Ashraf Habib / Getty Images

Technology Perspective

Towards greener water security

Membranes engineered with carbon nanotube fibres could become cost effective water filtration systems. 

23 December 2019

More than half the fresh water used in Saudi Arabia comes from the sea. Low rainfall and limited groundwater supplies have spurred the kingdom to turn seawater into potable water. Today, the kingdom is the world’s largest producer of desalinated water, with a daily production of 5 million cubic meters, and ambitions for 8.5 million by 2025. But, the cost is enormous — estimated at more than .5 million annually to produce around 76,000 cubic meter per day.

Given the energy demands and cost of the current desalination methods, the country has been looking to develop and invest in new technologies to provide water for its growing population.

KACST has teamed up with the University of Cambridge in the United Kingdom through their Center of Excellence for Advanced Materials and Manufacturing (CAMM), KACST’s National Center for Membrane Technology (NCMT) and the Macromolecular Materials Laboratory (MML) at Cambridge. Through the center, which is part of the larger Joint Centers of Excellence Programs (JCEP), the researchers engineer more efficient water treatments using carbon nanotube (CNT) fibres as membranes for reverse osmosis (RO), the process of pressurizing seawater through membranes that separate out the salt. 

CNTs are hollow tubules made of carbon atoms with outstanding features. Their strength, coupled with their flexibility, have outperformed any other material. They also have chemical properties that are ideal for desalination. They let water pass through, but block salts and pollutants. In particular, the motion of water through their smooth hollow interior is exceptionally fast and frictionless, requiring very little energy input. 

“Carbon nanotubes are like tiny water pipes that allow water molecules to move through them orders of magnitude faster than through any other known channel of comparable size,” explains Bandar AlOtaibi, KACST professor and CAMM co-director. “Compared with current porous materials, membranes made with carbon nanotubes offer higher permeability for their high aspect ratios. They have more channels for water to pass through; like sieves with more holes.” CNTs are also naturally antibacterial and self-cleaning, which offers longevity.

“The kingdom has turned to more innovative ways to provide water to support its development plans. This CNT-based technology has a tremendous potential to become one of the most efficient water purification system,” says AlOtaibi. Currently, 60 percent of the kingdom’s water requirements comes from desalination, 20 percent of which is produced via RO.

“We start from seawater, say from the Arabian Gulf, which has above average salinity, and we need to bring it down to match the potable water standards,” says AlOtaibi. “Current RO systems achieve 99 percent salt rejection with a low water flux, and are therefore energy intensive. We aim to increase the water flux while maintaining the same levels of salt rejection.” 

Combining the expertise of Cambridge’s scientists in CNT fibre production and that of KACST researchers in reverse osmosis (RO), the team, led by Professor Saad Aljlil at KACST’s National Center for Membrane Technology, is planning to construct a new CNT production facility in KACST, similar to the one used in Cambridge, and then use these CNT fibres as pores for RO membranes. 

One of the most advanced method in CNTs production was devised in the Macromolecular Materials group in Cambridge, led by Professor James Elliott, and is known as ‘floating catalyst chemical vapour deposition,’ or simply, the ‘Cambridge process.’ In this technique, sources of carbon (i.e. toluene) in liquid form are injected into a furnace, together with sources of iron and sulphur. They are all vaporized, and carried into the reaction zone by a flow of hydrogen gas. While in other techniques the substrates are placed inside the reaction chamber, with this method all molecules are floating inside the reactor, enabling the continuous production of long, highly aligned CNT fibres. It is also a one-step method that produces and assembles the CNTs at the same time, with no further processing necessary. As the carbon feed is continuously injected, individual CNTs are formed, form bundles and fibres, and are then extracted from the furnace on to a spinning spool.

More specifically, iron and sulphur are necessary as bases for carbon atoms to arrange into CNTs, and to control the CNTs’ diameter and layer number. When subjected to temperatures above 400°C, the source of iron (i.e. ferrocene) starts to break down and release its atoms. Then, as the iron particles travel through the furnace, they start to agglomerate into clusters. The bigger these aggregates are, the larger the diameter of the resulting CNTs. A source of sulphur (i.e. thiophene) is added to promote CNT production and hinder the agglomeration of iron atoms. The number of layers and diameter of CNTs is controlled by the amount of sulphur: the smaller the amount, the fewer the layers. In this way, the researchers can obtain single, double and multi-walled CNTs with variable diameters. 

KACST researchers are planning to soon build a three-zone horizontal furnace capable of reaching a temperature of 1500°C. The design allows the independent control of the input and output temperatures, opening possibilities such as controlling the point at which the iron source breaks down, or lowering the extraction temperature without changing the reaction temperature. By tuning different synthesis parameters, the team aims to produce and test different types of CNTs. 

Once the CNT fibres are created, the scientists will need to insert them in an impermeable matrix, and use them as membranes for RO. The next challenge will be to choose the best arrangement of CNT fibres within the membrane. A viable option is to align all CNTs, creating the so-called ‘carpet’ or ‘forest’ geometry. Another possibility is to have them randomly oriented. “At the moment, it is hard to tell which is the best architecture, but intuitively, single-wall CNTs aligned in parallel to the direction of the water flow would look like the most reasonable choice. The single wall should offer the best water-permeability, and the vertical alignment should represent a less tortuous path for water molecules to pass through. But, sometimes intuition and nature go in opposite directions,” explains the CAMM co-director. 

Tested in Cambridge, the current prototype RO membrane is able to withstand water pressures up to 60 bar and water flows between 2 and 25 liters per minute. The objective is to decrease the necessary water pressure and increase the water flow. The experimental aspect is complemented by computer simulations to better understand how variations in pressure difference, CNTs’ length, diameter, density, chemical functionalization and other factors affect the transport of ions and water molecules through the membrane. 

The promising technology is still in a preliminary stage. Before bringing it to market, it must be competitive in cost and efficiency. “CNT membranes are now costly to produce, especially for large-scale uses, but if their production cost is adjusted by other desirable properties — like self-cleaning, high water permeability, desalination capacity, robustness, anti-fouling, and so forth — then they have the potential to become attractive for industry,” says AlOtaibi.  

This project is part of the KACST-Cambridge University Joint Center of Excellence (JCE-CAMM), which was founded in 2011 and includes research on superconductors, innovative manufacturing technologies and renewable fuels. 

As part of KACST’s efforts towards water sufficiency, they have also signed an agreement in 2015 

with the Advance Water Technology Company (AWTC) for
a solar-powered water desalination plant in Khafji. Taking advantage of techniques developed by the KACST/IBM Joint Center for Nanotechnology Research collaboration, the plant aims to reach a production capacity
of 60,000 cubic meters per day. 

“It is nice to think that the water you are using for cooking, laundry and agriculture, did not cost a fortune to make, and did not leave a large carbon footprint behind,” concludes AlOtaibi. 


  1. Li, Y. L., Kinloch, I. A., et al. Direct spinning of carbon nanotube fibers from chemical vapor deposition synthesis. Science, 304(5668), 276-278 (2004). 

    2. Koziol, K., Vilatela, J., et al. High-performance carbon nanotube fiber. Science, 318(5858), 1892-1895. (2007).

     | article