Comparison of cryogenic and membrane oxygen production implemented in the Graz cycle

Highlights

• A membrane-based oxygen production cycle is coupled with the Graz Cycle.

• Two membrane cases are identified and compared with the cryogenic base case.

• High membrane areas (in the order of $10^5\text{–}10^6$ m${}^2$) are needed.

• Thermal efficiency improvements of 0.61 to 2.30% are achieved.

Abstract

One of the most promising technologies to decrease greenhouse gas emissions is carbon capture and storage (CCS). Oxy-fuel combustion, in which high-purity oxygen mixed with flue gases is used to burn fuels, reduces the complexity of CCS systems. Several methods have been studied for oxygen production to feed the process, where cryogenic air separation is the most mature technology. However, it is a highly energy-intensive method, which motivates the research of other alternatives such as ceramic membranes.

In order to compare the performance of both oxygen separation methods, the coupling of an oxygen production cycle based on ceramic membranes with a high-efficient power cycle, the Graz cycle, is studied. This cycle initially operates using cryogenic air separation.

The calculations of both cycles are carried out using the simulation software IPSEpro. For the membrane cycle, two cases are identified, whose main difference is the method to reduce the oxygen partial pressure on the permeate side of the membrane: vacuum generation (Case 1) and membrane sweeping (Case 2). Both cases are optimized, considering the thermodynamic conditions of the membrane operation and its effects on the energy consumption of oxygen production. Membrane cases achieve 54.08% and 55.76% of net efficiency for Case 1 and 2, respectively, being 0.61% and 2.30% points higher than the base case. Furthermore, the differences in turbomachine performances and streams variations are discussed, considering the effects of energy integration of membrane cases.

Introduction

Currently, one of the challenges in the industry is the reduction of pollutant and greenhouse gas emissions. The International Energy Agency (IEA) [1] reports that pre-COVID levels of emissions will be reached again around 2023–2025 due to the recovery of economic activities.

In this sense, the power production sector is one of the leading sectors of pollutant gases production. In the European Union (EU), approximately 21 % of the total energy consumption is by means of electricity, whereas 40 % is generated burning fossil fuels [2]. Thereupon, governmental frameworks were introduced whose purpose is the reduction of pollutant and greenhouse gas emissions. The EU has presented the 2050 long-term strategy, aiming to be climate neutral by 2050, aligning with the actions proposed in the Paris agreement [3].

One of the strategies to reduce the emission of greenhouse gases such as

is Carbon Dioxide Capturing and Storage (CCS), which is a promising procedure in which

is separated, transported and stored in long-term storage isolated from the atmosphere [4]. The authors expect CCS technologies to be a key in some industrial sectors as cement production, where 60 % of the

production is unavoidable due to the nature of calcination, an important part of the process [5]; also, in the power production sector, where most of the thermal energy sources are fossil fuels like coal and natural gas [6].

From this perspective, oxy-fuel combustion is seen as one of the most promising technologies to enable easy sequestration of

in industry. In oxy-fuel combustion, the combustion is performed using a mixture of oxygen and a thermal buffer (normally recycled flue gas composed of a

/

mixture), used to control the combustor temperatures instead of regular air. This removes nitrogen from the process, leading to a flue gas of higher

concentration, where

can be easily captured after a previous dehydration [7]. In addition to that, it has been reported that oxy-fuel combustion almost totally avoids

production, which can be up to 40 times smaller than with conventional combustion when using the same oxygen content, due to the removal of nitrogen from the comburent [8].

Oxy-fuel combustion has been studied for different types of fuels, exhibiting acceptable cycle efficiencies and high rates of carbon capture. Shi et al. [9] studied the use of different solid fuels as coal, lignite and sawdust in oxy-combustion power plants: for all cases, more than 97 % of the

could be captured during this operation. An oxy-combustion liquefied natural gas (LNG) power plant was studied by Liang et al. [10], where two auxiliary power generation sub-systems were also implemented. They compared the combustion in

/

and

/

atmospheres, which were chosen as thermal buffer gases. Efficiencies of 58.78 % were reached in the study, where 94.8 % of the

was captured with a purity of 97.2 %.

A study similar to the case above was performed with natural gas as fuel by Cai et al. [12], using an auxiliary subsystem of power production, and performing the combustion using recycled flue gas, pure

or steam mixed as alternative thermal buffers. A purity of 98 % was obtained for the captured

, achieving a maximum cycle efficiency of 67.6 % for the

/recycled flue gases case due to the high temperature of recirculation, exhibiting as the best option in this study.

In these mentioned cases, high-purity oxygen production was performed using cryogenic methods, the most common method in industry. In this method, the air is separated in a multi-column cryogenic distillation process by taking advantage of the different boiling temperatures of its main components, nitrogen and oxygen. The state-of-the-art method can produce oxygen with a purity of 99.5 %, recovering 97.85 % of the oxygen in the air. However, this is a highly energy-intensive method, causing a penalty of 3 % to 4 % of input energy and an efficiency reduction up to 8.5 % in processes where oxy-fuel combustion is applied [13].

For this reason, another oxygen production method studied to improve oxy-fuel cycle efficiencies is the use of ceramic membranes. Due to a pressure gradient between the feed and permeate sides, these membranes have an oxygen diffusion process through their crystal lattice, favored by high temperatures (preferably $>$ 700 °C), which increase the vacancy sites in the lattice, improving the oxygen flux [14]. There are two different modes of operation for these membranes, called 3-end and 4-end, which differ in the method to create the required differential oxygen partial pressure. The driving force for the 3-end is created by generating a vacuum in the permeate side, while for the 4-end a sweeping gas is used in the permeate side to reduce the oxygen concentration, creating the partial pressure gradient [15].

Skorek-Osikowska et al. [16] presented two cases for a 460 MW oxy-fuel combustion power plant where the integration of a hybrid membrane-cryogenic installation was compared with a cryogenic air separation unit, both used for oxygen production. After a thermodynamic evaluation, it was found that the addition of a membrane to the system improved the first law efficiency by 1.1 %, reducing the auxiliary power consumption by 13 %. Portillo et al. [15] compared the performance of cryogenic and membrane methods to produce oxygen for a coal-power plant. They found a 5 % improvement in the energy efficiency and a lower specific

capture when membranes were used compared with the cryogenic case. Additionally, an advanced zero-emission power plant using a ceramic membrane to produce oxygen was studied by Gunasekaran et al. [7], where the influence of the membrane operation temperature was examined evaluating quantities such as power plant production and efficiency. Maximum efficiency of 53.2 % was achieved for 850 °C membrane temperature capturing 100 % of the

produced during the operation of the plant, where it was found that reducing the temperature could be favorable in terms of efficiency, but drastically affecting the power output as the oxygen production for fuel combustion is affected.

This being said, it is seen that the implementation of membranes could improve an existent oxy-fuel power plant efficiency when the oxygen production is performed using cryogenic methods. In these cases, a proper energetic integration could improve the power plant efficiency, making it more competitive. In fact, membrane implementation can be performed in systems where it is required to reduce pollutant emissions and high-temperature streams are presented to feed the membrane, as it is seen in Serrano et al. [17]

Considering this, a study on the Graz Cycle is performed, which is an oxyfuel power cycle of highest efficiency developed at Graz University of Technology [18], [19], [20], [21], [22], [23], [24], whose primary oxygen source is a cryogenic air separation installation. An oxygen production cycle based on a mixed ionic–electronic conducting (MIEC) ceramic-based membrane is integrated into the cycle to produce the required oxygen instead of the original method, considering the alternatives of 3-end and 4- end membranes. Therefore, a comparative study of three cases is presented: a base case as it was determined in its last optimization [11], where a cryogenic process is implemented for oxygen production; a case where a 3-end membrane is used in the oxygen production cycle (Case 1), and a case where a 4-end membrane is implemented (Case 2).

The main objectives of this paper are:

• Optimize the main operation parameters of an oxygen production cycle using membranes coupled to the Graz Cycle for both proposed cases.

• Compare the performance of the three mentioned cases, considering the power plant production, the efficiency of the whole system and the energy cost of oxygen production.