Hydrogen production from CH4 dry reforming over bimetallic NieCoAl2O3 catalyst

Bimetallic 5%Nie10%CoAl2O3 catalyst was synthesized using impregnation method and evaluated for methane dry reforming reaction at different reaction temperatures. NiO, Co3O4 and spinal metal aluminates, namely, CoAl2O4 and NiAl2O4 phases were formed on gAl2O3 support surface during calcination process. 5%Nie10%CoAl2O3 catalyst exhibited reasonable surface area of 86.93 m2 g1 with small crystallite dimension of less than 10 nm suggesting that both Co3O4 and NiO phases were finely dispersed on the surface of support in agreement with results from scanning electron microscopy (SEM) measurement. Temperatureprogrammed calcination measurement indicates the complete thermal decomposition and oxidation of metal precursors, viz. Ni(NO3)2 and Co(NO3)2 to metal oxides and metal aluminates at below 700 K. Both CH4 and CO2 conversions were stable over a period of 4 h onstream and attained an optimum at about 67% and 71%, respectively at 973 K whilst H2 selectivity and yield were higher than 49%. The ratio of H2CO was always less than unity for all runs indicating the presence of reverse wateregas shift reaction. The activation energy for CH4 and CO2 consumption was computed as 55.60 and 40.25 kJ mol1, correspondingly. SEM micrograph of spent catalyst detected the formation of whiskerlike carbon on catalyst surface whilst D and G bands characteristic for the appearance of amorphous and grap

Trang 1

Hydrogen production from CH4 dry reforming over bimetallic Ni-Co/Al2O3 catalyst

Article · June 2017

DOI: 10.1016/j.joei.2017.06.001

CITATION

1

READS

81

6 authors , including:

Some of the authors of this publication are also working on these related projects:

Glycerol Dry Reforming View project

NAFOSTED View project

Ji Siang

Universiti Malaysia Pahang

6 PUBLICATIONS 5 CITATIONS

SEE PROFILE

Sharanjit Singh Universiti Malaysia Pahang

SEE PROFILE

Osaze Omoregbe

University of Benin

SEE PROFILE

Nguyen Huu Huy Phuc Toyohashi University of Technology

SEE PROFILE

All content following this page was uploaded by Dai-Viet N Vo on 04 December 2017.

Trang 2

Hydrogen production from CH 4 dry reforming over bimetallic

Ni eCo/Al2 O 3 catalyst

Tan Ji Sianga, Sharanjit Singha, Osaze Omoregbea, Long Giang Bachb,

Nguyen Huu Huy Phucc, Dai-Viet N Voa,d,*

a Faculty of Chemical & Natural Resources Engineering, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia

b Center for Advanced Materials Research, Nguyen Tat Thanh University, Ho Chi Minh City, Viet Nam

c Department of Electrical and Electronic Information Engineering, Toyohashi University of Technology, 1-1 Hibarigaoka, Tempaku, Toyohashi,

Aichi 441-8580, Japan

d Nguyen Tat Thanh University, 300A Nguyen Tat Thanh Street, Ward 13, District 4, Ho Chi Minh City, Viet Nam

a r t i c l e i n f o

Article history:

Received 21 April 2017

Received in revised form

1 June 2017

Accepted 1 June 2017

Available online xxx

Keywords:

Bimetallic NieCo catalyst

Methane dry reforming

Hydrogen

Syngas

a b s t r a c t

Bimetallic 5%Nie10%Co/Al2O3catalyst was synthesized using impregnation method and evaluated for methane dry reforming reaction at different reaction temperatures NiO, Co3O4and spinal metal alu-minates, namely, CoAl2O4and NiAl2O4phases were formed ong-Al2O3support surface during calcina-tion process 5%Nie10%Co/Al2O3catalyst exhibited reasonable surface area of 86.93 m2g1with small crystallite dimension of less than 10 nm suggesting that both Co3O4 and NiO phases were ﬁnely dispersed on the surface of support in agreement with results from scanning electron microscopy (SEM) measurement Temperature-programmed calcination measurement indicates the complete thermal decomposition and oxidation of metal precursors, viz Ni(NO3)2and Co(NO3)2to metal oxides and metal aluminates at below 700 K Both CH4and CO2conversions were stable over a period of 4 h on-stream and attained an optimum at about 67% and 71%, respectively at 973 K whilst H2selectivity and yield were higher than 49% The ratio of H2/CO was always less than unity for all runs indicating the presence of reverse wateregas shift reaction The activation energy for CH4and CO2consumption was computed as 55.60 and 40.25 kJ mol1, correspondingly SEM micrograph of spent catalyst detected the formation of whisker-like carbon on catalyst surface whilst D and G bands characteristic for the appearance of amorphous and graphitic carbons in this order were observed on surface of used catalyst by Raman spectroscopy analysis Additionally, the percentage of ﬁlamentous carbon was greater than that of graphitic carbon

1 Introduction

Increasing concerns about anthropogenic greenhouse gas emissions and the depletion of petroleum-based energy have renewed in-terests in the study of natural gas reforming processes for hydrogen production Hydrogen has been considered as a promising and green energy for replacing fossil fuels since it exhibits superior energy capacity (ca 120.7 kJ g1) to that of other common and alternative fuels[1]

and the combustion of H2results in the formation of water as a sole by-product[2] In addition, H2is used as a building block for producing synthetic fuel and valuable chemicals, namely, FischereTropsch synthesis (FTS)[3e5]and methanol production [6] The conventional approach for hydrogen production is methane steam reforming However, this method generates H2/CO ratio of 3 unfavourable for FTS, requires a higher H2O/CH4 ratio for increasing H2 yield and produces a considerable amount of greenhouse gas, CO2 [7] Hence, dry reforming of methane (DRM) has received signiﬁcant attention since it converts CO2to value-added products and forms syngas with H2/CO ratio of less than unity favoured for downstream methanol and FT synthesis[8] Co and Ni-based catalysts are conventionally employed for

* Corresponding author Faculty of Chemical & Natural Resources Engineering, Universiti Malaysia Pahang, Lebuhraya Tun Razak, 26300 Gambang, Kuantan, Pahang, Malaysia Fax: þ60 9 549 2889.

E-mail address: vietvo@ump.edu.my (D.-V.N Vo).

Contents lists available atScienceDirect Journal of the Energy Institute

j o u r n a l h o m e p a g e : h t t p : / / w w w j o u r n a l s e l s e v i e r c o m / j o u r n a l o f t h e e n e r g y

-ins t it ute

http://dx.doi.org/10.1016/j.joei.2017.06.001

Trang 3

hydrogen production in DRM owing to their high catalytic activity and selectivity[7,9] However, these catalysts may be deteriorated by carbon deposition and sintering during catalytic reaction[7] CoeNi supported on TiO2reportedly exhibited high catalytic performance and resisted to coke formation due to the presence of NieCo alloy[10] However, the synergetic effect and knowledge about bimetallic NieCo catalyst are still limited for DRM Particularly, the inﬂuence of operating parameters on catalytic performance, activation energy for product formation and reactant consumption as well as the nature of carbonaceous deposition on catalyst surface are important knowledge for kinetic studies and reactor design Thus, the objective of this research was to investigate the physicochemical attributes of 5%Nie10%Co/

Al2O3catalyst and examine the effect of reaction temperature on activity and selectivity of DRM

2 Materials and methods

2.1 Catalyst preparation

Gamma-Al2O3support supplied by SigmaeAldrich Chemicals was pretreated at 973 K in air for 5 h with a heating rate of 5 K min1to ensure thermal stability Bimetallic 5%Nie10%Co/Al2O3catalyst was prepared by co-impregnation method using calculated amounts of Co(NO3)2$6H2O, Ni(NO3)2$6H2O precursor solutions and pretreatg-Al2O3support Both aqueous metal precursor solutions were simulta-neously poured on the calcinedg-Al2O3support stored in a 150 ml glass beaker Afterwards, the resulting slurry was magnetically stirred for

3 h under ambient conditions and subsequently dried for 24 h inside an oven (Carbolite) at 373 K The dried solid was then calcined in air at

773 K for 5 h at a ramping rate of 10 K min1in the same apparatus followed by crushing and sieving to the average particle size of 100mm to obtain 5%Nie10%Co/Al2O3catalyst

2.2 Catalyst characterization

Multi-point BET surface area measurements were determined using N2physisorption at 77 K in a Thermo Scientiﬁc Surfer unit The TGA Q500 unit from TA Instruments was employed for temperature-programmed calcination (TPC) runs of uncalcined catalyst Catalyst sample was initially heated from ambient temperature to 373 K with a heating rate of 10 K min1in N2ﬂow of 100 ml min1and held isothermally

at this temperature for 30 min to remove moisture and volatile compounds It was subsequently heated to 1023 K in 100 ml min1ﬂow of 20%O2/N2mixture with different heating rates of 10e20 K min1 After keeping constantly at that temperature for 30 min, catalyst sample was cooled down to room temperature in the same gas mixture X-ray powder diffraction (XRD) measurements ofg-Al2O3support and 5%

Nie10%Co/Al2O3catalyst were conducted in a Rigaku Miniﬂex II system using Cu target at 30 kV and 15 mA with wavelength,l¼ 1.5418 Å. Samples were scanned from 3 to 80 with a scan speed of 1min1and a step size of 0.02 Scanning electron microscopy (SEM) mea-surements were conducted for morphological analysis of both fresh and used 5%Nie10%Co/Al2O3catalysts in a Carl Zeiss AG-EVO®50 Series unit The accelerating voltage was kept at 7 kV whilst platinum holder was employed during analysis Raman spectroscopy measurement of spent catalyst was performed in a JASCO NRS-3100 system in ambient air with a 532 nm green laser and the laser power was kept at less than 5 mW

2.3 Methane dry reforming experiment

Reaction runs were performed in a temperature-controlled tubularﬁxed-bed reactor (L ¼ 17 in and O.D ¼ 3/8 in) with CO2:CH4ratio of 1:1 at temperature range of 923 Ke973 K and atmospheric pressure About 0.1 g of catalyst with average catalyst particle size of 100mm mounted by quartz wool in the middle of stainless steel reactor was used per run Gas hourly space velocity (GHSV) was maintained constant

at 36 L gcat1h1for all runs to minimize transport-disguised kinetics Before conducting methane dry reforming evaluation, catalyst was reduced in situ in 50%H2/N2mixture of 60 ml min1at 973 K with a ramping rate of 5 K min1and kept isothermally at this temperature for

2 h Gaseous products were measured in an Agilent 6890 Series gas chromatograph equipped with both thermal conductivity (TCD) and ﬂame ionization (FID) detectors

The catalyst performance was evaluated in terms of reactant conversion (XCH4and XCO2), consumption rate (rCH 4), selectivity (Si, i: H2or CO) and yield of products (Yi) as provided in Eqs.(1)e(5)

Xjð%Þ ¼F

In

j FOut

j

FIn

j

rn

mol g1 s1

¼F

In

j FOut j

Sið%Þ ¼P ri

YH2ð%Þ ¼ F

Out

H 2

Trang 4

YCOð%Þ ¼ FCOOut

FIn

CH 4þ FIn

CO 2

where FIn

j and FOut

j are inlet and outlet molarﬂow rates (mol s1), respectively and W

Cat.is the weight of catalyst (g)

3 Results and discussion

3.1 Textural properties

As seen inTable 1, it was inevitable that the impregnation of active metals on support surface led to a reduction in surface area, average pore volume and pore diameter in comparison with that of calcinedg-Al2O3support This observation was rationally due to the successful incorporation of both Ni and Co metal oxides on the porous structure ofg-Al2O3support and hence decreasing its textural parameters, namely, average pore volume and pore diameter The drop in BET surface area of catalyst was also owing to the presence of metal particles on the pore entrance of support However, 5%Nie10%Co/Al2O3catalyst still exhibited reasonable surface area of 86.93 m2g1with average pore volume and pore diameter of 0.20 cm3g1and 11.21 Å, respectively comparable with other studies[11,12]

3.2 Temperature-programmed calcination

Fig 1shows the derivative weight proﬁle during temperature-programmed calcination of uncalcined catalyst with different heating rates

of 10e20 K min1 The derivative weight proﬁles was stable beyond 700 K indicating the complete thermal decomposition and oxidation of Ni(NO3)2and Co(NO3)2metal precursors to metal oxides and metal aluminates

As seen inFig 1, each thermal proﬁle has 2 characteristic peaks (P1 and P2) located at different temperatures regardless of ramping rates The high intensity peak located at low temperature of about 478 K was due to the decomposition of metal nitrates to metal oxides during calcination (see Eqs.(6) and (7))

NiðNO3Þ2/NiO þ 2NO2þ1

CoðNO3Þ2/CoO þ 2NO2þ1

The second peak (P2) was assigned to the subsequent oxidation of CoO and NiO phases to Co3O4and NiCo2O4, respectively at about 558 K

as given in the corresponding Eqs.(8) and (9) In fact, this observation is in line with the report from Foo et al.[8]

Table 1

Summary of textural properties ofg-Al 2 O 3 support and 5%Nie10%Co/Al 2 O 3 catalyst.

Sample BET surface area (m 2 g1) Average pore volume (cm 3 g1) Average pore diameter (Å)

Fig 1 Derivative weight proﬁles of 5%Nie10%Co/Al 2 O 3 catalyst during temperature-programmed calcination at different heating rates.

Trang 5

The small intensity peaks observed at high temperature range of 600e700 K suggested the formation of metal aluminates[13]may be described by:

In addition,Fig 2shows that peak temperature for both P1 and P2 shifted linearly to higher temperature with increasing heating rate and the derivative weight proﬁles exhibited similar shapes for all ramping rates This would suggest that the decomposition and oxidation reactions followed the same reaction mechanism irrespective of heating rates[14] Thus, activation energy, Ea(kJ mol1) and associated pre-exponential factor, A (s1) for the production of metal oxides and metal aluminates may be estimated from Kissinger equation[14]in order

to examine the activity of these solid decomposition and oxidation reactions:

ln b

T2

p

!

¼ ln

AR

Ea

withb, Tpand R being heating rate, peak temperature and universal gas constant, respectively The linear regressions of experimental data to Kissinger equation(12)exhibited a reasonablefit with correlation coefficient of essentially unity Thus, the associated Arrhenius parameters can be estimated from the slope and intercept of the plots for lnðb=T2Þ vs: ð1=TpÞ (seeFig 3) and are summarized inTable 2 The activation energy was about 89.51 and 80.03 kJ mol1for peaks P1 and P2, respectively The relatively low activation energy for peaks P1 and P2 would suggest the facile decomposition and oxidation processes of Ni(NO3)2and Co(NO3)2metal precursors to the correspondingfinal metal oxides and metal aluminates

3.3 X-ray diffraction measurement

The XRD patterns of calcined Al2O3support and 5%Nie10%Co/Al2O3catalyst are shown inFig 4 The Joint Committee on Powder Diffraction Standards (JCPDS) database was employed for analysing X-ray diffractograms[15] X-ray diffraction measure was conducted for calcined Al2O3support for comparison purpose As seen inFig 4a,g-Al2O3phase was detected with typical peaks at 2qof 19.28, 32.50, 37.16, 45.84, 60.22and 66.94 The discrete peaks located at 2q¼ 19.28and 31.46corresponded to the formation of Co3O4phase whilst NiO phase was detected at 36.98and 65.34 The characteristic peak observed at 2qof 31.46for bimetallic 5%Nie10%Co/Al2O3catalyst was ascribed to the formation of NiCo2O4phase The peaks belonging to spinal NiAl2O4(2q¼ 36.98, 45.00and 59.66) and CoAl2O4(2qof 31.46, 36.98, 45.00, 59.66and 65.34) phases were also observed consistent with other studies[8,12] Additionally, XRD results are in agreement with the results obtained from temperature-programmed calcination as seen inFig 1

The crystallite sizes of Co3O4and NiO phases were estimated from Scherrer equation (see Eq.(13))[16]as about 7.65 and 7.75 nm, respectively Both Co3O4and NiO phases possessed small crystallite size of less than 10 nm indicating theﬁne metal dispersion on support surface

Fig 2 Peak temperature versus heating rates during temperature-programmed calcination of 5%Nie10%Co/Al 2 O 3 catalyst.

Trang 6

with dpbeing crystallite dimension whilstl, B andqare wavelength, peak width and Bragg angle, respectively

3.4 Methane dry reforming evaluation

The transient proﬁles of CH4and CO2conversions for methane dry reforming at different reaction temperatures from 923 to 973 K with

CO2/CH4¼ 1 are shown inFig 5 It is apparent that catalyst performance was stable within 4 h on-stream for all reaction temperatures Both

CH4and CO2conversions were improved from 48% to 67% and 56% to 71% with increasing temperature, respectively In addition, CO2

Fig 4 X-ray diffractograms of (a) calcined alumina support and (b) 5%Nie10%Co/Al 2 O 3 catalyst.

Table 2

Summary of activation energy, E a and pre-exponential factor, A values for peak P1 and P2.

Fig 3 Estimated activation energy for the formation of metal oxides and NiCo 2 O 4 phases over 5%Nie10%Co/Al 2 O 3 catalyst during temperature-programmed calcination.

Trang 7

conversion was always greater than that of CH4regardless of reaction temperature suggesting that CO2was simultaneously consumed by the reverse wateregas shift and/or CO2gasiﬁcation reactions (Eqs.(14) and (15), respectively) during dry reforming of methane reaction[9]

In fact, the general mechanistic pathway for methane dry reforming reaction is reportedly a two-step process, in which H2is initially generated via CH4dehydrogenation into carbonaceous CxH1 x(x 1) species (see Eq.(16)) subsequently gasiﬁed by CO2reactant for producing CO gas as shown in Eq.(17) [17,18]

Table 3

Summary of catalytic activity for dry reforming of methane over Ni- and Co-based catalysts reported in literature.

Catalysts Operating conditions CH 4 conversion (%) CO 2 conversion (%) References

T (K) CH 4 :CO 2 ratio TOS a (h)

a TOS: time-on-stream.

b n.m.: not mentioned.

Fig 5 Effect of reaction temperature on (a) CH 4 and (b) CO 2 conversions at CO 2 /CH 4 ¼ 1:1.

Trang 8

5x 1 2

CxH1xþ xCO2/2xCO þ

1 x 2

In order to assess the efﬁciency of this bimetallic NieCo/Al2O3 catalyst, the catalytic performance of several Ni- and Co-based catalysts is summarized inTable 3for comparison purpose The bimetallic NieCo/Al2O3catalyst in this study exhibited superior CH4

and CO2conversions to those of monometallic Ni-based and Co-based catalysts Additionally, in comparison with Pt-promoted and Mn-promoted Ni/Al2O3catalysts, the bimetallic NieCo/Al2O3catalyst showed a comparable catalytic activity This observation could reveal its potential and promising use for replacing noble metal catalysts in large-scale methane dry reforming production from an economic standpoint

The selectivity and yield of hydrogen were also stable with time-on-stream as shown inFig 6 It is apparent that an increase in reaction temperature from 923 to 973 K signiﬁcantly enhanced YH 2by up to 60% (seeFig 6b) whilst the slight improvement of SH2was observed as seen inFig 6a Temperature of 973 K seems to be the optimal reaction temperature within temperature range of 923e973 K since it possessed greatest reactant conversions and exhibited the highest H2selectivity and yield of about 49.5% and 58.0%, respectively

Figs 7 and 8show the effect of reaction temperature on the corresponding reactant consumption rates and formation rates of gaseous product at CO2/CH4¼ 1 It is evident that consumption and formation rates increased with growing reaction temperature Hence, the

Fig 6 Effect of reaction temperature on (a) selectivity and (b) yield of H 2 at CO 2 to CH 4 ratio of 1:1.

Trang 9

Fig 9 Estimation of Arrhenius parameters for methane dry reforming at CO 2 :CH 4 ¼ 1:1.

Fig 8 Effect of reaction temperature on formation rates of H 2 and CO at CO 2 :CH 4 ¼ 1:1.

Fig 7 Effect of reaction temperature on consumption rates of CH 4 and CO 2 at CO 2 :CH 4 ¼ 1:1.

Trang 10

activation energy, Eaand pre-exponential, A can be computed from Arrhenius plots (as seen inFig 9) with correlation coefﬁcient, R2of 0.90e0.99 As seen inTable 4, activation energy for both reactants and products was relatively low and varied from 40 to 63 kJ mol1

Fig 10a shows H2:CO ratio at different reaction temperatures and the constant feed composition of CO2:CH4¼ 1:1 Regardless of reaction temperature, it is evident that H2/CO ratio was less than unity appropriate for using as feedstock for FT synthesis to generate long-chain hydrocarbons reasonably owing to the reverse wateregas shift reaction given in Eq.(14) [9] In fact, the ratio of H2production rate to

CH4consumption rate (rH2=rCH 4) was inferior to two (seeFig 10b) suggesting the formation of other hydrogen-containing products, viz

H2O arising from the reverse wateregas shift reaction The ratio of H2/CO also increased with rising temperature and approached to nearly unity at 973 K The enhancement of H2/CO ratio was possibly due to the increasing rate of CH4dehydrogenation reaction given by;

Fig 10 Effect of reaction temperature on (a) H 2 :CO ratio and (b) r H 2 =r CH 4 of methane dry reforming at CO 2 :CH 4 ¼ 1:1.

Table 4

Arrhenius parameters for methane dry reforming reaction at CO 2 :CH 4 ¼ 1:1.

Species Activation Energy, E a (kJ mol1) Pre-exponential factor, A (s1)

Định dạng
Số trang	13
Dung lượng	2,74 MB