Max O’Neill summarises his final-year research project which investigated and analysed forward osmosis including membrane characterisation and the variables which contribute to the efficiency of the process.

Forward osmosis (FO) is the transfer of water across a semi-permeable membrane due to the osmotic pressure difference of the two solutions. Osmotic pressure is the pressure needed to stop the transfer of water through a semipermeable membrane to produce pure water (Feher, 2017).

The water is transferred from the feed solution (low osmotic pressure) to the draw solution (high osmotic pressure) through the membrane in an attempt to reach equilibrium. The efficiency of FO can be measured by the amount of water that travels through the membrane area per hour (l/m2.h) otherwise known as water flux.

FO does not require a large energy input, making it attractive for industries as a method of purifying water. Although FO requires an additional step to separate the purified water from the draw/brine solution it is expected that it will require significantly less energy than the alternative high energy reverse osmosis systems. This creates the possibility of having a sustainable way to produce potable water globally.

Figure 1: Principle of Osmosis, Feed is the low Osmotic Pressure, Brine is the high Osmotic Pressure and Δπ is the difference of Osmotic Pressure

The use of FO as an application has not yet been implemented at an industrial scale. There are several factors that still need to be resolved. One of the challenges for FO would be the unknown scalability factor of FO at an industrial scale.

The process of scaling up the application is a difficult task as the effectiveness of the bench-scale or pilot-scale systems might not carry forward into the industrial scale.

Another challenge which is hindering the emergence of FO as an industrial application is the lack of a standard method of membrane characterisation. Without the standardised method of characterisation, it leads to a lack of commercial competition for the development of new membranes.

The purpose of the project is to tackle these challenges and further progress FO to help reach the stage where it can be implemented at a larger scale. To help fulfil this an application for FO was selected to illustrate the different industries FO could be applied to. The main aims of the project were:

  • Selection and experimental application of the chosen method for FO membrane characterisation.
  • The experimental investigation and analysis of variables and their effect on FO efficiency in the dairy processing industry. Using Urea as a proxy for skimmed milk.
  • Sustainability Report on both the environmental impact and industrial economics.

When carrying out the experiments the following parameters were recorded to calculate water flux:

Time; change in mass

For the chosen characterisation method as laid out by Tiraferri et al the following additional parameters were recorded:

Temperature; flowrate; concentration

The chosen characterisation method was as laid out in the research paper (Tiraferri et al,. 2013). The research paper also provided the additional resource of an excel template to execute the characterisation method.

This characterisation method consisted of using a range of concentrations for the draw solution during an experiment. This experiment was repeated four times to ensure that the values obtained for the structural parameters are accurate.  

This method produced accurate results for the structural parameters of the FTS H2O CTA FO membrane which was acquired from Sterlitech. The experimental results obtained were compared to values obtained from literature which deduced the accuracy of the experimental results.

For this method of membrane characterisation is defined by Water permeability coefficient (A), Solute resistivity (K), Solute permeability coefficient (B) and Structural parameter (S).

Table 1 Experimentally obtained Structural Parameters for the FTS H2O CTA FO membrane

These experimentally obtained structural parameters in Table 1. were then used to produce mathematical models for the FO membrane.

Figure 4 Mathematical Models of the FO membrane for Water flux

The mathematical models are developed from the transport equations obtained from Tiraferri et al. and Lee et al. It can be seen in Figure 4. the models both have a high level of accuracy between 0 – 1 mol.

As the models continue to predict water flux over the increase molarity, it can be seen that Lee et al. model begins to overestimate the effect of external factors has on water flux.

The Tiraferri model, however, is more accurate in its predictions of water flux. There are several factors associated with this and definite conclusions should not be made from this data. It must be questioned if the two hour time allocation to the characterisation method is truly representative of a real-life model.

Investigation and analysis of variables on water flux

The investigation of the effect of the variables on FO efficiency was carried out for high low ranges of temperature, concentration and flowrate of the draw solution.

The experiment was carried out using a design of experiment and two replicants to ensure the results were unbiased and accurate. Figure 5. shows that the combination with the highest efficiency was high temperature, high draw concentration, and high draw flowrate.

Figure 5 Run Set-ups effect on Water flux

From this initial investigation, the best set-up for the efficiency of FO would be considered as being the high level of each variable working in conjunction. This assumption was further analysed by using Minitab as a method for statistical analysis.

By using this method of analysis there was no assumptions to be made as all the results were found statistically. It can be seen in Figure 6. that in fact, the three variables working in any form of conjunction do not have a significant effect on water flux.

It was discovered that each variable working independently has a significant effect. The variable of concentration can be seen as having the most significant effect on water flux. This indicates that the use of the variable concentration may be the most beneficial variable in terms of increasing efficiency.


The sustainability aspect of FO is an important factor as to why this process is attractive to industries. The environmental side of the process was assessed by completing life cycle assessments on the main materials required.

These materials were either found to be environmentally friendly or environmentally friendly alternative materials could be used in the future when the process has developed further.

The energy cost savings associated were calculated for the replacement of falling film evaporators in the dairy industry with the FO process.

As illustrated below in Figure 7, the FO process is compared to mechanical vapour recompression (MVR), the next cost-effective alternative to the sole use of falling film evaporator.

The MVR consumed 70% less energy than a falling film evaporator while FO process consumed an impressive 88-95% less energy. The assumptions for the payback period were as follows:

Throughput of 370,000 kg/year; Initial Cost for MVR set-up: €5,000

Cost of Industrial Electricity 0.13€/kWh; Initial Cost for FO set-up: €5,000

Figure 7  Break even Graph

Conclusions and recommendations

The membrane characterisation method and results were found to be accurate. The method developed by Tiraferri et al. could potentially be used as the industry standard.

Experimental results matched Tiraferri et al, model across the full range while Lee et al, model was accurate at low concentration ranges but not high. Better transport equations are needed to accurately model FO for wider concentration ranges.

The independent use of the concentration variable may increase efficiency. The use of concentration in conjunction with other variables may be a cost effective way of increasing efficiency: this warrants investigation. The various parameters should  be investigated across a wider range of values.

FO is beneficial both environmentally and economically making this process a possible sustainable alternative to methods of water separation.

Author: Max O’Neill is a chemical and biopharmaceutical engineering student at Cork Institute of Technology. 


Dr Aisling O’Gorman who gave me advice throughout the project and always went that bit extra to ensure I had the best possible chance of success; Dr Phil Kelly, Teagasc.


1.) Feher, J (2017) '2.7 - Osmosis and Osmotic Pressure', in Feher, J. (ed.) Quantitative Human Physiology (Second Edition). Boston: Academic Press, pp. 182-198.

2.) Lee, J, Choi, JY, Choi, J-S, Chu, KH, Yoon, Y and Kim, S (2017) 'A statistics-based forward osmosis membrane characterization method without pressurized reverse osmosis experiment', Desalination, 403, pp. 36-45.

3.) Tiraferri, A, Yip, N, Straub, A, Romero-Vargas Castrillón, S and Elimelech, M (2013) 'A method for the simultaneous determination of transport and structural parameters of forward osmosis membranes', Journal of Membrane Science, 444, pp. 523-538.