Intensified Ethylene Production via Chemical Looping through an Exergetically Efficient Redox Scheme

Summary

Ethylene production via steam cracking of ethane and naphtha is one of the most energy and emission-intensive processes in the chemical industry. High operating temperatures, significant reaction endothermicity, and complex separations create hefty energy demands and result in substantial CO₂ and NO_x emissions. Meanwhile, decades of optimization have led to a thermally efficient, near-“perfect” process with ∼95% first law energy efficiency, leaving little room for further reduction in energy consumption and CO₂ emissions. In this study, we demonstrate a transformational chemical looping–oxidative dehydrogenation (CL-ODH) process that offers 60%–87% emission reduction through exergy optimization. Through detailed exergy analyses, we show that CL-ODH leads to exergy savings of up to 58% in the upstream reactors and 26% in downstream separations. The feasibility of CL-ODH is supported by a robust redox catalyst that demonstrates stable activity and selectivity for over 1,400 redox cycles in a laboratory-scale fluidized bed reactor.

Graphical Abstract

Highlights

• •Chemical-looping oxidative dehydrogenation intensifies ethylene production
• •Detailed process analysis shows ∼87% emission reduction versus steam cracking
• •Reduction in exergy losses leads to substantial energy savings
• •Superior performance of a robust redox catalyst is demonstrated

Chemical Engineering; Industrial Chemistry; Chemical Reaction Engineering

Introduction

Carbon capture and utilization has received significant attention for reducing anthropogenic emissions (E. Boot-Handford et al., 2014, Figueroa et al., 2008, Li et al., 2013, Tola and Pettinau, 2014, Volkart et al., 2013). However, most research activities in this area have focused on reducing CO₂ emissions from power and heat generation. This is understandable, since this sector alone accounts for approximately 31% of the 32 billion tons (CO₂ equivalent/year) of greenhouse gases emitted worldwide (Global Emissions, 2017). Meanwhile, the manufacturing sector, contributing 21% of the global greenhouse gas emissions (Friedrich, 2019), is often overlooked. For example, ethylene production via steam cracking releases over 1 ton of CO₂ per ton of olefin produced (Tao Ren, 2006). With a worldwide demand in excess of 150 million tons/year (Ethylene Global Supply Demand Analytics Service, 2018), ethylene production is one of the largest contributors to greenhouse gas emission in the manufacturing sector. Despite the high energy and carbon intensities associated with steam cracking, the operation is widely believed to be near optimal considering the continued research and improvements on the technology over the past 80+ years (Andrews and Pollock, 1959, Brown et al., 1983, Griffin and Moon, 1964, Heynderickx et al., 2001, Ludwig, 1951, Masaaki and Masaaki, 1967, Plehiers et al., 1990, Ranjan et al., 2012, Ruckaert et al., 1978, Sato and Ohnishi, 1971, W, 1963). Thermal efficiencies of up to 95% have been achieved through optimized furnace design and heat/steam integration (Zimmermann and Walzl, 2000). Although conventional wisdom would have indicated little room for further efficiency improvements in ethylene production, we recently reported a chemical looping–oxidative dehydrogenation (CL-ODH) scheme showing the potential to significantly reduce the energy consumption and CO₂ emissions compared with steam cracking (Haribal et al., 2017). In CL-ODH, as the name suggests, ethane is oxidatively dehydrogenated to ethylene and water, using the lattice oxygen of a chemical looping catalyst (reduction of the catalyst via Reaction 1a). Regeneration of this catalyst in air (Reaction 1b) provides the heat for ethylene formation.

CL-ODH (Reduction)

C₂H₆ + Me_yO_x → C₂H₄ + H₂O + Me_yO_x-1 (Reaction 1a)

CL-ODH (Regeneration)

Me_yO_x-1 + ½·O₂ → Me_yO_x (Reaction 1b)

In this study, using comprehensive second law analysis in conjunction with process models, we show the potential of CL-ODH to intensify the production of ethylene from ethane. It offers comprehensive proof-of-concept for CL-ODH (Haribal et al., 2017, Li and Neal, 2017, Neal et al., 2016, Sofranko et al., 2016, Yusuf et al., 2017), based on recent breakthroughs in redox catalyst development and process intensification. A net fuel demand reduction of up to 81% can be attained, leading to corresponding CO₂ emission savings. The near order-of-magnitude reduction in energy consumption and emissions confirmed in this work primarily results from the curtailed external fuel demand facilitated by the intrinsic advantages of CL-ODH in terms of (1) superior olefin yield, (2) removal of hydrogen (as water) before compression and refrigeration, (3) avoidance of steam usage, (4) significantly lowered operating temperature, and (5) advanced energy integration scheme. On a global scale, with the potential for saving >3 quadrillion BTU and cutting back over 100 million tons of CO₂ each year, the CL-ODH approach warrants further investigations as one of the potential “wedges” leading to a more carbon-neutral society (Socolow and Pacala, 2006). We also report a robust and highly selective prototype redox catalyst, which demonstrates exceptional performance, stability, and fluidization properties, that is significantly better than our previously reported Mg₆MnO₈ model catalyst system (Neal et al., 2016, Yusuf et al., 2017). High ethane conversion and ethylene selectivity were obtained over 1,400+ redox cycles in a laboratory-scale fluidized bed reactor. Performance data of this prototype redox catalyst confirms the potential of CL-ODH for appreciable exergy savings in both the upstream reaction section (up to 3.26 GJ/ton HVP or High Value Products, accounting for the produced C₂ and higher olefins), and downstream separation section (up to 0.89 GJ/ton HVP).

Results and Discussion

The CL-ODH Approach to Ethylene Production

The need for transformative approaches to optimize ethylene production is well recognized and extensively investigated. Although the thermodynamic first law efficiency of conventional steam cracking is as high as ∼95% (Zimmermann and Walzl, 2000), our second law analysis in the present study indicates a significant exergy loss, due to fuel combustion (cracking furnaces) and low energy quality from heat recovery. The downstream compression and cryogenic separation in conventional cracking leads to additional exergy losses. Pushing the cracking units to higher per-pass ethane conversions could reduce compression/separation loads. However, this is not practical owing to reaction equilibrium limitations (Reaction 2) and the propensity of ethylene to undergo secondary reactions, which eventually forms coke on the inner surfaces of cracking tubes.

Ethane cracking/pyrolysis

C₂H₆↔ C₂H₄+ H₂ ΔH_1123K = 143 kJ/mol; K_1123K = 2.35 (Reaction 2)

These factors lead to a net energy demand of 9.75 GJ per ton of HVP with an estimated exergy loss of 13 GJ/ton HVP. Membrane technology is promising for lowering separation costs (Bernardo and Drioli, 2010, Bessarabov et al., 1995, Lin et al., 2018), but it does not address the appreciable exergy losses and carbon emissions incurred in the cracking process. Several alternative ethylene production technologies have been investigated to tackle the upstream exergy losses (see Figure 1). Oxidative coupling of methane (OCM), which converts methane to ethane, ethylene, and water in a set of net-exothermic reactions, has received renewed attention owing to the abundance of inexpensive natural gas produced from North American shale (Arndt et al., 2012, Chua et al., 2008, Jenkins, 2012, Ahari et al., 2011, Elkins and Hagelin-Weaver, 2013, Elkins and Hagelin-Weaver, 2015). However, the low single-pass ethylene yields (<20% ethylene) and considerable ethane by-product (Ahari et al., 2011, Elkins and Hagelin-Weaver, 2015, Elkins and Hagelin-Weaver, 2013) require intensive downstream separation and recycle of unconverted methane and ethane.

Figure 1.

Comparison of Ethylene Production Techniques (Y Is Yes and N Is No)

In the ethane ODH route (Reaction 4), ethane is selectively oxidized to ethylene and water, which can potentially reduce reaction exergy losses and downstream separation costs (Henning and Schmidt, 2002, Al-Ghamdi et al., 2013a, Al-Ghamdi et al., 2013b, Argyle et al., 2002, Qiao et al., 2014). The removal of hydrogen as water pushes the system toward higher equilibrium conversions. The water product can be efficiently removed by condensation, lowering downstream separation loads. Single-pass ethylene yields as high as 78.3% (∼87% selectivity) have been reported for the Mo, V, Te, Nb mixed oxide “M1” catalysts (Xie et al., 2005). However, the use of gaseous oxygen increases parasitic energy consumption from cryogenic air separation (Castle, 2002, Smith and Klosek, 2001) and raises safety concerns. It is also difficult to maintain a combination of high selectivity and high conversion at the same time, limiting single-pass yields. Although several promising co-feed catalysts have been identified (Henning and Schmidt, 2002, Botella et al., 2004, Cavani et al., 2007, Cavani and Trifirò, 1995, Sanchis et al., 2017, Santander et al., 2014, Xie et al., 2005), few have demonstrated industrially satisfactory single-pass yields (e.g., >50%) at commercially relevant conditions (>1:1 ethane: diluent by volume). Figure 2 maps the performance of these alternative routes with respect to ethane steam cracking.

Figure 2.

Representative Ethylene Selectivity versus Ethane Conversion Chart

OCM

2·CH₄ + O₂ → C₂H₄ + 2·H₂O·········ΔH = −280.3 kJ/mol (Reaction 3)

Ethane ODH

C₂H₆ + ½·O₂ → C₂H₄ + H₂O·········ΔH = −105.5 kJ/mol (Reaction 4)

The CL-ODH process eliminates the challenges of traditional ODH and OCM (Haribal et al., 2017, Neal et al., 2016, Yusuf et al., 2017). The sensible heat in the oxidized redox catalyst particles (Reaction 1b) drives the endothermic gas-phase cracking reactions (Reaction 1a). As illustrated in Figure 3, conventional cracking would result in an inevitable exergy loss of >1.7 GJ per ton of produced ethylene, under an ideal scenario of (1) 100% ethylene yield, (2) perfect heat utilization for cracking, and (3) zero steam dilution. In comparison, even without accounting for its several practical advantages, CL-ODH reduces this minimum exergy loss by 27%. This exergy-saving primarily stems from the in situ hydrogen combustion occurring in CL-ODH versus the combustion of externally supplied methane, in conventional cracking. At an industrial scale, these CL-ODH reactions will be carried out in a circulating fluidized bed system with continuous circulation of redox catalyst particles. Compared with conventional ODH, CL-ODH can improve the ethylene yield and process safety while eliminating the costly and energy-intensive cryogenic air separation step.

Figure 3.

Idealized Exergy Conversion/Loss for Cracking and CL-ODH Schemes at 850°C (GJ/Ton Ethylene)

Redox Catalyst Performance and Demonstration

Our previous work has demonstrated a Mg₆MnO₈-based, model redox catalyst with excellent redox kinetics and single-pass ethylene yields of up to 68% (Neal et al., 2016). In this work a significantly improved, prototype CL-ODH redox catalyst (Sofranko et al., 2016) was demonstrated for improved catalyst selectivity, activity, and physical stability. Structural promotion is essential in ensuring sufficient physical and chemical stability of the catalyst, which is crucial for fluidization in a reactive atmosphere at 850°C without considerable attrition or unacceptable drops in activity. Details with respect to the redox catalyst composition can be found in Example 1 (Bed D) in Sofranko et al. (2016). To obtain a preliminary determination of the chemical/redox stability of this new, prototype redox catalyst, it was first tested in a packed-bed U-tube reactor for 115 redox cycles at 1,200 h⁻¹ GHSV/1.6 h⁻¹WHSV (Gas Hourly and Weight Hourly Space Velocity) and 850°C (see Supplemental Information 1.1). The prototype demonstrated up to 74% olefin yield with high C₂₊ selectivity, over multiple cycles, as shown in Figure 4 (also see Table S1). The product distribution from CL-ODH is similar to that from thermal cracking (see Figure 4 and Table S6), with key differences in higher yields toward C₃₊ (particularly 1,3-butadiene) and formation of small amounts of CO₂ and CO (∼4% total yield on a carbon basis). This can be attributed to the combustion of hydrogen by the redox catalyst, which shifts the equilibrium of gas-phase cracking reactions toward heavier products. Overall, more than 75% of H₂ was converted to water, which is sufficient to meet the energy demand of the overall reaction. The prototype is highly stable over 115 cycles, which is indicative of its long-term stability. Over the course of the reactions, a slight loss of selectivity (89.6% versus 88.2% C₂₊) is observed but is accompanied by a slight increase in overall conversion (83.1% versus 86.6% conversion). Fluidized-bed testing demonstrates the chemical and physical stability of the prototype for 1,400 cycles during 10 days of continuous operation. As illustrated in Figure 5 (also see Figures S1 and S2), the redox catalyst is highly promising for the proposed CL-ODH process.

Figure 4.

Product Selectivity (Carbon Basis) and Ethane Conversions in Ethane Steam Cracking versus That Obtained Using the Prototype Catalyst after 5 and 115 Cycles in the U-tube Reactor

(See Figure S1 and Table S1).

Figure 5.

Fluidized-Bed Redox Catalyst Performance

(A) Conversion and selectivity of prototype redox catalyst over 1,400 cycles. (B) Ethylene selectivity versus ethane conversion in a fluidized bed at multiple gas hourly space velocities, over 1,400 cycles (typical conditions: 16 g catalyst, 845°C, 15%–30% ethane, GHSV: 2000 hr⁻¹ – 3250 hr⁻¹); (C) 1,400-cycle fluidized bed testing demonstrating superior physical stability of the prototype redox catalyst.

Process Analysis: Steam Cracking versus CL-ODH

Depending on the gas residence time and solids hold-up in the fluidized bed reactor, CL-ODH offers flexibility in terms of ethane conversion and overall process exothermicity, since ethane conversion in CL-ODH is not limited by a reaction equilibrium as is the case with steam cracking. Detailed comparisons of CL-ODH at 85% ethane conversion (ODH 85) and steam cracking are performed through AspenPlus process simulations (see Supplemental Information 1.2). Additionally, we have also analyzed two other ethane conversion cases, i.e., ODH 67 (67% conversion based on experimental data) and ODH 99 (to evaluate an idealized scenario via extrapolation of experimental data). Compared with our previous study (Haribal et al., 2017), the current model provides significantly greater detail with respect to heat integration and downstream separations.

Table 1 provides a section-wise comparison of energy consumption (also see Table S7). The simulation indicates energy demands of 21.13 GJ/ton HVP for steam cracking versus 5.09 for ODH 85. This corresponds to a net decrease of 15.7 GJ/ton HVP or 76% in primary energy consumption (see Figure S7). It is noted that both CL-ODH and steam cracking co-produce H₂ and CH₄. For CL-ODH, a notable amount of CO is also produced (see Figure 4, Tables S5 and S6). These compounds can either be credited to the process as fuels or purified and sold as by-products. Accounting for these by-products as fuel (using the Lower Heating Value or LHV) produces a credit of 11.38 and 7.30 GJ/ton HVP for steam cracking and ODH 85, respectively. The fuel credit for ODH 85 is partially offset by an additional ethane demand due to lower C₂₊ selectivity. Accounting for these credits, ODH 85 produces a net reduction in fuel demand of 81%.

Table 1.

Section-wise Energy/Fuel Demand of Steam Cracking and ODH 85

		Net Energy Demand/Recovery (GJTh/ton HVP)
		Steam Cracking	ODH 85
Demand	Radiant Zone of the Cracking Furnace	6.63
	ODH Reactor-Regenerator Pair	NA	−1.3
	Preheating and Heat Recovery	9.55	2.62
	Quencha	−3.92	−2.93
	Compression	3.31	2.04
	Refrigeration	2.37	1.31
	Hydrocarbon Separation	3.19	2.85
	CO2 recovery and Acetylene Removal	0	0.5
	Total Demand	21.13	5.09
Fuel credits and penalties	Fuel Gas By-product (CO, H2, and CH4) (LHV)	−11.38	−7.3
	Extra Ethane Feed	0 (by definition)	4.04
Net demand		9.75	1.83

(Also see Table S7).

^aSteam cracking units and the proposed CL-ODH scheme recover a significant amount of heat from reactor furnaces and the product system quench.

Second Law Analysis

Despite the near-perfect thermal efficiencies reported for steam cracking (Tao Ren, 2006, Zimmermann and Walzl, 2000), the source of energy savings achieved by CL-ODH was determined using a detailed second law thermodynamic (i.e., exergy) analysis using AspenPlus (see Supplemental Information 1.3). Exergy analysis is particularly useful in light of the different energy qualities associated with the feedstock, fuels, by-products, and steam. Table 2 compares the section-wise exergy losses in the CL-ODH cases with steam cracking. With respect to steam cracking, ODH 85 has a net exergy saving of 4.26 GJ/ton HVP with the prime share occurring in the reactors (2.84 GJ/ton HVP). The current simulation indicates little exploitable energy (0.093 GJ/ton HVP) in the exhaust stream for steam cracking, confirming high thermal (first law) efficiency. However, in the cracker, a large destruction of exergy occurs in the radiant/cracking and convective zones combined, amounting to 7.6 GJ/ton HVP. As such, even at the reported efficiencies of up to 95% (Zimmermann and Walzl, 2000) (which requires partial condensation in the flue gas stream), sizable exergy losses are inevitable in conventional cracking owing to the irreversibility of fuel combustion and indirect heat transfer, as well as the low quality of the heat recovered.

Table 2.

Section-wise Exergy Loss Analysis

Section			Lost Work (GJ/Ton HVP)
			Steam Cracking	ODH 67	ODH 85	ODH 99
Upstream	Radiant Zone		4.90
	ODH Reactor-Regenerator Pair		NA	1.64	2.06	2.03
	Power Generationa	Preheating and Heat Recovery	2.73	2.21	2.15	2.11
		Quench	1.54	1.30	1.18	1.06
Downstream		Power Generation Block	1.12	1.44	1.34	1.24
	Compression		0.48	0.34	0.31	0.27
	Refrigeration		0.70	0.45	0.35	0.31
	CO2 and Acetylene Removal		0.10	0.18	0.22	0.18
	Separation		1.43	1.44	1.13	1.05
Total			13.00	8.99	8.74	8.26
% Reduction				30.8	32.7	36.5

(Also see Table S10).

^aThe Heat Recovery and Quench sections produce steam utilized for power generation.

The idealized analysis in Figure 3 indicates potential exergy savings using hydrogen instead of methane as the fuel. Although substitution of fuels cannot be made without changing other process conditions, a direct comparison of the exergy versus lower heating values of H₂ and methane indicates that the use of hydrogen as fuel is responsible for 1.4 GJ/ton HVP in apparent exergy savings. The remaining 2.1–2.4 GJ difference between the CL-ODH and steam cracking cases, in the reactor/preheating sections, is attributed to the significantly improved heat integration and elimination of steam dilution. Of particular note for steam cracking is the 350°C temperature differential between the radiant zone of the fire box and the highest temperature inside the cracking coils. This corresponds to a major irreversibility. In contrast, CL-ODH directly utilizes the sensible heat in the oxygen carrier particles to supply the heat required for ethane cracking, thereby minimizing the irreversibility.

The exergy savings of CL-ODH in the downstream of the process, illustrated in Figure 6 and Table 2 (also see Figure S7), is relatively straightforward. The removal of hydrogen as condensed water in CL-ODH prominently reduces the volume of gas that must be compressed and refrigerated (37% volume reduction). Combined with a higher per-pass yield, the quench, compression, and refrigeration sections in CL-ODH can result in up to a 0.9 GJ/ton HVP reduction in exergy loss. In spite of the higher overall downstream power demand, the exergy loss in the power generation section of steam cracking appears to be less than that of the CL-ODH cases. This results from the need to burn more fuel directly for power generation due to less heat recovery from the CL-ODH reactors. However, the improved power generation in steam cracking does not offset the large exergy losses in the furnace. When exergy losses in the pre-heating/heat recovery and quench are included in power generation, CL-ODH reduces exergy loss by 0.44–0.97 GJ/ton HVP (see Table 2) in these sections.

Figure 6.

Process and Exergy/Lost Work (LW) Schematics

(A) Steam cracking of ethane and (B) CL-ODH of Ethane (Results are in GJ/ton of high value products).

Exergy savings for CL-ODH in the separation sections are more limited than in the upstream section (see Table S10). This is attributed to a combination of (1) the need for CO removal before acetylene hydrogenation and (2) the inherently heavy separation demands imposed by polymer-grade ethylene specifications. The CL-ODH reaction product contains sufficiently high CO to poison the catalyst used for acetylene removal via selective hydrogenation (Battiston et al., 1982, Schbib et al., 1996). This necessitates the placement of acetylene removal (the de-acetylenizer) after the de-ethanizer, requiring an additional heat exchange load. Polymer-grade ethylene purity requirements (99.99%) impose a high exergy loss in the C2 splitter. This is the case even for the ODH 99 case with 99% ethane conversion (0.487 GJ/ton HVP). This underscores the potential impact of membranes and other advanced hydrocarbon separation technologies on ethylene production.

Broader Impact on CO₂ Emission Reduction

CL-ODH's higher exergetic efficiency also leads to substantial reduction in CO₂ emissions. A comparison of CO₂ emissions is given in Table 3 and Figure 7 (also see Figure S9). If fuel gas by-products are exported without credit, steam cracking gives CO₂ emissions of 1.26 ton/ton HVP, consistent with other simulations in the literature (Tao Ren, 2006). By comparison, ODH 85 emits only 0.45 ton/ton HVP, leading to a 64% reduction, as shown in Table 3. If the hydrogen is separated and recovered (at the cost of additional energy for pressure swing adsorption), it may be credited as a zero-carbon fuel against methane. Additionally, the CL-ODH cases also capture CO₂ in the product separation section, which, if beneficially utilized or sequestered, reduces the CO₂ emissions of the process. When hydrogen is credited as a by-product (at LHV) and the 0.093 ton of captured CO₂/ton HVP in the CL-ODH cases are credited, the ODH 85 emits 0.19 ton of CO₂/ton HVP. This represents a 78% reduction compared with steam cracking (0.88 ton CO₂/ton HVP). On a commercial scale of 1.5 million tons per annum (MTA) plant, this corresponds to a reduction of over 1 million tons of CO₂ each year.

Table 3.

CO₂ Production by Source (Ton/Ton HVP)

		Steam Cracking	ODH 85
CO2 Source	Fuel By-products Burneda	0.17	0.32
	External Fuel Burned	1.09	0.035
	CO2 from Reactor	NA	0.093
	CO2Produced	1.26	0.45
Credits	H2 Recovery Penalty	0.094	0.027
	Hydrogen Credit (LHV)b	−0.47	−0.14
	CO2 Capture Credit	NA	−0.093
Net CO2Emitted		0.88	0.19

^aCO and methane, excludes hydrogen, to eliminate fuel effects on heat recovery fuel gasses are treated as a credit against methane.

^bCalculated as CO₂ from methane versus the same LHV of hydrogen.

Figure 7.

CO₂ Emissions of Steam Cracking and ODH Processes under Various Assumptions:

(1) No H₂ Credit: H₂ is recovered and exported but not credited.

(2) Mixed Fuel Gas: H₂ is not recovered and is burned along with CO and CH₄ as fuel.

(3) with H₂ credit: H₂ is recovered and credited at 9 kg CO₂/kg H₂ (Spath and Mann, 2000).

(4) CO₂ Capture Credit: H₂ is recovered and credited at 9 kg CO₂/kg H₂ and CO₂ recovered from ODH product stream is beneficially utilized and not emitted.

Fuel gas and mixed fuel gas are calculated as a credit against methane burned to eliminate fuel composition effects on flue gas heat recovery. The Mixed Fuel Gas, ODH 99 case exports unused fuel gas at no credit.

Multiple justifiable assumptions can be made about the crediting of hydrogen as either a low-carbon fuel or an industrial feedstock. For example, although hydrogen can be utilized as a low-carbon fuel in conventional cracking, burning fuel-gas with high hydrogen concentrations can lead to high NO_x emissions (İlbas et al., 2005), requiring costlier emissions control. Additionally, in highly integrated areas, hydrogen can be a valuable feedstock for processes such as the production of low-sulfur fuel via hydrodesulfurization (Grange, 1980, Pecoraro and Chianelli, 1981). In this case of integrated chemical use, crediting hydrogen at a rate consistent with displacing steam methane reforming is reasonable (using 9 kg CO₂/kg H₂ [Spath and Mann, 2000]). However, although the various values of the hydrogen by-product credit can affect the absolute value of the overall net CO₂ emissions, it does not change the overall trend between cases. For lower credits, the purification and transportation challenges for hydrogen can make it less carbon intensive to burn the hydrogen as a fuel rather than export it. The trend in CO₂ emissions is consistent with the energy demand and exergy loss trends (CO₂ from Steam Cracking ≫ ODH 67 > ODH 85 > ODH 99), with the ODH 85 case giving a 60%–87% emission reduction and the ODH 99 case giving a 65%–94% reduction, depending on the assumptions used (see Figure 7). This confirms that the CO₂ reduction of CL-ODH is not an artifact of “fuel substitution” (H₂ versus CH₄). The ability to burn hydrogen without NO_x, while still providing some usable H_2, represents another advantage of CL-ODH.

Conclusion

In spite of decades of process optimization and high thermal (first law) efficiency, steam cracking remains an energy- and carbon-intensive process. This is due to large exergy losses incurred by fuel combustion, extensive heat transfer and quench requirements in the cracking furnace, as well as significant compression, refrigeration, and separation loads. Although conventional wisdom dictates the infeasibility of substantial efficiency improvement of steam cracking (which is already 95% thermally efficient), we show that ethylene production can be significantly intensified via exergy loss minimization. The transformative CL-ODH approach has the potential to produce ethylene with near-order of magnitude reduction in energy consumption and CO₂ emissions. The redox catalyst holds the key for the CL-ODH scheme. In this work, we demonstrate the technical feasibility of producing stable, fluidizable, active, and selective redox catalyst particles. Using a prototype redox catalyst, we demonstrate that >85% ethane conversion is achievable while maintaining high selectivity over 1,400 fluidized-bed redox cycles. Modeling of CL-ODH, based on experimental yields, demonstrates lower exergy loss (second law) per unit of HVP, compared with steam cracking. CL-ODH leads to substantial energy savings in the reactor sections. It also facilitates easier downstream processing owing to removal of hydrogen as condensable water and significant increase in ethane per-pass conversions compared with those attained in steam cracking. This improved efficiency leads to a CO₂ reduction of up to 87%. These findings not only support the feasibility of CL-ODH but also provide a useful guidance to design intensified chemical production processes with significantly lowered emissions. If adopted at a global level, this innovative process can reduce annual CO₂ emissions by over 100 million tons for ethylene production.

Limitations of the Study

The use of RStoic reactor models, along with other simplifying assumptions, may not fully capture complex behavior in a circulating fluidized bed reactor that could be revealed in future pilot scale testing.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Published: September 27, 2019