RESEARCH AREAS
1 Circular Economy of Biomass and Food Waste
(I)(A) Waste Management
Effective management and treatment of food and biomass waste is an increasingly prominent issue for countries worldwide. According to the NEA, approximately 800,000 tonnes of food waste are generated in Singapore annually, together with another 750,000 tonnes of horticultural wastes (https://www.nea.gov.sg/our-services/waste-management/waste-statistics-and-overall-recycling). To determine whether it would be more sustainable to use AD to treat these in Singapore, my team started by comparing the harvesting of energy and resources by incineration or by AD using a full life cycle assessment (LCA). Similarly, we also performed an environmental impact comparison to treat the cellulosic fraction of MSW (CF-MSW) in Singapore either by incineration or by AD. These findings have proven it critical to separate the wet wastes, such as food waste and horticultural waste, from the incineration process and treating it by AD. Effectively, this increases the efficiency of incineration and reduces the amount of bottom ash generation while converting food waste into biogas and fertilizer, thus making the entire waste management process more sustainable. As this research was done with the support of our National Environment Agency, the one that manages wastes in Singapore, the findings contributed to the Singapore Zero Waste Masterplan that was released in 2019, especially for food waste management. Separately, we have also developed a similar framework for managing wastewater to enable municipalities to make technological decisions with economic tradeoffs in mind.
Related publications: H. Tong et al., Renewable and Sustainable Energy Reviews 98, 163-178 (2018).; H. Tian et al., Renewable and Sustainable Energy Reviews 150, 111489 (2021).; J. T. E. Lee et al., Waste Management 78, 677-685 (2018); T.H. Tsui et al, Sci Total Env 823, 153616 (2022).
(I)(B) Waste Sorting
As a consequence of the above findings where it is more sustainable to separate the wet food waste from mixed wastes sent for incineration, we have proposed that it is critical to have effective waste sorting at the initial disposal stage. For example, as food waste is wet and has high amounts of Nitrogen, Phosphorus and Potassium that are detrimental to the incineration process, sorting is a crucial way to improve the efficiency of downstream treatment and resource recovery. Hence, my research team has studied waste sorting at the laboratory and pilot level on eventual hydrogen and methane generation. In addition, we collaborated with engineering companies in Singapore and China, under the Singapore-Shanghai Joint Innovation Call by Enterprise Singapore and the Science and Technology Commission of Shanghai Municipality, to design and fabricate a smart food waste sorting system that increases sorting performance since the efficiency of manual waste sorting is low. Through the E2S2-CREATE research programme (which will be elaborated in a later section), our research findings contributed to the policy by the Chinese government for compulsory waste-sorting into 4 streams in 46 cities by 2020 and most major cities by 2025. Similarly for Singapore, these findings contributed to the passing of a law to have food waste segregation in commercial and industrial premises by 2024, and eventually on-site food waste treatment by 2030.
Related publications: J. Zhang et al., Applied Energy 257, 113988 (2020); J. Zhou et al, Waste Management 114, 221-232 (2022).
(II) Pre-Treatment Technologies
One of the interesting outcomes of segregating food waste from general municipal solid waste is our studies into sources of food waste and its various types. From commercial premises, food waste tends to be more heterogeneous with a mix of vegetables, carbohydrates and proteins (meat). However from industrial premises, food waste can be quite homogeneous in significant quantities, such as spent coffee grounds, orange peels, okara (from tofu production), spent grains (from breweries), and even returned stale bread. For effective AD of these wastes, pre-treatment is needed for those with high lignocellulosic amounts. Through multiple, highly fruitful collaborations, we have studied various pre-treatment techniques to enhance downstream operations. These include biological pre-treatment (A/P Loh Kai Chee, NUS), microwave-assisted alkali pre-treatment (Prof. Wang Chi-Hwa, NUS) and wet oxidation pre-treatment (Prof. Birgitte Ahring, Washington State University) and etc. Separately, we have also developed a system integration of hydrothermal treatment and AD for recovering bioenergy and struvite from dewatered sewage sludge which was demonstrated to enhance ultimate biomethane yield by 38%.
Related publications: J. Zhang et al., Energy 137, 479-486 (2017).; X. Kan et al., Energy Conversion and Management 158, 315-326 (2018).; J. T. E. Lee et al., Bioresour Technol 332, 125033 (2021).; J. T. E. Lee et al., Energy Conversion and Management 225, (2020).; L. Mao et al., J Environ Manage 293, 112981 (2021).
(III)(A) AD Development and Optimization
The major focus of our research expertise lies in the development and optimization of AD processes. While AD is well-known and widely used commercially and industrially, the majority are for very large-scale systems in the agricultural and water treatment sectors. For food waste treatment in cities like Singapore and Shanghai, our modelling studies (see Topic VII below) showed that these large AD plants are not feasible due to the lack of space and the high cost of land. Therefore, we proposed smaller-scale AD bioreactors for managing food waste on-site, or mid-sized decentralized systems. There are many challenges is for implementing these successfully as conventional AD is slow with long SRT and low methane production. Therefore, we have focused on improving the effectiveness of smaller AD units and increasing energy recovery by developing multi-stage bioreactors and adding biochar. Notably, we have explored anaerobic co-digestion with several types of waste streams such as waste activated sludge (WAS) and discovered powerful synergies which resulted in higher treatment performance and methane production. We also investigate and optimize different types of AD processes such as high-solid AD systems (HSAD) and thermophilic AD process. In addition to the modification of regular process parameters, we also investigate how AD is affected by other factors such as the presence of plastics and wooden chopsticks, and additives such as biochar, or bio-augmentation with supplementary microbes.
(II) 
Figure 1. Our patented 3-stage anaerobic digester for food waste digestion (I). Algal photobioreactors reducing nitrogen and phosphorus in AD discharge (II).
Related publications: J. Zhang et al., Energy 137, 479-486 (2017).; L. Zhang et al., Bioresour Technol 285, 121333 (2019).; L. Zhang et al., Energy Conversion and Management 223, (2020).; W. Li et al., Applied Energy 209, 400-408 (2018).; L. Zhang et al., Bioresour Technol 302, 122892 (2020).; L. Mao et al., Bioresour Technol 294, 122177 (2019).
(III)(B) Pilot Scale Studies
While most AD research, including ours, are conducted at the bench-scale up to 5L, we were given a mandate to study issues related to implementing real-world systems as part of the key deliverables for the E2S2-CREATE programme. Therefore, we developed a 1000 L pilot-scale digester to anaerobically degrade food waste for biogas production, and collaborated with colleagues including Profs Loh Kai Chee (to study microbial consortia), Wang Xiaonan and Adam Ng (for optimization and modelling), and Wang Chi-Hwa (for combination of AD with gasification). The team initially designed, fabricated, and operated a mesophilic AD waste-to-energy system for treating mixed, post-consumption food waste on-site. The self-sustaining AD system was built in 2018 in a 20 foot container with initial deployment outside a canteen on the NUS campus, in one of the residential buildings of Raffles Hall. Biogas derived from the decentralized AD were used to generate electrical and thermal energy for nearby residents through combined heat and power (CHP) generation technology with the electricity being used to charge mobile phones and laptops while the heat was used to operate the bioreactor at 35-40°C. This trial was conducted for 2 years, demonstrating the feasibility, both economically and environmentally, for managing segregated food waste on-site. As we built this system to be fully mobile in a 20-foot container, we subsequently moved it for deployment in Jurong Island at a cafeteria for workers for 6 months, and then at an airport caterer production site (SATS Ltd).
Arising from the successful demonstration, my research Team was funded in 2020 by the Ministry of National Development (MND) through the National Environmental Agency (NEA) to scale up the AD system to 10 m3 working volume at East Coast Lagoon Food Village (ECLFV). This project is a pilot demonstration for AD treatment for all of the food waste generated at ECLFV, which is between 100-300 kg per day. In addition to proving the technological feasibility to generate biogas and subsequently electricity for the hawker centre, this project also studied the effectiveness of automated sorting system to remove plastic contaminants from the food waste stream in conjunction with behavioural interventions to encourage diners and hawkers and cleaners to help segregate food waste from non-food waste. The psychological study was conducted together with a colleague from the Department of Psychology, Prof Jia Lile. The eventual real-world demonstration validated a number of our bench-scale scientific findings, like the effectiveness of adding biochar to increase methane production and the need for pre-treatment of vegetable waste. However, we also discovered multiple implementation challenges including safety and regulatory issues, varying waste load over the period of 2 years (especially as this study started during the height of the Covid-19 pandemic), and equipment failure and breakdowns.
Related publications: L. Zhang et al., Energy Conversion and Management 209, (2020).; L. Zhang et al., Waste Management 75, 270-279 (2018).; J. Zhang et al., Energy Conversion and Management 228, (2021).
(IV) Utilization of By-Products
AD is an effective biological process to recover valuable materials from biological wastes that can then be used to produce food in a circular economy. One product my team worked on is the production of volatile fatty acids (VFAs) for which my research team has developed a two-phase organic glass mesh assisted leachate bed bioreactor (TPOGM-LBR) to explore the continuous production of VFAs using food waste as feedstock for growing yeasts as proteins. The research team has also studied on the addition of plastics on reactor performance during acidogenic fermentation of food waste for production of VFAs and discovered that addition of HDPE and PS increased total VFA yields by 28% and 47%, respectively.
Another by-product of AD is the nutrient-rich anaerobic digestate that can be used as a fertilizer for agriculture. The team is also working on the agronomic properties of untreated digestate and biochar-filtered digestate residue to overcome the concerns of phytotoxins in digestate. The team is also working on the recovery of nitrogen and phosphorus nutrition from anaerobic digestate by natural superabsorbent fiber-based adsorbent and reuse as an environmentally friendly slow-release fertilizer. With the increasing research into black soldier flies as a means of eating up food waste, we have also shown that black soldier fly larval frass derived from food waste could be a sustainable alternative to inorganic fertilizers.
Related publications: L. Zhang et al., Waste Management 109, 75-84 (2020).; L. Zhang et al., Bioresource Technol 337, 125481 (2021).; J. C. Cheong et al., Sci Total Environ 715, 136789 (2020).; S. Song et al., Waste Management 136, 143-152 (2021).; L. Zhang et al., Waste and Biomass Valorization 11, 5223-5237 (2019).; J. K. N. Tan et al., Waste Management 130, 155-166 (2021).
(V) Energy and Resource Recovery
For many years, the major output of the AD process is considered to be the biogas, comprising mostly of methane and carbon dioxide. For 1 tonne of biomass waste, AD can produce enough biogas to generate between 100-800 kWh of electricity, depending on multiple factors such as the type of waste, the digestion efficiency, the biogas generator efficiency and etc. In addition to the AD process optimization that was described above, we also looked into downstream power generation efficiencies. To improve energy recycling efficiency and realize full utilization of biomass (enhanced energy and material recovery), gasification is added as post-treatment for AD residues. In collaboration with Profs Wang Chi-Hwa (NUS) and Dai Yanjun (SJTU), our research team developed the coupling of AD and gasification hybrid systems (a two-stage hybrid AD-gasification system) for energy recovery, in which AD acted as pre-treatment to convert biodegradables into biogas followed by gasification converting solid residue into syngas. This coupling also enabled the recovery of biochar from the gasifier, which can be used to enhance the AD process and further increase energy recovery.
Due to the many possible parameters to study for AD and gasification processes, our team collaborated with Prof Wang Xiaonan (NUS) on machine learning and AI to further increase energy recovery through inverse design and in-depth interpretation. This also allows us to make full use of our experimental data collected since 2012 to speed up our research in AD.
Related publications: P Gao et al, Energy 79, 522-535 (2015); Z. Yao et al., Energy 124, 133-145 (2017).; X. Kan et al., Bioresource Technol 228, 77-88 (2017).; J. Zhang et al., Energy 152, 214-222 (2018); L. Zhang et al, Bioresource Technol 331, 125051 (2021); J. Zhang et al, Energy Conv Management 228, 113654 (2021); J Li et al, ACS Env Sci Eng 2, 642-652 (2022); J Li et al, Green Chem Eng, in press (2022).
(VI) Microalgae and CO2 Capture
While AD is considered to be a more sustainable method to manage food and biomass waste, the biological conversion process will produce CO2, comprising between 20-40% of the biogas produced. Even the generation of electricity from biogas will produce CO2 that should be recovered if the ultimate goal of the waste management is to be carbon neutral. Therefore, we have studied microalgae since 2005 as a biological carbon capture and utilization method, with the microalgae that can be subsequently used as food or fuel. Microalgae have attracted global interest for the bioproduction of fuels and high-value chemicals due to their high growth rate and simple nutrient requirements as compared to plants. In addition, it also consumes CO2 significantly faster than plants, thus giving is strong potential for carbon capture.
We have studied the symbiotic interactions between bacteria and microalgae and discovered that they could enhance algal growth by consuming excess oxygen and releasing phytohormones. Subsequently, we optimized CO2 usage during outdoor microalgae cultivation to maximize the conversion of CO2 to biomass in bubble column photobioreactors. Because microalgae also have the potential to clean up wastewater, we have also worked on using them for CO2 biofixation and phycoremediation of municipal wastewater. In this aspect, we have found that the digestate from AD being rich in nutrients can be used to grow microalgae, thus cleaning up the digestate for discharge.
Related publications: Z. Guo et al., Journal of Applied Phycology 26, 1483-1492 (2013).; Z. Guo et al., Bioresource Technol 186, 238-245 (2015).; R. Chaudhary et al., International Journal of Environmental Science and Technology 15, 2183-2192 (2017).; R. Chaudhary et al., Environ Technol 41, 617-626 (2020).; S.K. Parakh et al, Bioresource Technol 315, 123822 (2020).
(VII) Waste Management Modelling and LCA
Life cycle assessment (LCA), as suggested by the International Organization for Standardization (ISO), has been employed by many researchers as a standard and scientific tool to assess the environmental impacts of a process/product throughout its life cycle. In collaboration with Prof Adam Ng (NUS), we conducted an LCA of both fossil and renewable-based electricity generation for four environmental impact categories, in the context of Singapore. My team has also conducted an LCA to assess several AD scenarios for treating eatery FW in Singapore. Through these findings, we have proven to the Singapore Government that food waste should be treated separately from other municipal solid wastes, as a more sustainable way to manage waste by increasing the energy recovery efficiency of incinerators while producing more biogas from the co-digestion of food waste with sewage. This was a factor in planning by NEA the new Integrated Waste Management Facility (Tuas Nexus) which locates a PUB sewage-food waste co-digestion plant next to the new incinerator.
As described above, one of the key deliverables for E2S2-CREATE is to provide government agencies such as NEA with decision support systems on waste management technologies. Our team together with Profs Adam Ng, Michel Cardin (NUS), Kua Harn Wei (NUS), Peng Yinghong (SJTU) and Liu Xiao (SJTU), used urban metabolism analysis (UMA) to develop models for managing municipal solid wastes in Singapore and Shanghai. One major output of the model is the suggestion of minimum AD bioreactor size to ensure economic and financial viability in addition to technological feasibility, thus validating our hypothesis for using decentralized AD to manage segregated food waste.
Related publications: S. You et al, Bioresource Technol 218, 595-605 (2016); A. Quek et al., Energy Policy 122, 388-394 (2018).; H. Tong et al., Applied Energy 225, 1143-1157 (2018); H.W. Kua et al, Resource Conserv Recycl 181, 106265 (2022).
2 Biomimetic Materials and Devices
(I) Tissue Engineering
What are the limitations to current tissue engineering systems? Recent strategies of engineering an organ in vitro starts with a biodegradable polymer scaffold designed to simulate the extracellular environment of the body. Autologous or allogenous cells would then be seeded onto the scaffold to proliferate and differentiate. Finally, the cell-polymer construct would be incubated within a bioreactor until a form suitable for implantation is achieved. However, there are many problems encountered to date with this approach, such as: (a) sizes of scaffold being small due to difficulties in seeding cells, in vascularization and in internal necrosis; (b) dedifferentiation of cells after seeding onto the scaffold due to loss of cell-cell contact, extracellular matrix and dissolved growth factors; (c) the inability of random culture of mixed cell population to organize and form a complex tissue; and (d) the poor integration of the engineered tissue into the host organ.
It is my unique strategy, and the core hypothesis of our group since I started at NUS in 2001, that the use of microspheres as scaffolds can overcome all of the limitations above by mimicking cell development at the embryonic stage. As the embryo forms a microspherical blastocyst, cells provide 3-dimensional contact and biochemical cues to differentiate and proliferate. We have found, as have been published in our recent works, that polymeric microspheres are indeed very versatile and suitable for complex soft tissues such as the liver. In particular, (i) they can be easily fabricated from different polymers to tailor the degradation rates for the required growth rates of the cells, (ii) their surfaces can be easily modified using a variety of methods to attach peptides, proteins and carbohydrates to control cell adhesion, migration and proliferation, (iii) they can encapsulate various growth factors for controlled release at different time points to regulate cell differentiation and proliferation, and (iv) different types of cells can be cultured on different batches of microspheres for subsequent controlled co-cultures for organization into the required tissue structures. By providing cells with all the environmental cues as found in vivo, we are taking a biomimetic approach where we hypothesize that the cells can re-organize themselves from isolated cells into tissues as they would during the developmental stage.
An MEng student, Chaw Su Thwin, was the original contributor in starting this project in 2001, while a PhD student, Zhu Xinhao, has subsequently laid the foundation of successfully developing a microsphere-tissue construct comprising liver cells and fibroblasts. This was followed by 2 PhD students, Chen Wenhui and Anjaneyulu Kodali. Wenhui grew neuronal cells that showed axonal extension across microspheres, mimicking the nerve cell contacts made in the fetal stage of development; and in the meantime, Anjaneyulu grew adipose derived stem cells on the microspheres to differentiate into bone, fat and liver cells, truly mimicking the embryo development. This combination of adult stem cells with growth factors on microspheres is the culmination of our hypothesis, with a publication in Macromolecular Biosciences in 2014 described in section 3.2.1 below. Subsequently, another PhD student, Liang Youyun took this work and combined it with collagen gels to mimic cancer cells, developing an understanding of the stiffness of matrices that control cancer proliferation. This early thrust of my research was funded by my start-up NUS AcRF Tier 1 grant in 2001 that resulted in 5 publications, 7 conference presentations and one invited lecture. There were two follow-up grants from NUS AcRF Tier 1 in 2007 and 2011 that enabled me to continue the work, resulting in 8 publications, 11 conference presentations, 1 keynote lecture (APCChE 2014) and 1 plenary lecture (SMS 2011)
(II) Collagen-mimetic peptides
Following from our work above, I have been keenly interested in designing a biodegradable, biocompatible and biomimetic material to be used either to form the microspheres or as gels to hold the microspheres together. Arising from this is a collaborative project with Prof Michael Raghunath and Prof Li Jun (both from the Division of Bioengineering, NUS) in which a joint group project within the Faculty of Engineering was started through the joint funding of 3 NUS AcRF Tier 1 grants in 2005. This project was first lead by a PhD student, Khew Shih Tak, in which a biomimetic collagen was designed firstly using peptides and subsequently by coupling with dendrimers. The culmination of these works resulted in the design of a material that can behave like collagen but without the drawbacks of natural collagen. The results of the work have been published in 6 papers, and also presented in 7 conference papers.
(I)
(II) 
Figure 2. Adipose derived stem cells growing on gelatin microspheres (I), and highly organized bundles of CMPA fibrils (II).
The initial work lead to an interesting idea of better mimicking collagen using peptide amphiphiles. By combining a bioactive collagen-mimicking portion of the peptide amphiphile with self-assembling portions, we have developed a material which very closely mimics collagen, both biologically and physically. Cell adhesion and detachment studies have shown a 99% mimicry of the collagen mimetic peptide amphiphiles (CMPA) to natural collagen, while self-assembly process shows the physical mimicry of collagen fibrils and bundles. This work was initially done by a PhD student, Luo Jingnan, with a seminal paper in ACS Nano in 2011, described in more detail in section 3.2.1. I was also invited for a keynote lecture in the 9th World Congress of Chemical Engineering in 2013 for this work. Following from these, another 2 students, Chen Yiren and Sushmita Sundar, also applied the use of CMPAs to further mimic elastin and fibronectin, developing the biomimetic peptides into a solution that gels on demand within 5-10 seconds. This was funded through a project with Michael Raghunath in 2009, and another with Jiang Jianwen in 2012. Overall, we had 2 publications and 8 conference presentations from this work.
(III) Biomimetic Devices
In recent years, my interest in tissue engineering has shifted from growing replacement organs to forming small tissue pieces as we learnt about the complexities of tissue organization and growth into complete organs. As we have shown to be successful in constructing tissue-like cells on microspheres, I have started a group at NUS in 2015 to develop organs-on-a-chip with colleagues from NUS (Saif Khan, Chen Chia-Hung, David Leong, Lim Thiam Chye), NTU (Sierin Lim, Kang Yuejun), and UIUC (Kong Hyun-Joon), with other potential collaborators like Samir Mitragotri (Harvard) and Alireza Khademhosseini (MIT). Our strategy is to build an interlinked-system of different tissues, mimicking the connectivity of different organs with 2-dimensional tissue pieces. As this is a relatively new initiative, we have been applying for grants to support this research since 2016.
Figure 3. Preliminary design of interlinked organs-on-a-chip to mimic body function.
3 Biomimetic Membranes
In the area of biomimetic membranes, the original research started with molecular imprinting to mimic antibodies. Subsequently with our success in imprinting proteins onto particles, we developed molecularly imprinted membranes for continuous protein separation. This eventually lead to the idea of mimicking cellular membranes using proteins, which is one of our best works in the field of biomimetic membranes.
(I) Molecular Imprinting
The recognition of any molecule is one of Mother Nature’s secret behind the monitoring and co-ordination of various reactions and activities, like respiration, reproduction, and antibody recognition. Molecular imprinting is currently thought of as a feasible technique in imparting molecular recognition properties to synthetic materials. This is a technique which involves the formation of binding sites in a synthetic polymer matrix that are of complementary functional and structural character to its ‘substrate’ molecule. Conventionally, molecular imprinting is carried out through the preparation of a bulky imprinted polymer using the non-covalent approach. However, there are limitations to this method. Firstly, bulk imprinting produces imprinted polymer of sharp and irregular shape which limits its application. Secondly, it is only limited to small template molecules as it will not be easy for the macromolecules to diffuse through to reach the binding sites during rebinding. Thirdly, due to the fragile and sensitive nature of proteins, it is not easy to achieve a good imprinting effect for protein molecules since the reaction environment can easily denature and unfold the template molecules. Lastly, although bulk imprinting is easily carried out in the laboratory, this process is not suitable to be employed at the industrial scale due to its poor thermal dispersion.
I have started research in 2004 in molecular imprinting of proteins for three reasons. The main reason is due to my interest in using growth factors for tissue engineering as above, and a major problem is in obtaining sufficient quantities at low cost for our use. The high cost of proteins including growth factors is mostly due to the purification steps, and having knowledge of molecular imprinting, I believed that this method could be used to lower the production cost. Secondly, with our skills developed from fabricating microspheres, I hypothesized that using polymeric nanoparticles can significantly enhance the success of large molecule imprinting, eliminating the limitations of conventional imprinting as described above. Finally, Prof Bai Renbi (then the Division of Environmental Science and Engineering, NUS) had interest in protein adsorption and had invited me to collaborate in a project on protein separation supported by an NUS AcRF grant. Lead by my PhD student, Tan Chau Jin, we have fabricated polymeric nanoparticles that have shown an unprecedented ability to selectively adsorb the template protein from a mixture of proteins in aqueous solution. The results for this work have resulted in 5 publications, 3 conference presentations, 2 invention disclosures leading to one patent granted in 2009. Our success in molecularly imprinted nanoparticles is the first in the world for highly selective protein separation in a complex aqueous mixture, and the leading paper, discussed in section 3.2.1, published in Analytical Chemistry in 2008 is among my best work to date.
Arising from this, we subsequently were awarded a Bill & Melinda Gates Foundation Exploratory Grant in their first call in 2009 to develop viral imprinted nanoparticles for prevention in virus infection. Together with a PhD student, Niranjani Sankarakumar, we showed that our particles can hinder the infection of E. coli bacteria by M13 bacteriophages, mimicking the capture of viruses by antibodies or white blood cells in nature. This is the first time ever that a polymeric particle was shown to catch viruses, even in a bacteria model, to stop their infection of cells. The resulting 2 publications and 6 conference presentations. In 2014, I collaborated with Prof Christina Chai (Dept of Pharmacy, NUS) to develop these molecularly imprinted nanoparticles for the detection of algal metabolites and toxins, through a grant from NRF EWI. We have filed an invention disclosure in 2017 and were awarded a provisional patent for the MIPs, receiving interest from one of the largest analytical companies, Shimadzu Inc., to commercialize the technology.
Figure 4. Molecularly imprinted nanoparticles for protein separation (I) and aquaporin biomimetic membranes (II).
(II) Aquaporin Biomimetic Membranes
Our skills in molecularly imprinting proteins onto polymeric nanoparticles have lead to ideas on molecularly imprinting proteins onto membranes for continuous separation. In an unpublished work in 2009 by a PhD student, Wang Honglei, she demonstrated that these protein imprinted membranes can specifically separate out a two-protein mixture. As Honglei was co-supervised by my colleague in ChBE, Prof Neal Chung, one of the top 3 membrane experts in the world, he suggested a collaboration with him and another colleague from the Department of Biochemistry, Prof Jeyaseelan Kandiah, on using aquaporins imprinted onto membranes for water purification.
Aquaporins are transmembrane water channels that exists in all living cells, from single cell bacteria to eukaryotic cells like mammals and humans. First discovered by Peter Agre in 1994, for which he won the Nobel Prize, aquaporins have been found to allow only water molecules to pass through and at a very high rate. Therefore, many have theorized that aquaporins can be used as water channels in synthetic water purification membranes for seawater desalination, potable water purification and even wastewater treatment. However, no research lab or commercial company has successfully developed such a membrane that mimics cells. I formulated a strategy to use polymeric and lipid vesicles with aquaporins incorporated into the vesicle membranes in a way to mimic the way mangrove tree filters seawater with its root cells. Instead of using living cells, the vesicles behave like cells, allowing only water to permeate, while an impermeable barrier prevents salts from entering. We were the first group in the world to thus successfully fabricate such a biomimetic membrane with aquaporins, having the highest flux of pure water that is 10 times more than commercial reverse osmosis membranes. After applying for a patent in 2012, we published this method and membrane in the Journal of Materials Chemistry A in 2013. From an initial grant in 2009 by the NRF Environment and Water Industry Programme Office (EWI) for $3.7 million, this work lead to a subsequent larger grant of $8.8 million in 2012 to improve the membrane, scale-up the production of both the membranes and aquaporins, and test the stability of the membranes in various environmental conditions. In this large collaborative project, my own group published 11 out of the 15 papers with 2 patents awarded. We are now discussing the licensing of the patent with a company in China, AQ Biomimic Ltd while also preparing our own spin-off company to produce aquaporins at large scales. I was also invited to give a plenary lecture at the International Conference on Soft Materials (ICSM 2014) in Jaipur, India in 2014 for our work in biomimetic membranes.
4 Collaborative Research
Polymeric drug and gene delivery
The last thrust of my research is mainly an outcome of the need to incorporate controlled growth factor delivery into our biomimetic materials for tissue engineering and organ-on-a-chip devices. In this regard, I have been collaborating with Prof Wang Chi-Hwa (ChBE, NUS) and Dr Yang Yiyan (Institute of Bioengineering and Nanotechnology (IBN), A*STAR, Singapore) to learn about drug and gene delivery since 2001. Prof Wang is an expert in drug delivery for liver and brain cancer, with many publications on modeling drug release from micro- and nanoparticles. We co-supervised my PhD student, Zhu Xinhao, on integrating growth factor release from the polymeric microspheres. In addition to the tissue engineering applications as described above, we had also co-supervised a post-doctoral fellow, Dr Hu Yong, resulting in 2 publications on controlled drug release and their effects on hepatoma cells. We subsequently co-supervised another PhD student, Yan Weicheng, in 2013 who fabricated core-shell microparticles using electrohydrodynamic atomization for stroke treatment, who has been very productive with 6 publications to date.
The second collaboration, with Dr Yang, also started when I joined NUS in 2001. I collaborated with Dr Yang to co-supervise my very first PhD student, Liu Shaoqiong, on synthesizing stimuli-sensitive nanoparticles for cancer treatment. In return, I had then co-supervised Dr Yang’s MEng student, Chooi Kar Wai. As my expertise was in polymer synthesis, I acted as the main adviser for both students on the synthesis of PNIPAAm and its copolymers, while Dr Yang provided the funding, equipment and other resources. Through this collaboration, we have developed knowledge on stimuli-controlled drug and gene delivery, with the ability to characterize nanoparticles, their drug encapsulation strategies, and their effects on cells. The results were published in 5 papers (in which I agreed to be only the co-author for funding purposes) and presented in 3 conferences. We continued our strong collaboration since then, co-supervising 3 PhD students, Nikken Wiradharma from 2004 to 2009, Ke Xiyu and Willy Chin from 2011 until 2017. Our work expanded into peptide-mimetic drug delivery for anticancer treatment with Nikken (with 4 papers), polymeric micelles co-delivery of multiple drugs by Xiyu (2 papers), and antimicrobial polycarbonates by Willy (2 papers).