CC BY
11Journal of Siberian Federal University. Chemistry 1 (2012 5) 3-17
УДК 546
Dechlorinative Oligomerization of Multiply Chlorinated Methanes Catalyzed by Activated Carbon Supported Pt-Co
Vladimir I. Kovalchuka,* William D. Rhodesa and Mark A. McDonaldb
a Department of Chemical and Petroleum Engineering, University of Pittsburgh, Pittsburgh, PA 15261, United States b National Energy Technology Laboratory (NETL), United States Department of Energy, Pittsburgh, PA 15236, United States 1
Received 2.03.2012, received in revised form 9.03.2012, accepted 16.03.2012
The hydrodechlorination of dichloromethane, trichloromethane, and their mixtures catalyzed by a Pt-Co/C catalyst has been investigated in an effort to elucidate the chemistry associated with the generation of hydrocarbon oligomerization products. In the reaction of dichloromethane with hydrogen, the catalyst did not exhibit deactivation and maintained the steady-state activity within 18 h on stream at 523 K; whereas, when trichloromethane was added or converted in the absence of dichloromethane, significant deactivation occurred within the first 5 h on stream. Hydrocarbon oligomerization products were observed with all three reaction mixtures; the selectivity followed the order dichloromethane + dihydrogen < trichloromethane + dihydrogen < dichloromethane + trichloromethane + dihydrogen. The generation of ethane and propane was virtually independent of the reaction mixture composition. However, selectivity toward ethylene and propylene was significantly greater with the trichloromethane + dihydrogen and dichloromethane + trichloromethane + dihydrogen mixtures compared to the dichloromethane + dihydrogen feed. It was concluded that the saturated hydrocarbon products are formed by means of the alkyl mechanism of hydrocarbon chain growth; whereas, the alkenyl mechanism is responsible for the formation of the unsaturated hydrocarbons.
Keywords: hydrodechlorination, chloromethanes, dehalogenative oligomerization, Fischer-Tropsch synthesis, reaction mechanism, platinum, cobalt.
Introduction
The interest in halocarbon chemistry is motivated by the fact that the halogenated molecules remain important intermediates and products of the chemical process industries [1] and they have detrimental
* Corresponding author E-mail address: [email protected]
1 © Siberian Federal University. All rights reserved
impact on the environment [2,3]. Hence, low cost methods to detoxify unwanted halocarbons and their wastes can be considered a societal need. The value of such methods would be significantly increased if they generated valuable products. From this standpoint, dehalogenative oligomerization of Ci halocarbons to form unsaturated and saturated C2+ hydrocarbons or halocarbons is more valuable than the simple and well-established hydrodechlorination process in which halogen atoms in a C1 halocarbon are replaced by hydrogen to form methane [4-7].
Among halomethanes, the dehalogenation of CF2Cl2, CCl4, and, to the much lesser extent, CHCl3 has drawn attention. Dichlorodifluoromethane, a common refrigerant in the recent past, is still present in various applications; whereas, CHCl3 and CCl4 are widely used solvents and now are the most prevalent ground water contaminants [7].
It has been shown that dechlorinative oligomerization of CCl4 and CHCl3 to form C1-C7 hydrocarbons with selectivity greater than 90% is catalyzed by supported Pd [8-14] in the temperature range of 423-643 K and with chloromethane to H2 ratios of 5-15. Under similar conditions, Pt catalyzes the oligomerization of CCl4 to partially or completely chlorinated C2 hydrocarbons with selectivity ranging from 90 to a few percent [15-18] depending on the catalyst precursor, type of the catalyst support, and reaction conditions. There are no C2+ products in the Pt catalyzed hydrodechlorination of CHCl3 [8].
A comparative investigation of CF2Cl2 hydrodechlorination and hydrodehalogenative oligomerization catalyzed by activated carbon-supported Group VIII noble metals revealed the high oligomerization selectivity of Pd/C (~ 75%), which produced mainly saturated and unsaturated C2-C3 hydrocarbons at 523 K and CF2Cl2 to H2 mole ratio of 1 [19]. The oligomerization selectivity of Pt/C under the same reaction conditions was less than 5% [19].
The selectivity of supported Pt toward oligomerization products in the CF2Cl2 + H2 reaction changes dramatically in bimetallic catalysts. For example, addition of Co to Pt/C results in the steady-state selectivity toward C2+ hydrocarbons of 50%, with C2-C3 olefins and paraffins as the main oligomerization products (523 K, CF2Cl2 to H2 mole ratio of 1) [20]. For the Pt-Cu/C catalysts, the oligomerization selectivity under the same reaction conditions is a function of Pt/Cu atomic ratio and time on stream [21]. This selectivity is low initially but increases with time to exceed 70% toward C2+ hydrocarbons at steady state for the catalysts with Cu/Pt atomic ratio greater than 6.
Based on the research noted, it should be concluded that dehalogenative oligomerization of multiply halogenated methanes is not uncommon in the catalysis of halocarbon dehalogenation and under certain conditions could be considered an alternative route to hydrodechlorination. Surprisingly, investigations of dehalogenative oligomerization of multiply halogenated methanes are not abundant. Specifically, the mechanism of the hydrocarbon chain growth during C1 halocarbon dehalogenative oligomerization has not been studied. As the oligomerization product distribution follows the Anderson-Schulz-Flory statistics, it was speculated that the hydrocarbons are formed via a polymerization of surface C1 species [8,9,11] similar to the carbide mechanism of the Fischer-Tropsch synthesis [22]. However, no suggestion on the nature of the C1 species (CH, CH2, or CH3) was speculated [8,9,11]. This is quite understandable since the dehalogenative oligomerization investigation of CF2Cl2 or CCl4 [9-14,19,21] does not allow one to address the question of which CHx moieties couple to initiate C-C chain growth because both C-Cl and C-F bonds readily dissociate on the transition metal surface to form bare carbon atoms [9,19]. Subsequent interaction of the carbon atoms with adsorbed H atoms will
result in a set of CHx species of different stoichiometry; any possible combination of which may initiate the C-C oligomerization.
The objective of the present research is to shed light on the nature of the C1 species that are responsible for initiation and propagation of the hydrocarbon chain in the reaction of dehalogenative oligomerization of multiply halogenated methanes. It has been established that C-Cl bonds of chloromethanes readily dissociate on the surface of Group VIII metals before any carbon-hydrogen bond is broken [6,23,24]. Hence, it is reasonable to suggest that the dissociative adsorption of CH2Cl2 and CHCl3 on the metal surfaces will result in the formation of CH2 and CH species, respectively. Thus, the reaction of CH2Cl2 + H2, CHCl3 + H2, and CH2Cl2 + CHCl3 + H2 mixtures catalyzed by Pt-Co/C was investigated in an effort to differentiate between the so-called "alkyl" [25,26] and the "alkenyl" [27] mechanisms of hydrocarbon formation from multiply-halogenated methanes under conditions of halocarbon hydrodehalogenation. The catalyst choice was due to its high oligomerization selectivity in the CF2Cl2 + H2 reaction [20].
Experimental
Preparation of bimetallic Pt-Co/C
Activated carbon (Calgon Carbon BPLF3, 6-16 mesh) was crushed and sieved. A 24-60 mesh fraction (1400 m2 g-1, 2.4 nm average pore diameter) was used as the support. The catalyst was prepared by pore volume co-impregnation of the support with solutions of H2PtCl6-6H2O (Alfa, 99.9%) and CoCl2-6H2O (Aldrich, 98%) in 1 N aqueous HCl (EM Science). The concentrations of metal precursors in the impregnating solution were adjusted to obtain a metal loading of 0.3% Pt and 0.9% Co (1:10, atomic ratio). After impregnation, the material was allowed to equilibrate overnight before drying at ambient temperature and pressure for 24 h and then at 373 K for 2 h in vacuum (~25 Torr).
Kinetics experiments
Dechlorination of CH2Q2 (Aldrich, purity > 98%), CHCb (Aldrich, purity > 99%), and their mixture was performed at ambient pressure in a stainless-steel flow reaction system connected to a down-flow quartz microreactor (15 mm i.d.) in which the catalyst was supported on a quartz frit. The reactor zone containing the catalyst was heated by an electric furnace. The temperature of the catalyst was measured and controlled with an accuracy of ± 1 K with a temperature controller (Athena Series 6075). Gaseous reactants were metered using mass flow controllers (Tylan General, FC-280) and mixed prior to entering the reactor. Liquid CH2Cl2 and CHCl3 maintained at 274 K and 297 K, respectively, were metered into the reaction system via saturators; He was the carrier gas. Saturation was confirmed by varying the He flow rate through the saturators and quantifying the gas phase CH2Cl2 and CHCl3 by a gas chromatograph (GC) (HP 5890 series II).
Prior to the reaction, the catalyst was reduced in a mixture of H2 (20 ml min-1) and He (30 ml min-1) (Butler, each >99.99%) as it was heated from ambient temperature to 673 K at the rate of 25 K min-1 and then held at 673 K for 2 h. Next, the catalyst was cooled in He (30 ml min-1) to the reaction temperature. For the dechlorination reaction, 0.30 g of catalyst was used and the total flow of the reactant mixture through the reactor was 55 ml min-1. The flow consisted of CH2Cl2 (43,640 ppm), H2 (43,640 ppm), and He (balance) for the CH2Cl2 + H2 reaction; CHCl3 (43,640 ppm), H2 (43,640 ppm), and He (balance) for the CHCl3 + H2 reaction; and CH2Q2 (21,820 ppm), CHQ3 (21,820 ppm), H2 (43640 ppm), and
He (b alaace) for the; CH2Cl 2 + CHCl3 + H2 reaction. The aeaction temaarature war maintained at 523 K. The reactor effluent was analyzed on-line by a GC to identify the reaction products. The GC was equipped with a 9 m 80/100 Porapak Q packed column (Supelco) and a flame ionization detector (FID) capable of detecting o onc/ntratio ns > 1 ppm for all chlorocarbons and hydrocarbons involved in this study. Hydrogen chlorife, a reaction byproduct, wat not quantified.
The kinetics behavior of the three sets of reactants was compared at similar conversion levels f2-3%). The conversion (X was calculated as follows:
CcHxCly - YfliCi
X = -C-x 100%,
CcHxCly
wheae CcHxciy is thi molh fracthon of halomeahane(s) in the feed, nt and C,- are the member of carbon atoms in a product molecule and the mole fraction of the product j in effluent gas, respectively. As the true aonmersion of each reactant for ehe mixture of CH2 Cl2 and CHCl3 could nat be determined herein, rte comeined conversion was obtcined in this case, calculated as rhe mole fraction of CH2Cl2 and CHCl3 converted to reaction products. The selectivities (S) toward detectable carbon-containing products were cclculaOed as Sallows:
f n-— x100%.
Results
The Pt-Co/C Cataiyzed CI three C^2<Cl2 + i CHCl3 + H2, and CH2Cl2 + CHCl3 + H2 reactions to form eeactant-specific sets oa partia3ly and compktely dechloainaCed products . The pure support showed negligible activity1 initially ind was completely inacaive thereafter. The highest initial activity on a pea mhrs of catelyat I)asis waa obseraed for the CHCl3 + H2 reaction mixture; whe reas, with the CH2C12 + H2 mixture, the cstalyst exhibited Ihe loweat initial activity (Fig. 1). Thic activity order is smilar to that repoated elsewheae for suppchteC Pd [9].
TSe charactefislic feature of the; CH2C32 + H2 feed \a/it^ tte absence of catvlyst deactivation; after an initial 50% increase, ehe activity level attained steady-shaOe ft. 14 to 0.188 |imol gcat-i s-1) within 3 h od stream. In hontrast, with thres CHCl3 + H2 ^nd CH2(dl2 + CHC13 -a H2 fends, significant deactivation occurred (a Ccctoa of 32 and 20 during the first 5 h, respectivelf) until steady stare al essentially the same level was achiesed after ~10 o (in sVeam for Voth casas (Fig. ,1). The CH2Cl2 + CHCl3 + H2 mixlure clfo caused cfeactivatioc of PdiTiO2 as rrported elsewhete [o8]. The I>t-Co/C catalyst activities for all three reoction mixtures were very close (0,S0 to 0.17 |imol gcat-1 s-1) after 22 ho on stream (Fig. 1). Thin activity1 level range corresponds to the concersion range of 2-3%.
Product selectivities for the three different reaction mixtures were compared at the 2-h point since this is where all mixtures attained similar converrion levels. Toe reaction products were classified in two categories: C1 and C2+ (oligomerization). With the CH2Cl2 + H2 feed, CH4 and CH3Cl were the major and minor C1 products, respectively (Fig. 2). Methylene chloride and CH4 were the major and minor C1 products with the CHQ3 + H2 mixture (Fig. 2). In the case of CH2Q2 + CHQ3 + H2 feed, CH2Cl2 was both a reactant and a product. Hence, a compromise was necessary to extract meaningful product selectivity information. The approach taken was to determine the mjnjmum CH2Cl2 product
0.6
о
° n с ся 0.5
0.4
"o
о о
S с
0.3 0.2 0.1 0. 0
5 10 15
Time-on-stream, h
Fig. 1. Activity on per gram of catalyst basis vs. time-on-stream for Pt-Co/C; (□) - CH2Cl2 + H2, (▲) - CHCl3 +
H2, (V) - CH2C12 + CHC13 + H2
100
60
40
00
О 20
I I CH4 О ch3ci □ CH2C1.
22
i
li
Ji
CHjClj CHC13 CHjClj & CHC13
Chlorocarbon Reactant(s)
Fig. 2. C1 product distribution after 2 h on steam for the CH2Cl2 + H2, CHCl3 + H2, and CH2Cl2 + CHCl3 + H2 reactions catalyzed by Pt-Co/C
0
selectivity by assuming that none of the CH2Cl2 reactant was converted. Accordingly, for the case where both CH2Cl2 and CHCl3 were reactants, the major and minor Q products were CH4 and CH2Cl2, respectively.
The oligomerization products of the CH2Cl2 + H2, CHCl3 + H2, and CH2Cl2 + CHCl3 + H2 reactions catalyzed by Pt-Co/C consisted of saturated and unsaturated C2-C4 hydrocarbons (Fig. 3). The CH2Cl2 + H2 feed resulted in the total C2+ selectivity of approximately 5.5%; whereas, the oligomerization selectivities for the CHCl3 + H2, and CH2Cl2 + CHCl3 + H2 reaction mixtures were 9.5 and 11.5%, respectively. With the CH2Cl2 + H2 mixture, the C2+ product distribution indicates that the hydrocarbons may have formed according to the classic Andersoc-Schulz-Flory statistics for
'>
o
GO
+
O 2\-
CHft CHCl, CH,^ & CHCl3
Chlorocarbon Reactant(s)
Fig. 3. C2-4 product distribution after 2 h on stream for the CH2Cl2 + H2, CHCl3 + H2, and CH2Cl2 + CHCl3 + H2 reactions catalyzed by Pt-Co/C
o o
I I ethylene I I ethane propene Igjgjgl propane
É
1
i
tL
CH2Cl2 CHCl3 CHjClj & CHCl3
Chlorocarbon Reactant(s)
Fig. 4. C2-3 product distribution after 2 h on stream for the CH2Cl2 + H2, CHCl3+ H2, and CH2Cl2 + CHCl3 + H2 reactions catalyzed by Pt-Co/C
4
3
2
polymerization of surface Q species [29]. However, with the other two reaction mixtures, abnormally high selectivity toward C4 hydrocarbons as compared to that toward C2 and C3 was observed (Fig. 3).
A characteristic feature of the CH2Cl2 + H2 feed was the prevalence of saturated hydrocarbons among C2 and C3 oligomerization products (Fig. 4). The C2H6/C2H4 and C3H8/C3H6 mole ratios were 7.7 and 3.5, respectively. (The GC column employed in the present investigation did not allow determination of the olefin fraction for Ct hydrocarbons.) With the CHCl3 + H2 and CH2Cl2 + CHCl3 + H2 reaction mixtures, selectivity toward propylene exceeded that toward propane by a factor of 2-3; whereas, selectivities toward ethane and ethylene were comparable (Fig. 4).
16
14
lo 12
a
10
8
tc
el
le 6
G«
+ CN 4
C
2
0
I I 1 hour HH 2 hour
hi
rii
CHjClj CHCl, CHjClj & CHClj
Chlorocarbon Reactant(s)
Fig. 5. Oligomerization product selectivity at the early stage of time on stream for the CH2Cl2 + H2, CHCl3 + H2, and CH2Cl2 + CHCl3 + H2 reactions catalyzed by Pt-Co/C
10
o 8
6
o J2 <d oo
<D fl
Ji
^
w
v V V
V v
V
V 7
A A ▲
V
□ □□□□□□□
□ □□□□OOO
5
10
15
20
Time-on-stream, h
Fig. 6. Ethylene selectivity vs. time-on-stream for Pt-Co/C; (□) - CH2C12 + H2, (A) - CHC13 + H2, (7) - CH2C12 + CHC13 + H2
0
0
Oligomerization selectivity improved with time on stream for the CHCl3 + H2 and CH2Cl2 + CHCl3 + H2 feeds but remained essentially invariant for the CH2Cl2 + H2 mixture (Fig. 5). The most profound effect of time on stream was observed for the ethylene selectivity (Fig. 6). For the CH2Cl2 + CHCl3 feed, the ethylene selectivity increased from 1% after 1 h on stream to 8% after 14 h; with the CHCl3 + H2 feed, the increase was from 2 to 5 % for the same time on stream. Similar to the overall oligomerization selectivity, ethylene selectivity did not depend on time on stream for the CH2Cl2 + H2 mixture. Finally, the selectivity toward ethane did not depend on time on stream for all three reaction mixtures.
Discussion
A molecular-level understanding of the reaction pathways that lead to the formation of C2+ hydrocarbons from multiply-halogenated methanes is crucial for the development of economical routes to the elimination of chlorinated hydrocarbons by dechlorinative oligomerization. In addition, this understanding will contribute to the knowledge base of Q chemistry. However, the mechanism of the hydrocarbon chain growth during Q halocarbon dehalogenative oligomerization has not been studied. Although it had been speculated that the mechanism is similar to that of the CO+H2 (Fischer-Tropsch) reaction, without further details [8,9,11].
For the CO+H2 reaction two main mechanisms have been discussed in literature: the so-called alkyl [25,26] and the alkenyl mechanism [30-34]. According to the alkyl mechanism (Fig. 7), the chain is initiated by reaction of an H adatom and a methylene to form a methyl species that reacts with another methylene to form an ethyl species. Subsequently, methylene species add step-wise for chain growth. Lastly, chain growth is either terminated by p-elimination of hydride to form a-alkenes (primary reaction products), or hydrogenolysis of the metal-alkyl bond results in the formation of alkanes. According to the alkenyl mechanism (Fig. 8), ethylene forms via the coupling of the CH and CH2 species to form a surface vinyl radical followed by hydrogenolysis of the metal-vinyl bond. The C2+ olefins form via CH2 group insertion into the metal-vinyl or metal-alkylvinyl bond. And chain termination occurs with hydrogenolysis of the metal alkenyl bond. Both alkyl and alkenyl mechanisms of chain growth are supported by the results of theoretical calculations [35-39].
The present investigation provides evidence that different mechanisms are responsible for the formation of C2+ products from the CH2Cl2 + H2 and from the CHCl3 + H2 and CH2Cl2 + CHCl3 + H2 reaction mixtures. The kinetics results are consistent with the suggestion that with the former mixture, hydrocarbons form predominantly by means of the alkyl mechanism; whereas, the alkenyl mechanism of oligomerization product formation prevails with the latter two reaction mixtures.
Indeed, with the CH2Cl2 + H2 reaction mixture, methylene (CH2) should be a dominant C1 surface species during the dechlorination reaction as C-Cl bonds of hydrocarbons readily dissociate on Group VIII metals [6,23,24]. Even though both Pt and Co are able to dissociate hydrocarbon C-H bonds at temperatures of several hundred Kelvins to form surface CHx species [40-42], their further decomposition is significantly suppressed in the presence of hydrogen [23,24]. This is not surprising because the dissociation of C-H bonds is a reversible reaction [43] ; it is not likely to occur on a catalyst surface covered with hydrogen, as the dissociation of H2 is not an activated process on both Pt and Co [44,45]. Nevertheless, the classic alkyl mechanism of the C2+ hydrocarbon formation (Fig. 7) would hardly take place with the CH2Cl2 + H2 feed. The hydrocarbon chain initiation by the coupling of CH2 and CH3 (Fig. 7) does not seem likely because this elementary step has a high activation barrier on metal surfaces [37]. Formation of propylene by p-elimination of hydride from adsorbed alkyl species (Fig. 7) does not look likely either. This elementary step is reversible and should be suppressed on metal surfaces saturated with hydrogen. Thus, the only feasible option is that C2-C4 hydrocarbons in the CH2Cl2 + H2 reaction catalyzed by Pt-Co/C form by surface oligomerization of CH2 species followed by the hydrogenolysis of adsorbed (CH2)n with surface hydrogen to form alkanes. This is consistent with the prevalence of ethane and propane among C2 and C3 hydrocarbon products (Fig. 4).
An increase in total oligomerization selectivity and dramatically different olefin to paraffin ratios among C2 and C3 hydrocarbon products with the CHCl3 + H2 and CH2Cl2 + CHCl3 + H2 reaction
i> H2 H2
i A A
H?
A
H2 H2
?H3A A
CH3GH2CH2
H + CH3CH=CHj
J
Fig. 7. Schematic of the alkyl mechanism for the polymerization of surface methylene to surface alkyls. (Reproduced with permission from Ref. [27])
Fig. 8. Schematic of a catalytic cycle for the formation of alkenes, incorporating the alkenyl mechanism for the polymerization of surface methylene involving surface alkenyls. (Reproduced with permission from Ref. [27])
mixtures as compared to the CH2Cl2 + H2 feed (Fig. 3, 4) suggest the involvement of the CH species in surface oligomerization. The CH (methylidyne) species play a crucial role in the alkenyl mechanism of chain initiation (Fig. 8). If the formation of C2 surface species occurred solely by the alkyl mechanism, methylidyne would tasically be a surface spectator until hydrogenated to a methylene and no enhancement in oligomerizatioo selectivity would tie observed with the CHCl3--containing reeds. In addition, when CH species participate in hydrocarbon chain initiation, the chain termination occurs by additicn of hydrogen to the surface species resulting in olefins formation (Fig. 8), not paraffins as it is expected for the alkyl mechanism. Hence, the greater olefin-to-paraffin ratios for CHCl3-containing reaction mixtures, as compared to CH2Cl2 + H2 (Fig. 4), is consistent with the alkenyl mechanism of the oligomerization product formation.
Abnormally high selectivity toward C4 hydrocarbons in the CHCl3 + H2 and CH2Cl2 + CHCl3 + H2 reactions compared to that toward C3 products is another support for the alkenyl mechanism of oligomerization with the CHCl3-containing reaction mixtures. It has been shown, that dimerization of vinyl, CH=CH2, species is a facile reaction on metal surfaces [31,46,47]. Addition of vinylic probes to ruthenium- and rhodium-catalyzed Fischer-Tropsch reaction resulted in much higher incorporation of
the probes into C4 hydrocarbons than into C3 ones [31,47] suggesting that vinyl dimerization occurs at a greater rate than the CH2 insertion into metal-vinyl bond (Fig. 8).
Based on a lower reactivity of CH2Cl2 in hydrodechlorination reactions in comparison with CHCl3 [9] and the absence of CH3Cl among the Q reaction products with the CH2Cl2 + CHCl3 + H2 feed (Fig. 2), one may conclude that CH2Cl2 does not convert to reaction products being mixed with CHCl3. However, if CH2Cl2 did not convert, the olefin-to-paraffin ratios for the C2-C3 hydrocarbon products would be similar for the CHCl3 + H2 and CH2Cl2 + CHCl3 + H2 reaction mixture. However, the ratios are sufficiently lower for the latter feed (Fig. 4). This result is consistent with the formation of a fraction of the C2+ products by the alkyl mechanism resulting in paraffins, which is characteristic of the CH2Cl2 dechlorinative oligomerization as discussed above.
There are distinct differences in time-on-stream performance of the Pt-Co/C catalyst with different reaction mixtures. The CHCl3 + H2 and CH2Cl2 + CHCl3 + H2 feeds cause severe catalyst deactivation and a significant increase in oligomerization product selectivity with time on stream; whereas, with the CH2Cl2 + H2 mixture, both the catalyst activity and C2+ product selectivity remained essentially invariant during the course of the dechlorination reaction (Fig. 1, 5, and 6).
In general, there are two possible phenomena that may result in catalyst deactivation with time on stream: accumulation of chlorine adatoms on the metal surface and the formation of carbonaceous deposits [48]. The lack of marked catalyst deactivation with the CH2Cl2 + H2 feed provides strong evidence that surface chlorine does not affect catalyst activity. Moreover, the virtual independence of the activity on time on stream (Fig. 1) supports the suggestion (vide supra) that the dehydrogenation of surface methylene to the coke-precursor methylidyne species is not facile on the Pt-Co/C catalyst under the reaction conditions, even though CH is the most stable thermodynamically species on Co and Ru [36,49]. It is worth noting that, in the absence of catalyst deactivation, an increase in catalyst activity due to chlorine-induced metal redispersion [50-52] becomes observable for the CH2Cl2 + H2 feed at early times on stream (Fig. 1).
In contrast to the CH2Cl2 case, when CHCl3 is included in the reaction feed stream, significant deactivation occurs immediately (Fig. 1). The deactivation could be related to methylidyne surface species formation. However, the straightforward dissociation of CH moieties to form carbonaceous deposits is not likely. First of all, the dissociation is an endothermic process on metal surfaces [36,5355]. Secondly, as with any reversible surface reaction, it is disfavored by co-adsorbed hydrogen atoms. Most likely, the deactivation is caused by blocking the catalyst surface by high molecular-weight unsaturated hydrocarbons formed by surface polymerization of the CH species.
There is evidence in literature suggesting that halogen atoms affect the chemistry of hydrocarbon radicals on metallic surfaces [43,56-58]. However, the literature results are controversial in terms of how the adsorbed halogen atoms influence the CHx coupling chemistry. Results of reaction kinetics investigations suggest that surface halogen atoms promote coupling [5]; whereas, ultrahigh vacuum studies infer that adsorbed halogen atoms suppress coupling chemistry by blocking sites adjacent to the adsorbed CHx species [58,59].
The conclusion that surface halogen atoms favor the oligomerization of surface C1 species comes from the fact that the oligomerization selectivity in halomethane dehalogenation reactions increases with time on stream in parallel with an increase in the surface concentration of halogen atoms [5]. It is quite possible that at early times on stream - when the concentration of halogen adatoms is low -
the major fraction of CH species polymerize to form high molecular-weight hydrocarbons, which do not desorb, and the oligomerization selectivity stays low. In fact, it is an apparent selectivity because these heavy hydrocarbons are not taken into account at the selectivity calculation. As the catalyst equilibrates with the reaction medium, the concentration of halogen adatoms increases until it reaches steady state. This increase results in blocking a fraction of sites around CH surface species turning the polymerization of the CH species into di-, tri-, and tetramerization. As a consequence, the oligomerization selectivity increases as observed in the present investigation (Fig. 5, 6). In ultrahigh vacuum experiments, when CHx species on the single crystal surfaces are obtained by the dissociation of chloro- or bromomethanes, the concentration of halogen adatoms is high initially and does not change with time in the absence of H adatoms. The halogen concentration is too high to allow coupling reactions to occur.
Conclusion
The distribution of oligomerization products in the reactions of CH2Cl2, CH3Cl and their equimolar mixtures with hydrogen catalyzed by Pt-Co/C sheds light on the mechanism of hydrocarbon chain growth in the dehalogenative oligomerization of multiply halogenated methanes. The reaction of CH2Cl2 with hydrogen resulted in predominantly saturated C2+ hydrocarbons as the oligomerization products. When CHCl3 alone or in the mixture with CH2Cl2 was converted, olefins dominated among the hydrocarbon products. This is consistent with the suggestion that CH surface species play a crucial role in the formation of unsaturated C2+ hydrocarbons by means of the alkenyl mechanism, while the alkyl mechanism is responsible for the formation of paraffins by surface oligomerization of CH2 species. A concomitant effect of the CH species presence on the catalyst surface is catalyst deactivation, probably due to the facile polymerization of the CH moieties to form high molecular weight hydrocarbons. Continuous increase in oligomerization selectivity vs. time on stream with CHCl3-containing feeds was interpreted in terms of the accumulation of the Cl adatoms, which suppress the polymerization of CH moieties favoring their oligomerization.
Acknowledgment
This research was made possible by the NETL Regional University Alliance (NETL-RUA), an applied research collaboration that combines NETL's fossil energy expertise with the broad capabilities of nationally recognized, regional universities, in this case the University of Pittsburgh.
References
1. Kroschwitz, J. I.; Howe-Grant, M., Kirk-Othmer encyclopedia of chemical technology. 4th ed.; Wiley: New York, 1994; Vol. 11.
2. Molina, M. J.; Rowland, F. S., Stratospheric sink for chlorofluoromethanes. Chlorine atom-catalyzed destruction of ozone. Nature 1974, 249, (5460), 810-812.
3. Kalnes, T. N.; James, R. B., Hydrogenation and recycle of organic waste streams. Environmental Progress 1988, 7, (3), 185-191.
4. Bo, Y.; Gang, Y., Reductive Destruction of Environmental Pollutants by Hydrogenation Process. Progress in Chemistry 2009, 21, (1), 217-226.
5. Kovalchuk, V. I.; d'ltri, J. L., Catalytic chemistry of chloro- and chlorofluorocarbon dehalogenation: from macroscopic observations to molecular level understanding. Applied Catalysis A: General 2004, 271, (1-2), 13-25.
6. Fung, S. C.; Sinfelt, J. H., Hydrogenolysis of methyl chloride on metals. Journal of Catalysis 1987, 103, (1), 220-223.
7. Urbano, F. J.; Marinas, J. M., Hydrogenolysis of organohalogen compounds over palladium supported catalysts. Journal of Molecular Catalysis A: Chemical 2001, 173, (1-2), 329-345.
8. Mori, T.; Kikuchi, T.; Kubo, J.; Morikawa, Y., Hydrodechlorination of Trichloromethane to Higher Hydrocarbons over Pd/SiO2 Catalyst. Chemistry Letters 2001, 30, (9), 936-937.
9. Mori, T.; Hirose, K.; Kikuchi, T.; Kubo, J.; Morikawa, Y., Formation of Higher Hydrocarbons from Chloromethanes via Hydrodechlorination over Pd/SiO2 Catalyst. Journal of the Japan Petroleum Institute 2002, 45, (4), 256-259.
10. Golubina, E. V.; Lokteva, E. S.; Lunin, V. V.; Turakulova, A. O.; Simagina, V. I.; Stoyanova, I. V., Modification of the supported palladium catalysts surface during hydrodechlorination of carbon tetrachloride. Applied Catalysis A: General 2003, 241, (1-2), 123-132.
11. Dal Santo, V.; Dossi, C.; Recchia, S.; Colavita, P. E.; Vlaic, G.; Psaro, R., Carbon tetrachloride hydrodechlorination with organometallics-based platinum and palladium catalysts on MgO. Journal of Molecular Catalysis A: Chemical 2002, 182-183, 157-166.
12. Dal Santo, V.; Dossi, C.; Psaro, R.; Recchia, S.; Sordelli, L., Carbon tetrachloride hydrodechlorination with organometallics based platinum and palladium catalysts on MgO: EXAFS characterization and catalytic studies. Studies in Surface Science and Catalysis 2001, 140, 185194.
13. Lokteva, E. S.; Simagina, V. I.; Golubina, E. V.; Stoyanova, I. V.; Lunin, V. V., Formation of Ci-C5 Hydrocarbons from CCl4 in the Presence of Carbon-Supported Palladium Catalysts. Kinetics and Catalysis 2000, 41, (6), 776-781.
14. Lokteva, E. S.; Lunin, V. V.; Golubina, E. V.; Simagina, V. I.; Egorova, M.; Stoyanova, I. V., C-C bond formation during hydrodechlorination of CCl4 on Pd-containing catalysts. Studies in Surface Science and Catalysis 2000, 130C, 1997-2002.
15. Kim, S. Y.; Choi, H. C.; Yang, O. B.; Lee, K. H.; Lee, J. S.; Kim, Y. G., Hydrodechlorination of tetrachloromethane over supported Pt catalysts. Journal of the Chemical Society, Chemical Communications 1995, (21), 2169-2170.
16. Choi, H. C.; Choi, S. H.; Yang, O. B.; Lee, J. S.; Lee, K. H.; Kim, Y. G., Hydrodechlorination of Carbon Tetrachloride over Pt/MgO. Journal of Catalysis 1996, 161, (2), 790-797.
17. Choi, H. C.; Choi, S. H.; Lee, J. S.; Lee, K. H.; Kim, Y. G., Effects of Pt Precursors on Hydrodechlorination of Carbon Tetrachloride over Pt/Al2O3 Journal of Catalysis 1997, 166, (2), 284293.
18. Bae, J. W.; Park, E. D.; Lee, J. S.; Lee, K. H.; Kim, Y. G.; Yeon, S. H.; Hun, S. B., Hydrodechlorination of CCl4 over Pt/y-Al2O3: Effects of reaction pressure and diluent gases on distribution of products and catalyst stability. Applied Catalysis A: General 2001, 217, (1-2), 79-89.
19. Kulkarni, P. P.; Deshmukh, S. S.; Kovalchuk, V. I.; D'Itri, J. L., Hydrodechlorination of dichlorodifluoromethane on carbon-supported group VIII noble metal catalysts. Catalysis Letters 1999, 61, (3-4), 161-166.
20. Kulkarni, P. P.; Kovalchuk, V. I.; d'Itri, J. L., Oligomerization pathways of dichlorodifluoromethane hydrodechlorination catalyzed by activated carbon supported Pt-Cu, Pt-Ag, Pt-Fe, and Pt-Co. Applied Catalysis B: Environmental 2002, 36, (4), 299-309.
21. Chakraborty, D.; Kulkarni, P. P.; Kovalchuk, V. I.; d'Itri, J. L., Dehalogenative oligomerization of dichlorodifluoromethane catalyzed by activated carbon-supported Pt-Cu catalysts: effect of Cu to Pt atomic ratio Catalysis Today 2004, 88, (3-4), 169-181.
22. Storch, H. H.; Golumbic, N.; Anderson, R. B., The Fischer-Tropsch and Related Syntheses. John Wiley & Sons: New York, 1951; p 603.
23. Campbell, J. S.; Kemball, C., Catalytic fission of the carbon-halogen bond. I. Reactions of ethyl chloride and ethyl bromide with hydrogen on evaporated metal films. Transactions of the Faraday Society 1961, 57, 809-820.
24. Anderson, J. R.; McConkey, B. H., Reactions of methyl chloride and of methylene chloride at metal surfaces : II. Reactions over evaporated films of titanium and other metals. Journal of Catalysis 1968, 11, (1), 54-70.
25. Brady, R. C., III; Pettit, R., Reactions of diazomethane on transition-metal surfaces and their relationship to the mechanism of the Fischer-Tropsch reaction. Journal of the American Chemical Society 1980, 102, (19), 6181-6182.
26. Brady, R. C., III; Pettit, R., Mechanism of the Fischer-Tropsch reaction. The chain propagation step. Journal of the American Chemical Society 1981, 103, (5), 1287-1289.
27. Maitlis, P. M.; Quyoum, R.; Long, H. C.; Turner, M. L., Towards a chemical understanding of the Fischer-Tropsch reaction: alkene formation. Applied Catalysis A: General 1999, 186, (1-2), 363-374.
28. González, C. A.; Bartoszek, M.; Martin, A.; De Correa, C. M., Hydrodechlorination of light organochlorinated compounds and their mixtures over Pd/TiO2-washcoated minimonoliths. Industrial and Engineering Chemistry Research 2009, 48, (6), 2826-2835.
29. Anderson, R. B., Schulz-Flory equation. Journal of Catalysis 1978, 55, (1), 114-115.
30. Joyner, R. W., Mechanism of hydrocarbon synthesis from carbon monoxide and hydrogen. Journal of Catalysis 1977, 50, (1), 176-180.
31. Turner, M. L.; Long, H. C.; Shenton, A.; Byers, P. K.; Maitlis, P. M., The Alkenyl Mechanism for Fischer-Tropsch Surface Methylene Polymerisation; the Reactions of Vinylic Probes with CO/H2 over Rhodium Catalyst. Chemistry - A European Journal 1995, 1, (8), 549 - 556.
32. Ma, F.; Sunley, G. J.; Saez, I. M.; Maitlis, P. M., Evidence for vinylic intermediates in the Fischer-Tropsch reaction to give alkenes and alkanes. Journal of the Chemical Society, Chemical Communications 1990, (18), 1279 - 1280.
33. Turner, M. L.; Byers, P. K.; Long, H. C.; Maitlis, P. M., Vinyl initiation of Fischer-Tropsch polymerization over rhodium. Journal of the American Chemical Society 1993, 115, (10), 4417-4418.
34. Gaube, J.; Klein, H.-F., Studies on the reaction mechanism of the Fischer-Tropsch synthesis on iron and cobalt. Journal of Molecular Catalysis A: Chemical 2008 283, (1-2), 60-68.
35. Cheng, J.; Gong, X.-Q.; Hu, P.; Lok, C. M.; Ellis, P.; French, S., A quantitative determination of reaction mechanisms from density functional theory calculations: Fischer-Tropsch synthesis on flat and stepped cobalt surfaces. Journal of Catalysis 2008, 254, (2), 285-295.
36. Ciobica, I. M.; Kramer, G. J.; Ge, Q.; Neurock, M.; van Santen, R. A., Mechanisms for Chain Growth in Fischer-Tropsch Synthesis over Ru(0001). Journal of Catalysis 2002, 212, (2), 136-144.
37. Liu, Z.-P.; Hu, P., A New Insight into Fischer-Tropsch Synthesis. Journal of the American Chemical Society 2002, 124, (39), 11568-11569.
38. Ciobica, I. M.; Frechard, F.; van Santen, R. A.; Kleyn, A. W.; Hafner, J., A theoretical study of CHx chemisorption on the Ru(0001) surface. Chemical Physics Letters 1999, 311, (3-4), 185-192.
39. Zheng, C.; Apeloig, Y.; Hoffmann, R., Bonding and coupling of C1 fragments on metal surfaces. Journal of the American Chemical Society 1988, 110, (3), 749-774.
40. Marsh, A. L.; Becraft, K. A.; Somorjai, G. A., Methane Dissociative Adsorption on the Pt(111) Surface over the 300-500 K Temperature and 1-10 Torr Pressure Ranges. Journal of Physical Chemistry B 2005, 109, (28), 13619-13622.
41. Guczi, L.; Sarma, K. V.; Borko, L., Non-oxidative methane coupling over Co-Pt/NaY bimetallic catalysts. Catalysis Letters 1996, 39, (1-2), 43-47.
42. Koerts, T.; van Santen, R. A., The reaction path for recombination of surface CHx species. Journal of Molecular Catalysis 1991, 70, (1), 119-127.
43. Zaera, F., Reversibility of C1 hydrogenation-dehydrogenation reactions on platinum surfaces under vacuum. Langmuir 1991, 7, (10), 1998-1999.
44. Anderson, J. R., Structure of Metallic Catalysts. Academic Press: London, 1975; p 478.
45. Christmanna, K., Interaction of hydrogen with solid surfaces. Surface Science Reports 1988, 9, (1-3), 1-163.
46. Rodriguez, E.; Leconte, M.; Basset, J.-M.; Tanaka, K., Molecular approach to the mechanisms of C---C bond formation and cleavage on metal surfaces: Hydrogenolysis, homologation, and dimerization of ethylene over Ru/SiO2 catalysts. Journal of Catalysis 1989, 119, (1), 230-237.
47. Long, H. C.; Turner, M. L.; Fornasiero, P.; Ka^par, J.; Graziani, M.; Maitlis, P. M., Vinylic initiation of the Fischer-Tropsch reaction over ruthenium on silica catalysts. Journal of Catalysis 1997, 167, (1), 172-179.
48. Wiersma, A.; van de Sandt, E. J. A. X.; den Hollander, M. A.; van Bekkum, H.; Makkee, M.; Moulijn, J. A., Comparison of the Performance of Activated Carbon-Supported Noble Metal Catalysts in the Hydrogenolysis of CCl2F2. Journal of Catalysis 1998, 177, (1), 29-39.
49. Koerts, T.; van Santen, R. A., Mechanism of carbon—carbon bond formation by transition metals. Journal of Molecular Catalysis 1992, 74, (1-3), 185-191.
50. Birke, P.; Engels, S.; Becker, K.; Neubauer, H. D., Principles of Reactivation of Reforming Catalysts. Chemische Technik (Leipzig) 1979, 31, (9), 473-479.
51. Foger, K.; Jaeger, H., The effect of chlorine treatment on the dispersion of platinum metal particles supported on silica and y-alumina. Journal of Catalysis 1985, 92, (1), 64-78.
52. D'Aniello, M. J., Jr.; Monroe, D. R.; Carr, C. J.; Krueger, M. H., The redispersion of sintered Pt, Rh, and Pt/Rh catalysts. Journal of Catalysis 1988, 109, (2), 407-422.
53. Ciobica, I. M.; van Santen, R. A., A DFT Study of CHx Chemisorption and Transition States for C-H Activation on the Ru(1120) Surface. Journal of Physical Chemistry B 2002, 106, (24), 6200-6205.
54. Ciobica, I. M.; Frechard, F.; van Santen, R. A.; Kleyn, A. W.; Hafner, J., A DFT Study of Transition States for C-H Activation on the Ru(0001) Surface. Journal of Physical Chemistry B 2000, 104, (14), 3364-3369.
55. Paul, J.-F.; Sautet, P., Chemisorption and Transformation of CHx Fragments (x = 0-3) on a Pd(111) Surface: A Periodic Density Functional Study. Journal of Physical Chemistry B 1998, 102, (9), 1578-1585.
56. Bent, B. E., Mimicking Aspects of Heterogeneous Catalysis: Generating, Isolating, and Reacting Proposed Surface Intermediates on Single Crystals in Vacuum. Chemical Reviews 1996, 96, (4), 1361-1390.
57. Radhakrishnan, G.; Stenzel, W.; Hemmen, R.; H. Conrad; Bradshaw, A. M., The photon-induced reactions of chemisorbed CH3Br on Pt{111}. 1991, 95, (6), 3930-3938.
58. Fairbrother, D. H.; Peng, X. D.; Viswanathan, R.; Staira, P. C.; Trenary, M.; Fan, J., Carboncarbon coupling of methyl groups on Pt(111). Surface Science 1993, 285, (1-2), L455-L460.
59. Solymosi, F., Thermal stability and reactions of CH2, CH3 and C2H5 species on the metal surfaces. Catalysis Today 1996, 28, (3), 193-203.
Олигомеризация полихлорированных метанов в С2+ углеводороды
в присутствии Pt-Co/С катализаторов
В.И. Ковальчук% В.Д. Роудс% М.А. Макдональд®
a Департамент химической и нефтехимической технологии, Питсбургский университет, Питсбург, Пенсильвания 15261, США б Национальная лаборатория энергетических технологий, Государственный департамент энергии, Питсбург,
Пенсильвания 15236, США
Изучено дегидрохлорирование дихлорметана, трихлорметана и их смесей в присутствии Pt-Co/C катализатора с целью выяснения механизма образования высших углеводородов. Установлено, что в реакции дихлорметана с водородом при 523 К катализатор не подвергается дезактивации и сохраняет постоянную активность в течение, по крайней мере, 18 ч, в то время как присутствие трихлорметана в реакционной смеси вызывает значительную дезактивацию катализатора в течение первых 5 ч работы. Углеводородные продукты наблюдались во всех реакционных смесях с селективностью дихлорметан + водород < трихлорметан + водород < дихлорметан + трихлорметан + водород. При этом селективность по этану и пропану не зависела от состава реакционной смеси. В то же время селективность по этилену и пропилену была существенно выше в случае дехлорирования трихлорметана и дехлорирования смеси дихлорметана и трихлорметана, чем в случае дехлорирования дихлорметана. На основании полученных результатов сделано заключение, что насыщенные углеводороды образуются в соответствии с так называемым алкильным механизмом роста цепи, в то время как непредельные углеводороды образуются по алкенильному механизму.
Ключевые слова: дегидрохлорирование, хлорметаны, восстановительная олигомеризация, синтез Фишера-Тропша, реакционный механизм, платина, кобальт.
