3) A. Romero et al. / Applied Catalysis A: General 392 (2011) 19–27 23 80706050403020 0 200 400 600 800 1000 1200 1400 Cu O Cu H e ig h t (c ts ) C1 reduced Position (2ºTheta) 0 200 400 600 800 1000 1200 1400 BaCrO CuCr O H e ig h t (c ts ) C1 calcined C Graphite CuO CuCr O C2 calcined 80706050403020 C Graphite Cu CuCr O C2 reduced Position (2ºTheta) C Graphite CuO ZnO C3 calcined 80706050403020 C Graphite Cu ZnO C3 reduced Position (2ºTheta) Fig. 3. XRD patterns of calcined and reduced catalysts of C1, C2 and C3. In Fig. 5 water percentage yield vs. cyclohexanone percentage yield for each catalyst and temperature tested is shown. As can be seen the dehydration reactions are highly favored by C1 catalyst. These results for C1 are in accordance with the aforementioned NH3-TPD results. The dehydrating effects of alumina in the compo- 1 Y O N E ( % ) 1 Y O N E (% ) Fig. 4. Cyclohe temperature t sition of dehydrogenation catalysts have already been described in literature [4,7,25,26]. In C2 catalyst the amount of formed water is also remarkable. Nagaraja et al. [17] found that the largest amount of chrome (44% by weight in the form of copper chromite for C1 catalyst, and 26% for C2) in copper chromite catalysts induces dehydrating effects. From Fig. 5 it can be noticed the great influ- 1 1 ) 80 00 key Catalyst C1 C2 C3 a 0 20 40 60 0 20 40 60 80 00 6420 WSHV (h -1 ) key Catalyst C1 C2 C3 b xanone percentage yield as a function of WHSV for each catalyst and ested, (a) 290 ◦C and (b) 250 ◦C. Y H 2 0 ( % ) Y H 2 0 ( % Fig. 5. Water each catalyst a 8 0 2 key Catalyst C1 C2 C3 a 806040200 0 1 2 3 4 Y ONE (%) key Catalyst C1 C2 C3 b 0 2 4 6 percentage yield as function of cyclohexanone percentage yield for nd temperature tested, (a) 290 ◦C and (b) 250 ◦C. 24 A. Romero et al. / Applied Catalysis A: General 392 (2011) 19–27 Table 2 Detected and quantified impurities. Compound Acronym Formula MW CAS# Molecular structure Benzene 7 Cyclohexene 8 Phenol 9 2-Cyclohexe 9 Cyclohexano Cyclohexano 2-Cyclohexy 2-(1-Cycloh 2-Cyclohexy 2-Cyclohexy 2-(1-Cycloh 2-Cyclohexy ence of tem rises when remarkable In Figs. reaction at shown in F 2-cyclohexy 2-cyclohexy yield in Fig as molar pe As comp that the m BZN C6H6 CXEN C6H10 PhOH C6H6O n-1-one CXENONE C6H8O ne ONE C6H10O 9 l OL C6H12O 1 lidene-cyclohexanone CXECXONE C12H18O 1 exenyl) cyclohexanone l-cyclohexanone CXCXONE C12H20O 1 lidene-cyclohexanol CXECXOL C12H20O 1 exenyl) cyclohexanol l-cyclohexanol CXCXOL C12H22O 1 perature on the dehydration yield. Water production temperature increases. As expected, this effect is more for C1 and C2. 6–9 profiles of main impurities from dehydration 250 and 290 ◦C are shown. Cyclohexene yield is ig. 6, sum of 2-cyclohexylidene-cyclohexanone and l-cyclohexanone yield is given in Fig. 7, sum of lidene-cyclohexanol and 2-cyclohexyl-cyclohexanol . 8 and benzene yield in Fig. 9 yield values are given rcentages. ared with the results in Figs. 5 and 6 it is deduced ain impurity from dehydration reactions for C1 and C2 catalyst this impur sequently WHSV). Th ature, that dehydratio similar to cyclohexen temperatur Sum of C observed fo negligible a 8 71-43-2 2 110-83-8 4 108-95-2 6 930-68-7 8 108-94-1 00 108-93-0 78 1011-12-7 1502-22-3 80 90-42-6 80 100314-24-7 66500-79-6 82 6531-86-8 s is cyclohexene. A significant rate of formation of ity is obtained at low cyclohexanone yield and con- high cyclohexanol concentration (highest values of erefore, it is assumed, as it usually done in liter- cyclohexene is directly formed from cyclohexanol n. Influence of temperature on cyclohexene yield is the observed one for water yield. The amount of e produced in C3 catalyst is almost negligible at both es. XECXONE and CXCXONE shows a similar trend to that r cyclohexene. Rate of formation of both impurities is t low cyclohexanone yield (medium rich in cyclohex- A. Romero et al. / Applied Catalysis A: General 392 (2011) 19–27 25 0 2 4 6 8 10 a key Catalyst C1 C2 C3 Y C X E N (% ) 806040200 0 1 2 3 4 b key Catalyst C1 C2 C3 Y C X E N (% ) Y ONE (%) Fig. 6. Cyclohexene percentage yield as a function of cyclohexanone percentage yield for each catalyst and temperature tested, (a) 290 ◦C and (b) 250 ◦C. anol). Therefore, it is assumed that these are produced mainly from cyclohexanone. Consequently, the yield for this lumping specie is higher in C1 than in C2 while C3 shows the lower amount for the three cataly Y C X E C X O N E + C X C X O N A (% ) Y C X E C X O N E + C X C X O N A (% ) Fig. 7. Sum of percentage yie and temperatu 0,00 0,02 0,04 0,06 0,08 a key Catalyst C1 C2 C3 Y C X E C X O L + C X C X O L (% ) 806040200 0,00 0,02 0,04 0,06 0,08 b key Catalyst C1 C2 C3 Y C X E C X O L + C X C X O L (% ) Y ONE (%) Fig. 8. Sumof 2-cyclohexylidene-cyclohexanol and2-cyclohexyl-cyclohexanol per- centage yield as a function of cyclohexanone percentage yield for each catalyst and temperature tested, (a) 290 ◦C and (b) 250 ◦C. amount of these impurities as temperature rises, being this effect more remarkable in C1 and C2. The higher acidity due to the large amount of acid sites on alumina for C1 catalyst explains the highest y on sts considered. It is also noticed an increase on the 1,5 a key Catalyst activit 0,0 0,3 0,6 0,9 1,2 C1 C2 C3 806040200 0,0 0,1 0,2 0,3 0,4 0,5 0,6 b key Catalyst C1 C2 C3 Y ONE (%) 2-cyclohexylidene-cyclohexanone and 2-cyclohexyl-cyclohexanone ld as a function of cyclohexanone percentage yield for each catalyst re tested, (a) 290 ◦C and (b) 250 ◦C. 0 0 0 0 0 0 Y B Z N (% ) 0 0 0 0 0 Y B Z N (% ) Fig. 9. Benzen for each cataly dehydration reactions. ,00 ,04 ,08 ,12 ,16 ,20 a key Catalyst C1 C2 C3 806040200 ,00 ,02 ,04 ,06 ,08 b key Catalyst C1 C2 C3 Y ONE (%) e percentage yield as a function of cyclohexanone percentage yield st and temperature tested, (a) 290 ◦C and (b) 250 ◦C. 26 A. Romero et al. / Applied Catalysis A: General 392 (2011) 19–27 Y H 2 im p ( % ) 8 10 Y H 2 im p ( % ) key Catalyst C1 C2 C3 a Fig. 10. Hydro for each cataly The sum always mu CXCXONE b impurities cyclohexan CXECXONE sum of CXE cyclohexan A remar ene (Fig. 6) rich in cyclo Figs. 7 and 8 at equilibriu Benzene form cycloh agreement benzene pr other hand, orders of m produced. A rises. 3.4. Impuri By stoic the molar fl the reactor YH2 imp, wh also obtaine defined by YH2 = FH2 FOLo 0 1 2 3 a key Catalyst C1 C2 C3 Y P h O H (% ) 806040200 0,0 0,1 0,2 0,3 0,4 0,5 b key Catalyst C1 C2 C3 Y P h O H (% ) Y ONE (%) Phenol percentage yield as a function of cyclohexanone percentage yield catalyst and temperature tested, (a) 290 ◦C and (b) 250 ◦C. p = FH2 imp FOLo × 100 = 2 × YCXENONE + 3 × YPhOH + 2 × YBZN (5) valu ne y 806040200 0,0 0,2 0,4 0,6 0,8 1,0 1,2 Y ONE (%) key Catalyst C1 C2 C3 b 0 2 4 6 gen percentage yield from impurities as a function of cyclohexanone st and temperature tested, (a) 290 ◦C and (b) 250 ◦C. of CXECXOL and CXCXOL yield, shown in Fig. 8, is ch lower than that corresponding to CXECXONE and ut following a similar trend. Rate of formation of both is lower at low cyclohexanone yield (medium rich in ol) but the slope is higher than observed for the sum of and CXCXONE. Therefore, it is assumed that the lumped Fig. 11. for each YH2 im The hexano CXOL and CXCXOL is produced from condensation of one and cyclohexanol. kable finding is that significant amounts of cyclohex- are obtained from low cyclohexanone yields (media hexanol) but the profiles of condensation impurities in are growing exponentially as the reaction approaches m (media rich in cyclohexanone). yield is shown in Fig. 9. This impurity could be obtained exanol dehydration followed by dehydrogenation, in with that previously proposed in literature [7]. In fact, ofile agrees with the cyclohexene one in Fig. 6. On the the amount of this impurity formed in C1 and C2 is two agnitude lower than the corresponding cyclohexene gain, as temperature increases the yield of the impurity ties from dehydrogenation reactions hiometry a total yield of hydrogen, YH2 , defined as ow of hydrogen to molar flow of cyclohexanol fed to ratio, is calculated. Hydrogen yield from impurities, ich excludes the cyclohexanone produced, has been d. These hydrogen yields are given as percentages and the following expressions: × 100 = YONE + 2 × YCXENONE + 3 × YPhOH + 2 × YBZN (4) Y C X E N O N E (% ) Y C X E N O N E (% ) Fig. 12. 2-Cyc centage yield es obtained of hydrogen yield from impurities vs. cyclo- ield are shown in Fig. 10, for the three catalysts and 0,30 a key Catalyst 0,00 0,05 0,10 0,15 0,20 0,25 C1 C2 C3 806040200 0,00 0,02 0,04 0,06 0,08 0,10 b key Catalyst C1 C2 C3 Y ONE (%) lohexen-1-one percentage yield as a function of cyclohexanone per- for each catalyst and temperature tested, (a) 290 ◦C and (b) 250 ◦C. A. Romero et al. / Applied Catalysis A: General 392 (2011) 19–27 27 S H 2 im p 1,5 10 15 20 25 30 S H 2 im p key Catalyst C1 C2 C3 a Fig. 13. Selec percentage yie both tempe file of hydr Moreover, f yield and i equilibrium anone with increases al ties does. In Figs. 1 yields, resp are the mai this work. B As can b ties PhOH a Therefore, f that no diff Cu0, on deh Phenol i anda slight phenol amo but increas rium (medi is negligibl hexanol). T from cycloh hydrogen fr SH2 imp = Y values obta shown in F be seen in Fig. 13 higher hydrogen selectivity increases as follows C3>C2>C1, being this in agreement with the dehydration capacity obtained for these catalysts. 4. Conclusions Impurities from dehydration reactions were due to the pres- ence of alumina or chromium in the catalyst. Moreover, as acidity es so does the dehydration impurities. the file o ne y ts. T in de ount lly w ey ar um o oug rogen taly wled s wo tion thors supp nces itz, H. orn, J . A5, W . Witt 6, pp. ut, R. . Nikif . Lin . Sivar M. Me omer 806040200 0,0 0,5 1,0 1,5 10 15 20 25 30 Y ONE (%) key Catalyst C1 C2 C3 b 0,0 0,5 1,0 tivity to hydrogen from impurities as a function of cyclohexanone ld for each catalyst and temperature tested, (a) 290 ◦C and (b) 250 ◦C. rature tested. As can be seen in this figure, the pro- ogen from impurities is similar for the three catalysts. ormationofhydrogen isquite lowat lowcyclohexanone ncreases exponentially as the reaction approaches at . At these last conditions the media is rich in cyclohex- lower concentration of cyclohexanol. As temperature so the hydrogen produced due to formation of impuri- 1 and 12 are shown phenol and 2-cyclohexen-1-one increas For the pro hexano amoun active Am nentia that th maxim Alth dehyd with ca Ackno Thi Innova The au for its Refere [1] J. R ferk vol [2] H.A 199 [3] G. G [4] N.V [5] Y.-M [6] C.H [7] F.T. [8] A. R ectively, as percentage, vs. cyclohexanone yield. These n impurities from dehydrogenation reactions found in enzene yield was already shown in Fig. 9. een observed in Figs. 11 and 12 the profiles of impuri- nd 2-CXENONE are quite similar for the three catalysts. rom these results in Figs. 11 and 12, it can be inferred erences can be noticed for both copper sites, Cu+ and ydrogenation activity. s the most important impurity from dehydrogenation lymajor amount is generatedby theC3catalyst. Besides, unt produced is quite low at low cyclohexanone yield es exponentially as the reaction approaches at equilib- a rich in cyclohexanone). Rate of formation of phenol e at low cyclohexanone yield (medium rich in cyclo- herefore, it is assumed that phenol is produced mainly exanone, being this a remarkable finding. Selectivity to om impurities is obtained as: H2 imp YH2O (6) ined for SH2 imp vs. cyclohexanone percentage yield are ig. 13 for each catalyst and temperature tested. As can [9] D.V. Cesa 205–212 [10] V.Z. Fridm [11] P. Téténty [12] V.Z. Fridm [13] V.Z. Fridm [14] V. Siva Ku Reddy, K. [15] D. Ji, W. Z [16] B.M. Nag Reddy, B 345. [17] B.M. Naga Raju, K.S. [18] K.V.R. Ch 75–81. [19] J.M.Camp [20] M.V. Twi [21] Z. Wang, [22] G. Bai, X 2031–20 [23] R.G. Herm Catal. 56 [24] S. Menta, [25] F.M. Baut Romero, [26] F. García- [27] J.M. Rynk three catalysts used few differences are obtained in f impurities from dehydrogenation reactions vs. cyclo- ield, being phenol the impurity produced in higher herefore both species Cu+ and Cu0 can be considered hydrogenation and phenol formation. s of phenol and condensation impurities grow expo- hen reactions approach to the equilibrium suggesting e formed from cyclohexanone, which even explain the f cyclohexanone yield observed in Fig. 4. h the three catalysts tested were active in cyclohexanol ation it was noticed that higher activity was obtained st containing the smaller copper crystallite size. gment rk was funded by the Spanish Ministry of Science and under contract CTM 2006-00317 and PET 2008-0130. express their gratitude to UBE Corporation Europe S.A. ort and Süd-Chemie AG by catalyst supply. Fuchs, H. Kieczka, W.C. Moran, Caprolactam, in: F.Th. Cambell, R. Pfef- .R. Rounsaville (Eds.), Ullmann’s Encyclopedia of Industrial Chemistry, iley-VCH, Weinheim, 1986, pp. 31–50. cof, B.G. Reuben, Industrial Organic Chemical, John Wiley & Sons, Inc., 253–264. Jaeger, Chem. Eng. Sci. 37 (1982) 319–326. orova, K.A. Zhavnerko, Petrol. Chem. URSS 14 (1974) 25–31. , I. Wang, C.-T. Yeh, Appl. Catal. 41 (1988) 53–63. aj, M. Reddy, B. Kanta, P. Rao, Appl. Catal. 45 (1988) L11–L14. ndes, M. Schmal, Appl. Catal. A 163 (1997) 153–164. o, A. Santos, P. Yustos, Ind. Eng. Chem. Res. 43 (2004) 1557–1560. r, C.A. Peréz, V.M.M. Salim, M. Schmal, Appl. Catal. A 176 (1999) . an, A.A. Davydov, J. Catal. 195 (2000) 20–30. i, Z. Paál, J. Catal. 208 (2002) 494–496. an, A.A. Davydov, J. Catal. 208 (2002) 497–498. an, A.A. Davydov, K. Titievsky, J. Catal. 222 (2004) 545–557. mar, S. Sreevardhan Reddy, A.H. Padmasri, B. David Raju, I. Ajitkumar S. Rama Rao, Catal. Commun. 8 (2007) 899–905. hu, Z. Wang, G. Wang, Communication 8 (2007) 1891–1895. araja, V. Siva Kumar, V. Shashikala, A.H. Padmasri, S. Sreevardhan . David Raju, K.S. Rama Rao, J. Mol. Catal. A: Chem. 223 (2004) 339– raja, A.H. Padmasari, P. Seetharamulu, K. Hari Prasad Reddy, B. David Rama Rao, J. Mol. Catal. A: Chem. 278 (2007) 29–37. ary, K.K. Seela, D. Naresh, P. Ramakanth, Catal. Commun. 9 (2008) os-Martin,A.Guerrero-Ruiz, J.L.G. Fierro, J. Catal. 156 (1995)209–212. gg, Catalyst Handbook, 2nd ed., Wolfe, 1989 (Chapter 6). J. Xi, W. Wang, G. Lu, J. Mol. Catal. A: Chem. 191 (2003) 123–124. . Fan, H. Wang, J. Xu, F. He, H. Ning, Catal. Commun. 10 (2009) 35. an, K. Klier, G.W. Simmons, B.P. Finn, J.B. Bulko, T.P. Kobylinski, J. (1979) 407–429. G.W. Simmons, K. Klier, R.G. Herman, J. Catal. 57 (1979) 339–360. ista, J.M. Campelo, A. García, D. Luna, J.M. Marinas, R.A. Quirós, A.A. Appl. Catal. A Gen. 243 (2003) 93–107. Ochoa, A. Santos, Ind Eng. Chem. Res. 32 (1993) 2626–2632. owski, T. Paryjczak, M. Lenik, Appl. Catal. A Gen. 106 (1993) 73–83. Comparative dehydrogenation of cyclohexanol to cyclohexanone with commercial copper catalysts: Catalytic activity and impu... Introduction Experimental Chemicals and catalysts Catalytic activity Catalysts characterization Analytical methods Results and discussion Catalysts characterization Catalytic activity Impurities from dehydration reactions Impurities from dehydrogenation reactions Conclusions Acknowledgment References