Sage Journals: Discover world-class research

Abstract

Ethylene is recovered from ethylene/ethane mixture by adsorptive separation on CuCl₂ loaded activated carbon. When 36.7% of CuCl₂ were dispersed on activated carbon made from coconut shell (CSC) and calcined at 400℃ for some hours, a satisfactory adsorbent sample was prepared. The adsorption capacity of ethylene reached 23.55 mg/g and ethylene/ethane separation factor reached 7.6. Adsorbent characterization by various modern instruments demonstrated the reduction of CuCl₂ to provide Cu⁺ cation on CSC, which is responsible to form π-complexation bonds for the adsorption of ethylene. Mechanism was discussed. Nature of related active centers was suggested.

Keywords

Ethylene π-complexation adsorbent cuprous chloride (CuCl)

Introduction

Ethylene is one of the most important raw materials for the production of many petrochemicals, including plastics, rubbers and textiles. Therefore, its demand is rapidly increasing in the latest decades. Ethylene is now produced mainly by steam cracking of naphtha from crude oil or decomposition of ethane from natural gas. In order to save these raw materials, recovery of ethylene from dry gas of fluid catalytic cracking (FCC) and delayed coking units in a refinery is also under consideration. In all of the above technologies, ethylene and ethane separation play an important role in the preparation of polymer-grade ethylene. Cryogenic distillation is known to be a conventional and reliable measure for this separation. However, it is also energy-intensive due to the close relative volatilities between ethylene and ethane. A number of potential alternatives have, therefore, been investigated, among which adsorptive separation is commonly considered to be one of the most potential approaches. Zeolite has specialized regular channel dimension and polarity, and can be also modified by ion exchange, therefore, it is often considered to be an ideal adsorbent for the separation of ethylene and ethane as reported in the literature (Meng et al., 2010; Mofarahi and Salehi, 2013; Narin et al., 2014; Triebe et al., 1996). Recently, a zeolite imidazolate framework, which is very close to the structure of zeolite, has been used as an adsorbent for this purpose, wherein a gate-opening mechanism is suggested (Ying et al., 2014). Additionally, a metal organic framework (MOF) has also been reported (Martins et al., 2015). However, the most promising alternative is the adsorptive separation based on π-complexation (Choudary et al., 2002; Yang and Kikkinides, 1995). It is reported that π-complexes are formed between the double bond of ethylene and some Cu⁺ or Ag⁺ cation containing transition metal salts. The π-complexation bonds are stronger than Van der Waals forces formed by physical adsorption, so π-complexation adsorption can achieve high selectivity and high adsorptive capacity for ethylene. On the other side, those bonds are still so weak, that ethylene can be desorbed by raising the temperature or decreasing the pressure. Related chemical mechanism and related conditions for this π-complexation reaction have been discussed in the literature. Cuprous chloride (CuCl), which contains Cu⁺ cation(Cu(I)), is recommended to be the first choice (Huang et al., 1999; Van Miltenburg et al., 2006) in this work, because silver is more expensive and Ag⁺ is often apt to be reduced to silver metal. This metal salt is always dispersed over a support with a large surface area to provide more adsorption sites (Aguado et al., 2012). Several kinds of supports have been examined (Padin and Yang, 2000; Rege et al., 1998; Yang et al., 2002), but zeolite materials are also often considered to have the excellent performance and have been widely applied to support cuprous chloride (CuCl) (Van Miltenburg, 2007; Xie et al., 1990) to prepare an adsorbent for the separation of ethylene.

However, cuprous chloride is not soluble in water and it must be dissolved in a concentrate hydrochloride solution, which requires a special tank to prepare CuCl solutions to solve corrosion problems. On the contrary, cupric chloride (CuCl₂) is thoroughly soluble in water, while it is not able to produce π-complexation bonds with ethylene. Therefore, the adsorbent can be prepared by dispersing CuCl₂ water solution on the zeolite support first and then converting it to CuCl by succeeding reduction of CuCl₂ particles in hydrogen or carbon monoxide to provide Cu(I) (Xie et al., 1990).

Activated carbon (AC) is one of the conventional adsorbent and has been widely industrially applied. For example, it has been successfully applied to FCC dry gas separation to remove C₃ and heavier compounds to prepare ethylene/ethane starting materials for this work. However AC itself is not able to produce high purity ethylene. Because of the porous structure and various fundamental groups on the AC surface, it has been applied as a support for the dispersion of transition metal compounds to prepare an adsorbent for ethylene and ethane separation via complexation. Initially CuCl is still used to prepare CuCl/AC for this purpose (Mei et al., 2002). If CuCl₂ must be used instead of CuCl, a succeeding reduction step is considered to be necessary (Golden et al., 1992). More recently, it is observed that CuCl₂ can be reduced to CuCl by AC itself in the preparation steps of this adsorbent as exemplified in a paper (Zhang et al., 2008). Therefore, the succeeding reduction step to convert CuCl₂ into CuCl by hydrogen stream can be cancelled. As we have known, the reduction behaviors and mechanism for CuCl₂ to CuCl by AC support have not yet been studied. This work aims to highlight the effect of important preparation parameters on the dispersion and reduction results of cupper species on CSC support. Reduction mechanism and related active centers will be also addressed. This information would pave the way to prepare more effective CuCl₂/AC adsorbent for the recovery of ethylene from ethylene/ethane mixture.

Experimental

Preparation of the adsorbent

AC has varying pore dimensions. Adsorption is performed mainly in micropores, while the contribution by mesopores is rather small. Very few macropores is also formed in AC and they only play the role of passageway. All of these three pores link up each other. In addition, different oxygen-containing functional groups are exhibited on the surface of AC, including carboxylic group, lactonized carboxylic group, phenolic hydroxyl groups and carbonyl groups (Boehm, 2002; Lahaye, 1998). Modification of the above properties of AC by using hydrogen peroxide or nitric oxide has often been used in the literature (Gil et al., 1997).

In our previous work, it has been recognized that adsorption behaviors of CuCl₂/AC adsorbent are firstly determined by the source of the starting materials to produce that AC. In addition, AC made from coconut shell (CSC) support provides substantial advantages over other AC that is in line with the reported results (Zhang et al., 2008). To clarify the effect of reduction extent and related mechanism, original CSC without any chemical treatment is selected to be a support for CuCl₂ aqueous solution in this paper.

In this work, CSC was purchased from Shanghai activated company. It was pulverized and sieved to ϕ0.2–0.8 mm particles, then dried at 100℃ for 2 h before taken as CSC support. The length of the ϕ10 mm tube column is about 300 mm, therefore, it is possible to be a plug flow in the adsorption tube. CuCl₂/AC was prepared by using incipient wetness technique, which is reported to permit effective dispersion of a monomolecular layer of metal compounds and avoid from their decomposition (Yang et al., 2002; Xie et al., 1998). A desired concentration aqueous solution of CuCl₂ was prepared. A fixed volume of pre-dried CSC support was then impregnated in the solution at a fixed temperature for 8 h. After that, the solid product was dried at 110℃ for 6 h, and then it was purged with nitrogen stream at a flow rate of 60 mL/min in a tubular oven for 30 min under room temperature to thoroughly remove the oxygen therein. Afterwards, the tubular oven is temperature programmed heated at a ramp of 5℃/min to a fixed calcined temperature. After keeping for 4 h, the product was cooled down to room temperature.

Adsorption behaviors

To simulate the composition of the potential feed gas, an ethylene and ethane mixture in nitrogen (28:60:12 kPa) was used as a typical feedstock for the separation tests by adsorption in this work. Adsorbed amount and separation factor were measured in a conventional self-made fixed bed adsorption apparatus by using dynamic gas chromatograph method. It is composed of feed gas and adsorption system, which is connected to a thermal conductivity detector (TCD) of a gas chromatography (GC) system. A U-type adsorption tube (ϕ10 mm × 300 mm) is put in an electric thermostat to keep the temperature constant. Each of the adsorption or desorption step was preformed therein. Gas composition was analyzed by a GC9160 GC equipped with a Φ3 mm × 4 m GDX-502 packed column (copolymer of divinylbenzene and nitrogen containing heterocyclic monomers) and a TCD.

The adsorption capacity q_k was calculated from the related breakthrough curves as follows:

q_{k} = \frac{\int_{0}^{t} Q (y_{k, i} - y_{k, e}) \cdot P / RT d t - ɛ \cdot V \cdot P / RT}{m}

where Q is flow rate of the mixture, y_k,i is inlet concentration, y_k,e is exit concentration, ɛ is column porosity, V is column volume, m is the mass of absorbent.

Separation factor S for ethylene and ethane was also calculated from the related breakthrough curves as follows:

S = (x_{1} / y_{1}) / (x_{2} / y_{2})

where x₁ and x₂ are the concentration of ethylene and ethane in the adsorption phase at equilibrium, while y₁ and y₂ is that in the gas phase also at equilibrium, respectively.

In our tests, three times of measurements were performed to obtain each amount of adsorbed ethylene and separation factor and their average results were used. The results have good reproducibility, and the relative standard deviation is less than 5%.

Characterization

Surface area and pore structure were measured on an ASAP 2020 surface area and porosity analyzer. Surface area was measured by using BET method, while the pore volume of the micropores was calculated by t-plot method (Lippens and De Boer, 1965). XRD patterns were recorded on an X-ray diffractometer D/max2550V. Temperature programmed reduction (H₂-TPR) behaviors of the adsorbent was measured on a Micromeritics AutoChem II 2920 V4.02 set. H₂-TPR profiles are used for characterization of changes in valence of metal cupper species. In this test, 70 mg of dried samples were loaded to a U-type tube and pretreated in a high purity nitrogen flow at 200℃ for 30 min to remove the adsorbed water, and then dropped to 50℃. After the base line is stable, the tube is switched to a 10/90 of H₂/Ar gas line with a flow rate of 40 mL/min and was temperature programmed heated with a ramp of 10℃/min to 750℃. TCD was used to measure the signal for consumed hydrogen. Changes in functional groups were detected on a Nicolet 6700 FTIR spectrometer in the range of 4000–500 cm⁻¹ with a resolution of 4 cm⁻¹. The adsorbent sample was prepared as a KBr pellet, wherein AC:KBr was 1:400. Valence changes and surface element analysis of the cupper species on loaded AC were detected by X-ray photoemission spectra (XPS) (Padin and Yang, 2000), obtained on ESCAL-ab 220i-XL, produced by VG Scientific Inc. England. Measurement of the surface elements and their states on the absorbent sample were performed by using energy dispersion X-ray (EDX) analyzer from EDAX Inc., attached to S-3400N vacuum electric microscope, produced by Hitachi, Japan.

Results and discussion

Both CuCl₂ content on CSC support and the calcinations temperature are the important parameters in the preparation of CuCl₂/CSC adsorbent. At the same time, CuCl₂ must be reduced to CuCl to provide Cu⁺ cations (Cu(I)) and well dispersed on CSC surface for the separation of ethylene via complexation. They will all be studied as follows:

Representative breakthrough curve

Adsorption behaviors are confirmed with adsorption breakthrough tests. A 17.44 g of a 36.7% CuCl₂/CSC adsorbent sample is loaded, and the column porosity is 0.56. The feed gas is a mixture of ethylene/ethane/nitrogen (28:60:12) with a rate of 40 mL/min. The measurement is performed at 0.1 MPa and 313 K.

In Figure 1, it is seen that both ethylene and ethane are entirely adsorbed in the first few minutes, but ethane breaks through the tube column after 8 min. During the first 25 min, the adsorbent retains ethylene and the outlet stream is free from ethylene, while it breaks through afterwards. At last, the exit gas concentration returns back to that of feed gas after 50 min, demonstrating the saturation of that adsorbent sample. The amount adsorbed of ethylene is obtained by integration of the related breakthrough curve and the separation factor is calculated by the formula, as described in the above section (Aguado et al., 2012; Huang et al., 1999).

Figure 1.

Breakthrough curve for a mixture of ethylene/ethane/nitrogen (28:60:12) on 36.7% CuCl₂/CSC adsorbent.

Effect of loaded CuCl₂ contents

Effect of loaded CuCl₂ amount on C₂H₄/C₂H₆ adsorptive separation behaviors on CuCl₂/CSC adsorbents is examined. It is shown in Figure 2 that the adsorbed amount of ethylene is lower than that of ethane on the pure CSC support, resulting from the inherent physical adsorption of CSC itself. With the increase of loaded CuCl₂ content on CSC, the adsorbed amount of ethane sharply drops. On the other hand, the adsorbed amount of ethylene raises up initially, reaching the maximum at 36.7% and then gradually drops down. This decrease is probably because of their threshold limit and exceeding amount of CuCl₂ would plug the pores on the surface. These results are in line with the increasing adsorption separation factors, as seen in Figure 2.

Figure 2.

Effect of loaded CuCl₂ amount on ethylene/ethane adsorptive separation behaviors (calcination temperature: 673 K, adsorption conditions: P = 0.1 MPa, feed flow = 40 mL/min, T = 313 K).

The separation behaviors of CuCl₂/CSC adsorbents relate to their physico-chemical characteristics. Firstly, the relation of pore structures of CuCl₂/CSC with varying CuCl₂ content is listed in Table 1. It is shown that both specific surface area and pore volume rapidly decrease with increasing loaded CuCl₂ amount. At CuCl₂ content of 36.7%, the specific surface area and the pore volume of the micropores drop to 4.6% and 4.4% of that of the pure CSC, respectively. It must result in a significant decrease in the physical adsorption amount on these CuCl₂/CSC adsorbents. Therefore, adsorptive ability for ethane rapidly drops in Figure 2. However, the adsorbed ethylene amount on the adsorbent increased instead, demonstrating the prominent effect of adsorption via complexation up to 36.7% CuCl₂/CSC adsorbent. Secondly, the separation behaviors relate to reduction and dispersion results of the metal cations on the CSC support. XRD patterns of CuCl₂ loaded CSC adsorbents are shown in Figure 3. Two wider peaks on pattern (a) for pure CSC support are observed near value of 2θ of 24° and 44°, respectively. They are attributed to specialized peaks of the graphite microcrystals, demonstrating the disordered layer structure of graphite on pure CSC support (JCPDS-ICDD, 1995). When CuCl₂ is loaded on CSC and calcined at 400℃ in nitrogen stream, the peaks for graphite microcrystal significantly weaken with increasing amount of loaded CuCl₂, that is in line with the change of the microstructure. At loaded CuCl₂ content lower than 44%, no new specialized peak appears except that for graphite microcrystal of CSC itself, demonstrating good dispersion of the loaded metal salt. When the CuCl₂ content in the adsorbent reaches 55%, new specialized peaks appeared at 2θ of 28.520°, 47.434°, 56.287°, attributed to peaks for CuCl crystal faces of 111, 220, and 311, respectively. These results reveal the reduction from Cu(II) into Cu(I), but dispersion of the latter sharply declines. Therefore, peaks attributed to CuCl clearly appear on the XRD pattern of 55% CuCl₂/CSC, resulting in lower adsorptive ability as seen in Figure 2. Meanwhile, reduction extent of Cu(II) on 44% is not explained. It will be discussed in “Extent of reduction from Cu(II) to Cu(I) on 36.7%CuCl₂/CSC adsorbent” section.

Figure 3.

XRD patterns of CuCl₂/CSC adsorbents with varying CuCl₂ content (calcination temperature: 673 K, a: CSC, b: 15%CuCl₂/CSC, c: 25%CuCl₂/CSC, d: 3–36.7%CuCl₂/CSC, e: 44%CuCl₂/CSC, f: 55%CuCl₂/CSC).

Table 1.

Pore structural parameters of CuCl₂/CSC adsorbents.

Adsorbents	S_BET (m²·g⁻¹)	S_micro (m²·g⁻¹)	V (cm³·g⁻¹)	V_mirco (cm³·g⁻¹)	Average pore diameter (nm)
CSC	886.3	731.4	0.406	0.315	1.83
15%CuCl₂/CSC	491.3	374.3	0.231	0.160	1.88
36.7%CuCl₂/CSC	47.3	33.9	0.024	0.014	1.50
44%CuCl₂/CSC	34.4	13.3	0.020	0.005	2.41
55%CuCl₂/CSC	15.7	2.9	0.011	0.001	2.89

Effect of calcination temperature

Calcination is an important step in the preparation of modified AC adsorbent. In this step, changes in crystal particles and reduction extent of loaded cupper species have a strong influence on their dispersion and surface area.

In Figure 4, it is seen that at calcination temperature of 200℃, the adsorbed amount of ethylene on 36.7%CuCl₂/CSC adsorbent is only 2.8 mg/g with a C₂H₄/C₂H₆ separation factor of 2.1. When the adsorbent is calcined at 400℃, the adsorptive capacity reaches an optimum of 23.6 mg/g with a separation factor of 7.6. On the other side, they dropped to 19 mg/g and 6.2, respectively, at a temperature of 500℃.

Figure 4.

Effect of calcinations temperature in the preparation of 36.7%CuCl₂/CSC on ethylene/ethane separation behaviors (adsorption conditions: P = 0.1 MPa, feed flow = 40 mL/min, T = 313 K).

Effect of calcinations temperature on this adsorbent also relates to the physico-chemical characteristics. Firstly, it is seen in Table 2 that the surface area and pore volume of micropores of the sample, calcined at 200℃ are very small, demonstrating serious plugging of adsorbent channels and resultant bad dispersion of the metal salt. Of course, they must lead to a poor adsorptive ability. They are improved with increasing calcinations temperature. Higher calcination temperature, ranging from 300℃ to 500℃, is benefit to the diffusion and dispersion of the metallic compounds. However, the adsorbent ability reaches an optimum at 400℃. The reason why the adsorptive capacity is the best at a critical temperature and declines at 500℃ has not been further examined.

Table 2.

Pore structures of 36.7%CuCl₂/CSC adsorbents, calcined at different temperatures.

Adsorbents	S_BET (m²·g⁻¹)	S_micro (m²·g⁻¹)	V (cm³·g⁻¹)	V_mirco (cm³·g⁻¹)	Average pore diameter (nm)
36.7%CuCl₂/CSC-200℃	9.7	0.19	0.0100	≈0	4.13
36.7%CuCl₂/CSC-300℃	26.2	10.7	0.0144	0.004	2.20
36.7%CuCl₂/CSC-400℃	47.3	33.9	0.0240	0.014	1.50
36.7%CuCl₂/CSC-500℃	69.5	58.4	0.0385	0.032	2.21

Secondly, separation behaviors also relate to the changes in the extent of reduction and dispersion, which is drawn from characterization by XRD patterns. As shown in Figure 5, three peaks of uncalcined 36.7% CuCl₂/CSC appear at 2θ of 15.4°, 25.9° and 30.7°, respectively, attributing to the diffraction peaks of CuCl₂ crystals. For this adsorbent calcined at 200℃ under nitrogen stream, a weak peak at 2θ of 15.4° attributing to CuCl₂ is still observed. It indicates that part of CuCl₂ has not yet been reduced and has not well dispersed. On the other hand, no other peak except that for active carbon is observed for that sample, calcinated at 400℃ and 500℃. It is therefore concluded that active components can be well dispersed or no sintering of the metallic salt on CSC support is observed at calcination temperature of 400℃ and 500℃. In our work, calcination temperature is selected to be 400℃ to obtain the optimal separation behaviors.

Figure 5.

XRD patterns of 36.7%CuCl₂/CSC prepared at varying calcination temperatures.

From the results in this section, it is concluded that the surface area and pore volume of adsorbent decrease with increasing CuCl₂ content on CuCl₂/CSC calcined at 400℃, while both of them increases with increasing calcination temperature. From XRD patterns in Figure 5, it is seen that calcination temperature must be above 400℃ to convert Cu(II) into Cu(I) for each 36.7%CuCl₂/CSC. If CuCl₂ content is higher than 36.7%, dispersion of Cu(I) will be deteriorated as shown in Figure 3 and that is in line with the decrease in adsorption ability as shown in Figure 2. A 36.7%CuCl₂/CSC sample calcined at 400℃ has satisfactory dispersion, but the extent of reduction cannot be estimated from XRD patterns in Figure 3. It will be measured and discussed in the following section.

Extent of reduction from Cu(II) to Cu(I) on 36.7%CuCl₂/CSC adsorbent

The main component to form π complex adsorption is only Cu(I) in CuCl. The selective adsorption behaviors of CuCl₂/CSC adsorbent is related to the extent of the reduction of CuCl₂ to CuCl.

Results of surface elementary analysis of the adsorbent by XPS measurement are shown in Table 3. It is seen that the elements on CuCl₂/CSC adsorbent surface contain C, O, Cl, and Cu (Biniak et al., 1997; Kozlowski and Sherwood, 1986). Although Cl and Cu atomic ratio reaches about 2.2 on that adsorbent, calcinated at 200℃, the ratio drops to 0.96 or near 1 on that calcinated at 400℃. It demonstrates that Cu(II) on the adsorbent calcinated at 400℃ has entirely been reduced to Cu(I), so that gives satisfactory separation behaviors. It is in line with the result in Figure 4. Additionally, SEM-EDS measurement leads to the similar conclusion, because Cu and Cl atomic ratio also near 1 as listed in Table 4.

Table 3.

Some surface element analysis of 36.7%CuCl₂/CSC adsorbents calcined at different temperatures.

Adsorbents	Atomic percentage (%)				Mole ratio
Adsorbents	Cl	C	O	Cu	Cl/Cu
36.7%CuCl₂/CSC-200℃	10.72	76.31	8.09	4.88	2.2
36.7%CuCl₂/CSC-400℃	10.31	63.65	15.28	10.76	0.96

Table 4.

Main element analysis of CSC and 36.7%CuCl₂/CSC by EDS.^a

Sample	Parameter	C	O	Mg	Al	Si	Cl	K	Cu
CSC	Wt%	86.35	10.35	0.27	0.43	0.71	0	0	0
CSC	Atom%	89.05	7.44	0.22	0.16	0.40	0	0	0
36.7%CuCl₂/CSC	Wt%	34.3	5.64	0.09	0.00	0.13	24.0	0.51	35.3
36.7%CuCl₂/CSC	Atom%	64.3	7.70	0.07	0.00	0.10	15.2	0.28	12.3

Calcination temperature: 673 K.

Mechanism of the reduction from Cu(II) to Cu(I)

TPR is one of the effective method to study the valence of cupper metal species. Changes in valence of cupper species in reduction step from Cu(II)→Cu(I)→Cu(0) were monitored by using the related peaks, appeared at varying temperatures on the TPR profiles with hydrogen stream. By comparison with the results in the literature, the peak appeared at 320℃ in Figure 6 is attributed to the reduction reaction of Cu(I)→Cu(0) on the oxygen-containing functional groups of CSC, while that appeared at 420℃ is attributed to the reduction reaction of Cu(I)→Cu(0) on the carbon on CSC. Moreover, one more reduction peak appears at 220℃ on the TPR profile for the uncalcined CuCl₂/CSC sample. This peak is attributed to reduction of CuCl₂. These results also indicate that CuCl₂ has thoroughly converted to CuCl on the CuCl₂/CSC adsorbent after calcination at 400℃, because that peak at 220℃ has thoroughly disappeared.

Figure 6.

H₂-TPR profiles of CuCl₂/CSC adsorbents (calcination temperature: 673 K).

To further elucidate the mechanism for the reduction of Cu(II) to Cu(I) on CSC support. FTIR pattern measurement is performed for pure CSC and CuCl₂ /CSC samples as shown in Figure 7. As reported in the literature (Biniak et al., 1997), the peak at 1000–1200 cm⁻¹ is attributed to stretching vibration of phenolic hydroxyl groups, while the other one at 3000 cm⁻¹ is attributed to that of C–H groups on the poly-aromatic rings. It is observed in Figure 6 that both of these two peaks drop with increasing CuCl₂ content on CuCl₂/CSC adsorbents. It reveals that these two groups play the role in the reduction of Cu(II) to Cu(I) on CSC support and demonstrates the existence of two reduction active centers. This result is in line with the measurement by TPR, where two active centers are observed. This reaction is postulated to partially proceed from chlorine substitution reactions between CuCl₂ and para-position hydrogen atoms of C–H bonds on CSC poly-aromatic rings as follows:

Figure 7.

FTIR spectra of CSC and CuCl₂/CSC adsorbents (calcination temperature: 673 K).

This suggestion is proved by the fact that hydrochloride gas has been detected in the exit gas during calcination in our test.

Conclusions

CuCl₂/CSC adsorbent is effective for ethylene and ethane separation via π-complexation, because Cu(II) has thoroughly been reduced to Cu(I) by CSC support in the preparation stages in this work. CuCl₂ is soluble in water, while CuCl can only be dissolved in concentric hydrochloride solution. Therefore the corrosion problem resulting from the preparation of CuCl solution has been avoided.

Separation factor for the separation of ethylene and ethane on the CuCl₂/CSC adsorbent enhances with increasing CuCl₂ content thereon. Meanwhile, a maximum adsorbed amount of ethylene appears at CuCl₂ content of 36.7%. It is clear that π-complexation plays the predominant role.

Calcination temperature is another one of important preparation parameters. When it is performed at 200℃, CuCl₂ is not well dispersed on CSC support and the channels are seriously plugged that would give negative effect on the separation behaviors. At calcinations temperature of 400℃, reduction of Cu(II) to Cu(I) is proved to have been performed from our tests. They are well dispersed on CSC support and their separation ability is therefore apparently improved. However, it slowly drops at calcination temperature of 500℃.

It is postulated that there are two reduction active centers on CSC support. One is phenol-hydroxyl groups and the other is CH– groups connected to poly-aromatic rings.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Aguado

Bergeret

Daniel

(2012) Absolute molecular sieve separation of ethylene/ethane mixtures with silver zeolite A. Journal of the American Chemical Society 134(36): 14635–14637.

Biniak

Szymański

Siedlewski

(1997) The characterization of activated carbons with oxygen and nitrogen surface groups. Carbon 35(12): 1799–1810.

Boehm

(2002) Surface oxides on carbon and their analysis: a critical assessment. Carbon 40(2): 145–149.

Choudary

Kumar

Bhat

TSG

(2002) Adsorption of light hydrocarbon gases on alkene-selective adsorbent. Industrial & Engineering Chemistry Research 41(11): 2728–2734.

Gil

de La Puente

Grange

(1997) Evidence of textural modifications of an activated carbon on liquid-phase oxidation treatments. Microporous Materials 12(1): 51–61.

Golden TC, Kratz WC and Wilhelm FC (1992) Selective adsorbent for carbon monoxide or ethylene. Patent 5,126,310, USA.

Huang

Padin

Yang

(1999) Comparison of π-complexations of ethylene and carbon monoxide with Cu⁺ and Ag⁺. Industrial & Engineering Chemistry Research 38(7): 2720–2725.

JCPDS-ICDD (1995) JCPDS data set. USA: International centre for diffraction data.

Kozlowski

Sherwood

PMA

(1986) X-ray photoelectron spectroscopic studies of carbon fibre surfaces vii-electrochemical treatment in ammonium salt electrolytes. Carbon 24(3): 357–363.

10.

Lahaye

(1998) The chemistry of carbon surfaces. Fuel 77(6): 543–547.

11.

Lippens

De Boer

(1965) Studies on pore systems in catalysts: V. The t method. Journal of Catalysis 4(3): 319–323.

12.

Martins

VFD

Ribeiro

Ferreira

(2015) Ethane/ethylene separation on a copper benzene-1,3,5-tricarboxylate MOF. Separation and Purification Technology 149: 445–456.

13.

Mei

Liu

(2002) Study of activated carbon supported CuCl for ethylene/ethane separation by adsorption: Effects of oxidative treatment. New Carbon Materials 17(4): 33–37.

14.

Meng

Lin

CCH

Kuznicki

(2010) Separation of a binary mixture of ethylene and ethane by adsorption on Na-ETS-10. Chemical Engineering Science 65(11): 3494–3498.

15.

Mofarahi

Salehi

(2013) Pure and binary adsorption isotherms of ethylene and ethane on zeolite 5A. Adsorption 19(1): 101–110.

16.

Narin

Martins

VFD

Campo

(2014) Light olefins/paraffins separation with 13X zeolite binderless beads. Separation and Purification Technology 133: 452–475.

17.

Padin

Yang

(2000) New sorbents for olefin/paraffin separations by adsorption via π-complexation: synthesis and effects of substrates. Chemical Engineering Science 55(14): 2607–2616.

18.

Rege

Padin

Yang

(1998) Olefin/paraffin separations by adsorption: π-complexation vs. kinetic separation. AIChE Journal 44(4): 799–809.

19.

Triebe

Tezel

Khulbe

(1996) Adsorption of methane, ethane and ethylene on molecular sieve zeolites. Gas Separation & Purification 10(95): 81–84.

20.

Van Miltenburg A (2007) Adsorptive separation of light olefin/paraffin mixtures: dispersion of CuCl in Faujasite zeolites. TU Delft, Delft University of Technology.

21.

Van Miltenburg

Zhu

Kapteijn

(2006) Adsorptive separation of light olefin/paraffin mixtures. Chemical Engineering Research and Design 84(5): 350–354.

22.

Xie Y, Bu N, Liu J, et al. (1990) Adsorbents for use in the separation of carbon monoxide and/or unsaturated hydrocarbons from mixed gases. Patent 4,917,711.

23.

Xie

Zhu

Zhao

(1998) An important principle for catalyst preparation-spontaneous monolayer dispersion of solid compounds onto surfaces of supports. Studies in Surface Science and Catalysis 118: 441–449.

24.

Yang

Kikkinides

(1995) New sorbents for olefin/paraffin separations by adsorption via π-complexation. AIChE Journal 41(3): 509–517.

25.

Yang RT, Padin J and Rege SU (2002) Selective adsorption of alkenes using supported metal compounds. Patent 6,423,881, USA.

26.

Ying

Chen

Liu

(2014) Adsorption and separation of ethane/ethylene on ZIFs with various topologies: Combining GCMC simulation with the ideal adsorbed solution theory (IAST). Chemical Engineering Science 124: 144–153.

27.

Zhang

Zhao

Wang

(2008) Effect of CuCl₂ loading on CuCl₂-La(NO₃)₃/AC adsorbents for ethylene separation from the mixed-gas of ethylene/ethane. Natural Gas Chemical Industry 33(6): 27–32.

Adsorptive recovery of ethylene by CuCl 2 loaded activated carbon via π-complexation

Abstract

Keywords

Introduction

Experimental

Preparation of the adsorbent

Adsorption behaviors

Characterization

Results and discussion

Representative breakthrough curve

Effect of loaded CuCl2 contents

Effect of calcination temperature

Extent of reduction from Cu(II) to Cu(I) on 36.7%CuCl2/CSC adsorbent

Mechanism of the reduction from Cu(II) to Cu(I)

Conclusions

Footnotes

Declaration of Conflicting Interests

Funding

References

Effect of loaded CuCl₂ contents

Extent of reduction from Cu(II) to Cu(I) on 36.7%CuCl₂/CSC adsorbent