Scientists study the possibility of using reservoirs with adsorbing matrices for making use of the boil-offs occurring in LNG transported by vessels.

Optimising the thermodynamic aspects of the adsorption processes and the emptying of reservoirs in which natural gas is stored in adsorbing matrices: this is the main goal of the project Optimisation based on the Adjoint Method for Natural Gas Storage Systems, of the Research Centre for Gas Innovation – RCGI. Coordinated by Dr. Ernani V. Volpe, a RCGI member and a professor of the Mechanical Engineering Department of the Escola Politécnica of the USP, the project is going to use mathematical models to improve both the amount of gas stored in a given adsorbing matrix and its capacity of returning this gas for being used.

There are different ways to store natural gas. The most common are liquefying the gas (which generates liquefied natural gas, or LNG) or compressing it (known as compressed natural gas, or CNG). The use of the adsorption technology may be a good alternative to store and to transport natural gas without having to use high pressures (in the case of CNG), or resorting to a liquefaction plant to obtain LNG, which has to be kept at very low temperatures (about -160°C). The technology by which gas is stored in an adsorbing matrix is called ANG –Adsorbed Natural Gas.

The adsorption storage system consists in adhering the gas particles to the adsorbent wall, which can be activated carbon, but it can also be silica gel, activated alumina, synthetic or natural zeolites. “To allow the adsorbing matrix to capture the maximum gas possible and later allow the greatest retrieval possible of this fuel by the user, it is necessary to manage the heat transfers involved in the absorption and adsorbent emptying processes”, explains Volpe.

According to him, for optimising these processes, a physical model and an optimisation model, known as Adjoint Method, will be used. “The first model deals with the physics of outflow and answers the question: how does the physics of the processes occur? The second optimises the solution given by the physical model. The two models work with a solution of differential equations, employing the finite element method. To be solved, they need computational numerical methods, known as CFD (Computational fluid dynamics).”

At the end of the five-year duration of the project, the team, composed by Volpe, post-doc researcher Marcelo Hayashi, and doctoral researchers Bruno Chieregatti and João Brasil Lima, intends to deliver a software package (programmed by the researchers) that models the problem according to the inputs provided, besides presenting optimised solutions to a given objective function. These results will additionally permit to develop methodologies that allow operating the technology adequately.

“The idea is that users may insert the inputs regarding the matrices they intend to use, the operation conditions being followed, and the software returns the best processes possible for filling and emptying the tanks”, says Volpe. He states that, besides the temperature and pressure variables, the models created by the group shall work with inputs related to the porosity and to the permeability of the adsorbing matrix and, in case it is possible, altering the geometry of the tank.

Applications – One of the applications developed in the project will be used in the marine transport of LNG. “There is no perfect thermal insulation, as good as the system used in the tanks transporting LNG by sea may be. Hence, when the gas is transported in its liquid state, and due to the heat exchange with the environment, part of this gas returns to its vapour state, which we call boil-off. The estimated losses due to boil-off is of 0.1% of the volume transported per storage day. In a simple calculation, considering a typical 50-day ocean voyage, this represents about 4.88% of the total volume of gas stored in the vessel.  For example, in a vessel carrying 150 thousand cubic metres of natural gas, this corresponds to about 7.3 thousand cubic metres of gas evaporated.”

According to the professor, these boil-offs should be used. “The resulting gas can be used to move the vessel, which already occurs in some cases. It can be re-liquefied, which requires the vessel to have a re-liquefaction plant, increasing its operational cost; or in most cases, it is simply burnt and launched into the atmosphere, which is not advisable at a time of concern with emissions. Therefore, one of my students raised the possibility of storing it in a tank equipped with an adsorbing matrix, since the lowest the gas temperature, the better the adsorption process occurs. We thus intend to study this possibility, verifying its technical and economic viability.”