Both thermoresponsive flat membranes and core-shell microcapsule membranes, with (N-isopropylacrylamide) (PNIPAM) a porous membrane substrate and grafted poly N-isopropylacrylamide PNIPAM gates, were successfully prepared using a plasma-graft pore-filling polymerization method. PNIPAM was prepared to be grafted homogeneously onto the porous membrane sub-strates, in the direction of both the membrane thickness and surface. Regardless of the solute molecular size, temperature had an opposite effection diffusion coefficients of the solute across the PNIPAM-grafted membranes with low graft yields as opposed to those with high graft yields. The PE-g-PNIPAM membranes change from positive thermo-response to negative thermoresponse types with increasing pore-filling ratios at around 30%. Phenomenological models were developed for predicting the diffusion coefficient of the solute across PNIPAM-grafted membranes at temperatures, both above and be-low the lower critical solution temperature LCST. Predicted diffusional coefficients of solutes across both the PNIPAM-grafted flat and PNIPAM-grafted microcapsule membranes fit the experimental values. To obtain an ideal result for the diffusional thermoresponsive controlled release through PNIPAM-grafted membranes, the substrates strong enough to prevent any conformation changes are more suitable for preparing thermoresponsive membranes than weak ones.

Wehave successfully prepared monodispersed thermoresponsive core-shell hydrogel microspheres with a mean diameter of 200-400nm with poly (N-isopropylacrylamide-co-styrene) [P (NIPAM-co-St)] cores and poly (N-isopropylacrylamide) (PNIPAM) shells. The submicrometer-sized monodispersed P (NIPAM-co-St) core seeds were prepared by using a surfactant-free emulsion polymerization method, and the PNIPAM shell layers were fabricated onto the core seeds by using a seed polymerization method. The particle size, morphology and monodispersity, and thermoresponsive characteristics of the prepared microspheres were experimentally studied. In the preparation of P (NIPAM-co-St) seeds, with increasing the initiator dosage, the mean diameters and the dispersal coefficients were almost at the same levels at first; however, when the initiator dosage increased further to a critical amount, the mean diameters decreased drastically and the monodispersity became worse significantly. With increasing the stirring rate, the particle diameter decreased, and when the stirring rate was larger than 600 rpm, the monodispersity became worse obviously. With increasing the phase ratio, the mean diameter became larger simply, and the monodispersity became worse first and then became better again. With increasing the reaction time, the particle sizes nearly did not change, while the monodispersity gradually became better slightly. For the core-shell microspheres, with increasing the NIPAM dosage in the preparation of the PNIPAM shell layers, the mean diameters became larger simply, the monodispersity became better, and the thermoresponsive swelling ratio of the hydrodynamic diameters increased.

A glucose-sensitive microcapsule with a porous membrane and with linear-grafted polyacrylic acid (PAAC) chains and covalently bound glucose oxidase (GOD) enzymes in the membrane pores acting as functional gates was successfully prepared. Polyamide microcapsules with a porous membrane were prepared by interfacial polymerization, PAAC chains were grafted into the pores of the microcapsule membrane by plasma-graft pore-filling polymerization, and GOD enzymes were immobilized onto the PAAC-grafted microcapsules by a carbodiimide method. The release rates of model drug solutes from the fabricated microcapsules were significantly sensitive to the existence of glucose in the environmental solution. In solution, the release rate of either sodium chloride or VB12 molecules from the microcapsules was low but increased dramatically in the presence of 0.2mol/L glucose. The prepared PAAC-grafted and GOD-immobilized microcapsules showed a reversible glucose-sensitive release characteristic. The proposed microcapsules provide a new mode for injection-type self-regulated drug delivery systems having the capability of adapting the release rate of drugs such as insulin in response to changes in glucose concentration, which is highly attractive for diabetes therapy.

Wehave successfully prepared monodispersed thermoresponsive core-shell microcapsules with a mean diameter of about 4μm with a porous membrane and with linear-grafted poly (N-isopropylacrylamide) (PNIPAM) chains in the membrane pores acting as thermoresponsive gates. The preparation was carried out by using a Shirasu porous glass (SPG) membrane emulsification technique to prepare small-sized monodispersed oil-in-water emulsions and using interfacial polymerization to prepare the core-shell microcapsules with porous membranes. Plasma-graft pore-filling polymerization was used to graft linear PNIPAM chains into the pores of the microcapsule membranes. In the SPG membrane emulsification process, theoptimumsurfactant sodium dodecyl sulfate (SDS) concentrationandpoly (vinyl alcohol) stabilizer concentration could be selected by solely considering the monodispersity of the emulsion droplets. However, before the interfacial polymerization process was started, these two oncentrations needed to be monitored to avoid any aggregation of the microcapsules and, if necessary, the appropriate quantity of SDS or Tween 80 needed to be added to prevent the microcapsules from aggregating in the interfacial polymerization stage. The prepared PNIPAM-grafted monodispersed microcapsules with a mean diameter of about 4μm showed satisfactory reversible and reproducible thermoresponsive release characteristics. The release of both NaCl and vitamin B12 (VB12) from the PNIPAM-grafted microcapsules was slow at 25℃ and fast at 40℃, which is due to the closed/open state of the grafted "gates". The "on/off" ratio of the release rate of VB12 from the PNIPAM-grafted microcapsules was much larger than that of NaCl.

A thermo-responsive core-shell microcapsule with a porous membrane and poly (N-isopropylacrylamide) (PNIPAM) gates was prepared using interfacial polymerization to prepare polyamide core-shell microcapsules, and plasma-graft pore-filling polymerization to graft PNIPAMinto the pores in the microcapsule wall. The proposed thermo-responsive microcapsule could be a positive thermo-response controlled-release one or a negative thermo-response one by changing the PNIPAM graft yield. When the graft yield is low, the release rate from the microcapsules is higher at temperatures above the lower critical solution temperature (LCST) than that below the LCST, due to the opened/closed pores in the microcapsule membranes controlled by the PNIPAM gates. In contrast, when the graft yield is high, the release rate is lower at temperatures above the LCST than that below the LCST, due to the hydrophilic/hydrophobic phase transition of the PNIPAM gates.

A series of thermoresponsive gating membranes, with a wide range of grafting yields, were prepared by grafting poly (N-isopropylacrylamide) (PNIPAM) onto porous poly (vinylidene fluoride) (PVDF) membrane substrates with a plasma-induced pore-filling polymerization method. The effect of grafting yield on the gating characteristics of thermoresponsive gating membranes was investigated systematically. The results showed that the grafting yield heavily affected both the water flux responsiveness coefficient and the thermoresponsivity of the membrane pore size. When the grafting yield was smaller than 2.81%, both the flux responsiveness coefficient and the thermoresponsivity of the membrane pore size increased with an increase in the grafting yield; however, when the grafting yield was higher than 6.38%, both the flux responsiveness coefficient and the thermoresponsivity of the membrane pore size were always equal to 1; i. e., no gating characteristics existed anymore. Diffusional permeation experiments showed that two distinct types of temperature responses were observed, depending on the grafting yield. The diffusional coefficient of a solute across membranes with low grafting yields increased with temperature, while that across membranes with high grafting yields decreased with temperature. To get a desired or satisfactory thermoresponsive gating performance, the membranes should be designed and prepared with a proper grafting yield.

Membranes with porous substrates and functional gates that are responsive to environmental stimuli are attracting increasing interest from various fields. Their permeation properties can be controlled or adjusted by the gates according to the external chemical and/or physical environment, and they may find various applications e.g. in controlled drug delivery, bioseparation, chemical separation, water treatment, tissue engineering, and as chemical sensors. The functional gates for stimuli-responsive membranes serve as chemical valves, and have been reported to act in response to changes in environmental temperature, [1-5] pH, [5-10] ionic strength, [11] glucose concentration, [12, 13] electric field, [14] light, [15] redox properties, [16] or different molecules. [17-19] There are many cases in which environmental temperature fluctuations occur naturally, and in which the temperature stimuli can be easily designed and artificially controlled; therefore much attention has recently been focused on thermoresponsive membranes. [2-1] Up to now, almost all of the thermoresponsive gating membranes have featured positively thermoresponsive characteristics, that is, the membrane permeability increases with increasing environmental temperature, because all of the thermoresponsive functional gates were constructed from poly (Nisopropylacrylamide) (PNIPAM). [1-5] In these cases, the membrane pores change from a "closed" to an "open" state when the environmental temperature increases from below the lower critical solution temperature (LCST) of PNIPAM to above the LCST, as a result of the swelling/shrinking conformational change of the polymer. In certain applications, however, an inverse mode of the thermoresponsive gating behavior of the membranes is preferred. Here we report a novel family of thermoresponsive gating membranes with negatively thermoresponsive gating characteristics, that is, "opening" of the membrane pores is induced by a decrease rather than an increase in temperature.

We report the synthesis and characterization of monodispersed thermoresponsive hydrogel microspheres with a volume phase transition driven by hydrogen bonding. The prepared microspheres, composed of poly (acrylamide-co-styrene) (poly (AAM-co-St)) cores and poly (acrylamide)/poly (acrylic acid) PAAM/PAAC) based interpenetrating polymer network (IPN) shells, were featured with high monodispersity and positively thermoresponsive volume phase transition characteristics with tunable swelling kinetics, i.e. the particle swelling was induced by an increase rather than a decrease in temperature. The monodisperse poly (AAM-co-St) seeds were prepared by emulsifier-free emulsion polymerization, the PAAM or poly (acrylamide-co-butyl methacrylate) (poly (AAM-co-BMA)) shells were fabricated on the seeds by free radical polymerization, and the core-shell microspheres with PAAM/PAAC based IPN shells were finished by a method of sequential IPN synthesis. The microsphere size increased with increasing both AAM and BMA dosages. The increase of hydrophilic monomer AAM dosage resulted in a better monodispersity, but the increase of hydrophobic monomer BMA dosageled to a worse monodispersity. With increasing the crosslinker methylenebisacrylamide (MBA) dosage, the mean diameter of the microspheres decreased and the monodispersity became better. An equimolar composition of AAC and AAM in the IPN shells of the microspheres resulted in a more complete shrinkage for the microspheres at temperatures lower than the upper critical solution temperature. Both BMA and MBA additions depressed the swelling ratio of the hydrodynamic diameter of the microspheres.

A novel type of polyethersulfone porous microcapsule membranes with straight pores across the whole thickness was successfully prepared with a gel-sol phase inversion method for the immobilization of microbial cells. The mean diameter of the microcapsules was about 2mm, and the microcapsule membranes were full of straight finger pores. The pore size distribution was from 3.5 to 53.6m and the median pore diameter was 13.6m. The prepared microcapsules had large total pore volume, and total pore area and porosity (as large as 91.7%). The prepared porous microcapsules were used as carriers for the immobilization of microbial cells in the anaerobic wastewater biotreatment process. Immobilized anaerobic microbial cells were found both inside the membrane pores and in the inner spaces of microcapsules. The prepared porous microcapsules were proven to be efficient for the microbial immobilization under anaerobic conditions.

The pore size and permeability control of a glucose-responsive gating membrane with plasma-grafted poly (acrylic acid) (PAAC) gates and covalently bound glucose oxidase (GOD) enzymes were investigated systematically. The PAAC-grafted porous polyvinylidene fluoride (PVDF) membranes with a wide range of grafting yields were prepared using a plasma-graft pore-filling polymerization method, and the immobilization of GOD was carried out by a carbodiimide method. The linear grafted PAAC chains in the membrane pores acted as the pH-responsive gates or actuators. The immobilized GOD acted as the glucose sensor and catalyzer; it was sensitive to glucose and catalyzed the glucose conversion to gluconic acid. The experimental results showed that the glucose responsivity of the solute diffusional permeability through the proposed membranes was heavily dependent on the PAAC grafting yield, because the pH-responsive change of pore size governed the glucose-responsive diffusional permeability. It is very important to design a proper grafting yield for obtaining an ideal gating response. For the proposed gating membrane with a PAAC grafting yield of 1.55%, the insulin permeation coefficient after the glucose addition (0.2mol/l) was about 9.37 times that in the absence of glucose, presenting an exciting result on glucosesensitive self-regulated insulin permeation.