Soft Polyelectrolyte Hydrogel As Versatile Material in Different Application

Home , Polyelectrolyte, Sodium triphosphate

SOFT POLYELECTROLYTE HYDROGEL AS VERSATILE MATERIAL IN DIFFERENT APPLICATION

YUJEN WANG

Submitted in partial fulfillment of the requirements for the degree of Master of Science

Thesis Advisors: Dr. Gary Wnek, Dr. Joao Maia

Department of Macromolecular Science and Engineering

CASE WESTERN RESERVE UNIVERSITY

August 2016 CASE WESTERN RESERVE UNIVERSITY SCHOOL OF GRADUATE STUDIES

We hereby approve the thesis/dissertation of

______YU-JEN WANG ______candidate for the ______Master of Science______degree *.

(signed)______Gary E Wnek______(chair of the committee)

______Joao M Maia ______

______Alexander M Jamieson______

______

(date) ____ 06/30/2016______

*We also certify that written approval has been obtained for any proprietary material contained therein.

Table of Contexts Committee Approval Sheet…………………………………………………… 2

Table of Contents……………………………………………………………… 3

List of Tables………………………………………………………………….. 5

List of Figures…………………………………………………………………. 6

Acknowledgements …………………………………………………………… 7

Abstract…………………………………………………………………………8

Chapter 1 Introduction

1.1 Polyelectrolytes……………………………………………………………….10

1.2 Poly(acrylic acid)…………………………………………………….……….14

1.3 Previous Work from the Wnek Laboratory.....………………………………...15

1.4 Calcium Ions…………………………………………………………………..16

1.5 Adenosine Triphosphate………………………………………………………17

1.6 Chelators………………………………………………………………………18

1.7 Elactroactive Polymer and Hydrogel Rheology………………………………20

1.8 Poly(acrylic acid) hydrogel as a osmotic pressure-induced artificial muscle...22

1.9 Glycerol as an additive for PAA hydrogel……………………………………23

1.10 Summary…………………………………………………………………….24

Chapter 2 Methodology

2.1 Hydrogel Synthesis……………………………………………………………25

2.2 Calcium Titration and Chelation…….………………………………………..26

2.3 PAA thread Lengths as a Function of pH……………………………………..27

2.4 Electrospun PAA Tube…….………………………………………………….27

2.5 Differential Scanning Calorimetry……………………………………………29 3

2.6 Thermogravimetric Analysis………………………………………………….30

2.7 Rheology………………………………………………………………………30

2.8 Mechanical(Tensile) Testing………………………………………………….32

Chapter 3 Results and Discussion

3.1 PAA Hydrogel with Shrinkage and Expansion Cycle………………………...33

3.2 PAA Shrinkage Ability………………………………………………………..39

3.3 Chelation Abilities…………………………………………………………….40

3.4 Rheology………………………………………………………………………43

4. Conclusion ………………………………………………………………………..46

5. Future Work ………………………………………………………………………47

6. Bibliography..…………………………………………………………………….48

List of Tables

Table 1, Configuration of PAA gel………………………………………………… 17

Table 2, Ingredient of HBS…………………………………………………………..26

Table 3, Configuration of PAA with different percentage glycerol…………………31

List of Figures

Fig 1. Collection of Chelators………………………………………………………..18

Fig 2.CaCl2 and EDTA Reversible Titration Curve………………………………….33

Fig 3. PAA Hydrogel Thread Length variation vs Ph………………………………..37

Fig 4. The two stages of PAA gels………………………………………………...…38

Fig 5. The CaCl2 Titration Curve…………………………………………………….39

Fig 6, Chelators Titration Plot……………………………………………………..…41

Fig 7-1, PAA hydrogel frequency sweep storage modulus. …………………………44

Fig 7-2, PAA hydrogel frequency sweep loss modulus……………………………...44

Acknowledgement

I would like to thank my lab mates in different groups, Maia’s group and

Wnek’s group. For graduate students in Wnek group, they shared their experience with me and helped me setting up my experiments. Especially Anne Walker, she had a good experience in PAA hydrogel. Without her, I could spend at least half year more. For undergraduate students with me, I really appreciated for your diligence and curiosity. I was happy to share my perspective and experiment with you all. Chioma

Onukwuire, Sara Rudolph, Anita Venkataswamy were the three main people helping me complete the titration part. Stephanie Kresic was really helpful for the rheology part. Jae Young Choi was a freshman. Jae and I discovered some fundamental behavior of PAA hydrogels. With Maia’s group, Jesse Gadley and Parker Lee provided me with information about gel rheology. Arman Boromand was a very knowledgeable senior graduate student, who was willing to guide me through the theoretical parts.

Soft Polyelectrolyte Hydrogel as Versatile Material in

Different Application

Abstract

YUJEN WANG

The principal biochemical role of adenosine triphosphate (ATP) involves the driving of metabolic processes linked to its large, negative free energy of hydrolysis.

However, less well known is ATP’s ability to act as a chelator for divalent metal ions.

We recently have become very interested in poly(acrylic acid) (PAA) gels as a model system for biomimicking physiological systems such as muscles and nerves, especially the ability of PAA to switch from a compact to a swollen gel as the result of divalent/monovalent ion exchange. Swelling of calcium ion-containing gels can be accomplished using common chelators such as EDTA or citrate, but it was found in the present work that such gels efficiently re-swell using ATP as the chelator. This is of great interest because two biochemically-significant species, Ca2+ and ATP, can be used to switch PAA gels between compact and swollen states in water. A comparison of the relative abilities of EDTA, citrate and ATP to swell calcium ion-contracted gels is presented.

In addition polymer such as PAA are wet hydrogel materials and can serve as artificial muscles which have recently become a popular topic of research. While applying a voltage, hydrogels deform and translate electrical power to mechanical power. However, Ionic Polymer Metal Composites (IPMC) have to be wet in order to function well. PAA is a new candidate for IPMC, as it is a biocompatible polymer and can absorb a lot of water. Thus, crosslinked PAA hydrogel networks can serve as new

IPMC materials. Toward that end, this research also introduces glycerol as a component of PAA gels for solving two problems at the same time, namely mechanical and rheological modulation of such hydrogels.

1. Introduction

1.1 Polyelectrolytes

Polyelectrolytes are a group of water soluble polymers that include charges in their monomer units. They are different from ionomers, which have only a few charge groups within the polymer chains. For an understanding polyelectrolytes, a few concepts need to be considered. An electrolyte is a substance that produces an electrically conducting solution when dissolved in and at least partially ionized solvent such as water. There are strong electrolytes, which completely ionized in water, and weak electrolytes, which are ionize only incompletely. In order to describe how strongly an electrolyte is ionized, dissociation constant Ka and Kb are defined as

[H+][A−] [OH−][B+] K = where HA ↔ H+ + A−, and K = where BOH ↔ B+ + a [HA] b [BOH]

OH−. The dissociation constants do not change with solution condition in the same temperature, only the concentration of each species varies. Also, the ion product of water, Kw, controls the concentrations of hydrogen and hydroxyl ions in water. At

− + −14 20℃, Kw = [OH ] × [H ] = 10 . Thus, pure water without any solute will have a pH value of 7, which means [H+] = 10−7 = [OH−]. Recall that the pH value equals to

- log[H+].

[H+][A−] α2퐶 We may also write K = as = , where α is the degree of a [HA] 1−α ionization. Since pH = - log[H+] and [H+] = 훼퐶, then

훼 log = 푝퐻 + 푙표푔퐾 = 푝퐻 − 푝퐾 . 1−훼 푎 푎

Polyelectrolytes can also be divided into weak and strong types. Like weak electrolytes, weak polyelectrolytes are not dissociated fully. Polyelectrolyte gels are

10 known for volume changes when then imbibe solvent. Macroscopically, the morphology of polyelectrolyte network can be ionized. Microscopically, attractions and repulsions are caused by each monomer unit in polymer chains as well as small, charged particles, such as counterions. The Bjerrum length defines how close a counterion can reach to polymer chains. That is to say, the Bjerrum length equation is an energy balance, which balance the energy of individual particle with level temperature, so the energy of individual particle is the Boltzmann constant (푘퐵) times absolute temperature (Kelvin) 푇, and the Coulomb attraction force energy is

Coulomb’s constant times two charges over the distance of two charges.

푒2 λ퐵 = (1) 4휋휀0휀푟푘퐵푇

Equation 1,the Bjerrum Length equation, where λ퐵 is Bjerrum Length, e is the charge of electron, 휀0 is the permittivity of free space, 휀푟 is the dielectric constant, 푘퐵 is the Boltzmann constant, and T is the absolute temperature.

However, the Bjerrum length does not take salt concentration into account, so the Debye length (휄퐷) comes into play. The Debye length (휄퐷) is defined as

−1 휀0휀푟푘퐵푇 휄퐷 = 휅 = √ 2 (2) 2푁퐴푒 퐼 1 where Ι = Σ퐶 푍2, e is the charge of ion, 휀 is the permittivity of free space, 휀 is 2 푖 푖 0 푟 the dielectric constant, 푘퐵 is the Boltzmann constant, T is the absolute temperature, and 푁퐴 is Avogadro’s number. The surrounding ions in the system are taken into consideration. As to equations, the ionic strength is put into the equation. With polyelectrolyte system, not only the positive and negative charged ions are taken into

11 account but also the polyelectrolyte concentration and effective charges. The general

1 1 ionic strength equation (I) is Ι = Σ퐶 푍2. For polyelectrolyte, Ι = 퐶 푍2 + 2 푖 푖 2 푐푎푡푖표푛 푐푎푡푖표푛

1 퐶 푍2 + 퐶 ∥ 푍 ∥. Due to the attraction forces, counterions are going 2 푎푡푖표푛 푎푡푖표푛 푝표푙푦푚푒푟 푒푓푓 to condense on the charged polyelecytrolytes, which is called counterion condensation.

Counterion condensation is a phenomenon discovered by Manning in 1969 and discuseed in several papers [1-3]. It is based on the Bjerrum length, and a thermal energy balance with electrostatic energy. It was initially modeled as a linear chain of charged polyelectrolyte in an electrolyte solution, and counterions are attracted to the charged polyelectrolyte chains to reach an entropy minimum in the system. He proposed that the valence of the ions mattered, but the sizes did not. Instead of calculating charge-to-charge attraction, Manning introduced a linear rigid line with infinite length, so that the charge density was uniform. throughout every part of the chain. Then, a charge parameter, ξ, was introduced (eq. 3), which governed the occurrence of counterion condensation. If ξ ≥ 1, that is to say, larger than the Bjerrum length, the counterion condensation would occur [1-3]. The charge parameter is defined as

푒2 ξ = (3) 휀푘푇푏 where ξ is the charge parameter; e is the elemental charge; ε is dielectric constant in bulk solution; k is Boltzmann constant; b is about 0.71 nm. [31]

In 2004, Ripoll et al. proposed that multivalent ions would condense on the opposite charged polyelectrolyte is preference to monovalent ions, requiring a modification of the charge parameter equation. Specifically, the function would be inversely proportional to the charge of counterions. That is to say, the minimum critical value for condensation of multivalent counterions would be lower. Only after all the multivalent ions fully condensed, and the charge parameter was still smaller than 1 (the critical value for monovalent ions), would the monovalent ions would condense.

In 2008, Kundagrami and Muthukumar proposed a theory about competitive adsorption on flexible polyelectrolytes. They argued that polyelectrolyte chains would collapse if the charges of the multivalent cations equals to that of ionizable groups of the polymer. Also, they reported the multivalent counterions may overcharge the polyelectrolyte and another layer of counter charges would appear. Due to the layer overcharges, co-ion condensation would happen. They confirmed their theory by calculation of energy in the system. The plot started with the size decrease when the multivalent ions concentration increased for counterion condensed on the polyelectrolyte. Then, size increases with the ions concentration increase due to the overcharged that would increase the size, and another size decreased with co-ions condensation on the opposite charges on the polyelectrolyte-counterions complex.

Moreover, not only can counterion condensation happen, but also the ionic bridging effect could occur happen in the case of multivalent counterions. The bridging effect by multivalent counterions brings two close polyions chains more close together. When they reached their isoelectric point, the intramolecular collapse would occur [33]. 13

1.2 Poly(acrylic acid)

We recently have become very interested in poly(acrylic acid) (PAA) gels as model systems for biomimicking physiological systems as muscle and nerve [5]. PAA is very hydrophilic and has the form of lightly crosslinked particles, which is widely used in baby diapers for it can absorb many times its weight in water. Each acrylic acid unit has a carboxylic group in their side chains with pKa = 4.5 [6]. After neutralization, PAA is negatively charged. The negative charges along the polymer chains will be similar as actin or tubulin fibers, because these also carry negative charges.

N, N methyl-bisacrylamide is a good cross-linker to prepare a PAA hydrogel.

When hydrogel is neutralized, gel swells due to carboxylate ions repulsion. Another useful cross-linker is ethylene glycol, which is used in electrospun PAA formulation.

After spinning, PAA fibers are thermally treated to induce esterification and which results in crosslinking. Due to the porosity of electrospun PAA fibers mats, the fibers tend to swell when wet yet the electrospun mats remain porous with a large area to volume ratio. In contrast, a bulk hydrogel is dense and the Spaces between the chains are solvent. The water holding abilities experiment can prove this. Hydrogel can hold water more tightly than electrospun fibers, but when it comes to transport efficiency, electrospun fibers are better.

1.3 Previous Work from the Wnek Laboratory

Linghui Meng was a Ph.D student with Dr. Wnek in the department of

Macromolecular Science and Engineering at Case Western Reserve University. He used electrospun PAA to mimic properties of the polyanionic nanofibrous cortical layer (ectoplasm) of nerve [8]. Tube-shaped PAA nanofiber structures were made.

This idea actually originated from the work of Tasaki. He and others have suggested that electrophysiological processes of nerve excitation, and conduction are basically manifestations of abrupt phase transitions of the cytoskeleton in the cortical gel layer of the axon and can be mimicked using synthetic polyanionic hydrogel monoliths [9-

11]. Electrospun PAA fiber constructs were thus an attractive approach in studying synthetic nerves [12].

The approach of Meng et al. [8] involved electrospinning of PAA tubes which neutralized with base and then tubes shank to about 65% of their original lengths, and tubes soaked in 150 mM NaCl solution shrank with the least amount of added

The results were clear. Tubes shrank to about 65% of their original lengths. The tubes soaked in 150 mM NaCl solution shrank with the least amount of added CaCl2, and tubes in the same condition also expanded the fastest when subsequently treated with a chelator. However, no matter what the bathing NaCl concentration was, the tubes expand to about the same length.

These results are quite interesting. 140 mM NaCl is about the same concentration as physiological saline. It is likely that the bathing NaCl solution reduces the electrostatic repulsion of carboxylate anions, thus making Ca2+ binding easier and more cooperative, so HEPES buffered saline is used with hydrogel.

1.4 Calcium Ions

Calcium ions, Ca2+, are of the utmost importance, both intracellularly and extracellularly. Calcium ions play major roles in diverse phenomena including neurotransmission, muscle contraction, blood clotting and apoptosis, and act as an important second messenger. Over 99% of intracellular calcium is bound to molecules in the cytoplasm, or sequestered in membrane-bound organelles so that the intracellular “free”, not bound, calcium is very low ( 10−7 M inside vs 10−3 M outside). This is possible for a few reasons: (i) changes in free Ca2+ levels are used for intracellular signaling and control; (ii) the large gradient across the membrane promotes flux through ion channels; (iii) ATP-triggered calcium pumps are always active; (iv) a high level of free Ca2+is toxic and can cause cell death. On the other hand, the concentration of free extracellular calcium ions is about 1 mM. Changing the extracellular Ca2+ causes profound effects on neuronal function. For example, a decrease in Ca2+ results in a reduction of the threshold for excitation of nerve.

Conversely, an increase in Ca2+ causes reduced excitability. [13-14, 23]

1.5 Poly(acrylic acid) as a physiologically-mimicking material

In order to make a physiologically-mimicking material, understanding how nature works is important. For muscle calcium regulation, there is an organelle called the sarcoplasmic reticulum (SR) that is the calcium reservoir in the muscle cells. The

SR influxes calcium ions by an ATP-consuming pump on membranes and releases

16 calcium ions into the cytosol by ion channels when activated [15]. Then calcium ions bind with troponin in the myofibril that initiates muscle contraction. More specifically, tropomyosin and troponin work together to block myosin binding sites on actin in the resting stage. When calcium ions enter, they bind with troponin. A conformational change of troponin-tropomyosin complex occurs. The binding site is exposed so that myosin binds to actin, and contraction occurs [16].

PAA carries negative charges at physiological pH like tropomyosin and troponin.

When calcium ions bind, the counterion condensation may occur. If the PAA is neutralized by NaOH already, the charge parameter of the PAA chain for monovalent ions is 1, which is higher than the critical threshold for divalent ions (1/2). Thus, neutralized PAA (e.g., sodium polyacrylate) favors calcium ion binding over sodium ions due to electrostatic interaction. When associated with calcium ions, the charge parameter drops to 1/2 from 1, and the linear charge density changes as does the length of the PAA chain. As a result, a contraction occurs. With such an ion- exchanging system, PAA can be conceived as an interesting artificial muscle system by a cycle of addition and subtraction of calcium ions.

1.6 Chelators

Chelators are chemicals that can catch and trap metal ions like a claw. In fact, the word in Greek that chelator is derived from means claws. There are negative charges in chelators that enables the chelation of positive charge metal ions. In a

physical chemistry view, chelators has a higher affinity to metal ions compared to other chemicals in the solution. If there are multiple chelators in the same environment, the competitive chelation would happen, such as our case.

Chelators can compete for calcium ions with PAA as the latter can be considered a polymeric chelator. When chelators are added into the PAA-calcium chain system mentioned above, chelators would bind the ions that do not associate with PAA charges, and local calcium ion concentration decreases. And since the calcium ions leave the polyelectrolyte, the charge parameter increases, so does the length of PAA. By adding chelators little by little, calcium ions in the PAA network would be significantly reduced, and the length of PAA would fully recovered.

Common chelators such as EDTA, sodium citrate, and sodium triphosphate were used in this research. ETDA has a “claw-like” molecular structure that allows it to 18 easily bind to heavy metals and other toxins. This makes it an ideal candidate for chelation therapy which helps to treat lead and other kinds of heavy metal poisoning.

In order to administer this therapy, EDTA is injected intravenously and then, once in the bloodstream, EDTA can trap lead and other metals or toxins. This forms a compound that can be eliminated in the urine [20, 21].

Sodium citrate is mainly used as a food additive in order to preserve food or enhance its flavor. It can also be used to control the acidity of various substances as it is fairly strong base. Recently, it was reported that sodium citrate has a limited capacity for calcium depletion for anticoagulants when compared with EDTA, but it can chelate other ions [22].

Sodium triphosphate is known mainly for its ability to deal with hard water.

Since metal ions can interfere with the effectiveness of cleaning detergents, chelators like sodium triphosphate can be used to combine with the metal ions such as Ca2+ and

Mg2+ and prevent them from forming insoluble complexes with soaps. It can also be used as a polyanionic crosslinker in polysaccharide-based drug delivery [40].

The key interest in this thesis is the ability of ATP to act as a chelator for Ca2+ bound to PAA gels, and to compare its chelation ability to that of common chelators discussed above.

The “CRC handbook of Food Additives” has extensive data of stability constants

(log K1 ) of various metal chelates. Several commonly used chelators stability constants are provided including EDTA, citrate, polyphosphate (triphosphate) and

ATP. The K1 value of these chelators for calcium ions are 10.70, 3.5, 3.0, 3.60, respectively. In homogeneous solutions, the binding power of EDTA is significantly greater than the other three chelators as the stability constant is in logarithmic scale. 19

With the data provided, we can determine that EDTA will be the best chelator among these four.

1.7 Adenosine Triphosphate

ATP (adenosine triphosphate) is a coenzyme that plays a crucial role as energy currency. It can be formed through various natural processes including glycolysis, oxidative phosphorylation, and photosynthetic phosphorylation. It is the central compound for storing energy within the cell, since a living cell cannot store significant amounts of free energy without overheating. Once the energy is stored in the form of ATP, it can be transduced in many essential biological processes since it is an extremely versatile molecule both functionally and conformationally. ATP is able to form many types of intermolecular hydrogen bonds and has the ability to stack well due to the aromatic groups of the molecule. These properties allow ATP to bind well into enzyme reaction pockets as either a substrate or as an effector [28-30].

Additionally, the principal biochemical role of ATP involves the driving of metabolic processes linked to its large, negative free energy of hydrolysis [17]. However, ATP’s ability to act as a chelator for divalent metal ions is less well known [18]. The negative charges on ATP, a phosphoanhydride, makes it less vulnerable to nucleophilic attack by hydroxide ion or water than, for example, a carboxylic anhydride. The negative charge on ATP also accounts for its usual retention within the cell, its kinetic stability to hydrolysis, and its ability to bind to both metal ions and to positively charged components of proteins.

As for the chelation ability of ATP, there have been several studies over the last century. Tice studied a lead-ATP complexation in 1968. He determined that, in the presence of Pb2+, mitochondrial adenosine triphosphatase activity dropped to 60%

[19]. Although the dissociation constant was not directly measured, the paper gave agood explanation about a possible source of lead poisoning. Wilson and Chin also studied the chelation ability of ATP. They showed that binding constants decrease in order of Mg2+ > Ca2+ > Sr2+ and monovalent cations hindered MgATP2− complex formation [14]. There are no apparently studies researches comparing ATP chelation ability to other chelators, affording the opportunity to explore this in this thesis.

1.8 Poly(acrylic acid) hydrogel as a osmotic pressure-induced artificial muscle

Hydrogels can react to osmotic pressure and translate osmotic force into mechanical force. More specifically, under a voltage differential, counterions inside hydrogel can be transport to the oppositely charged electrode by the field driving force and accumulate. With such ion accumulation, an osmotic pressure develops, and a thermodynamic driving force is created that draws water molecules toward the region of accumulated ions to reduce the osmotic pressure difference. The deformation happens due to with unevenly distributed water and differential swelling inside a gel. That is to say, a mechanical force is generated by deformation. We called this kind of polymer as elactroactive polymer (EAP)

Ionic EAP such as PAA, can in principle work at low voltage such as 1 V but need to be wet. With electrode layers on both faces, ionic EAP can be called Ionic

Polymer Metal Composites (IPMC). The mechanism is similar to EAP. These metals on the both sides help electric power distributed well. [24]

Nafion is a material of recent focus as IPMC. Nafion has a fluorinated carbon backbone and fluorine-containing side chains with a sulfonic acid group. Fluorine is known for its strong electron-withdrawing ability, so this makes the sulfonic acid very acidic in terms of pKa. For Nafion IPMC, thin sheets of Nafion are coated with platinum or gold in both sides. After dipping in electrolyte solution, Nafion is ready for actuation [25]. Upon applying low voltages such as 0.5V, Nafion IPMC can bend in a few seconds till almost like a roll [26, 27]. Nafion shows a structural similarity to

PAA, with a negative charges in side chains, although PAA is very hydrophilic and the pKa of the –COOH group is much higher . Under these circumstances, and since

PAA can swell in water and retain moisture. We started to have an idea of using PAA as an artificial muscle actuator.

1.9 Glycerol as an additive for PAA hydrogel

PAA hydrogels are soft in wet form but hard and fragile in dry form. It is hard to prevent gels from losing water over time. There are few ways for preventing water loss in the system. Coating is one of the options, but moisture inside the gels could cause delamination in the long run, and gel might lose it desirable properties.

Operating by immersion in water is another option which might maintain desirable properties, but would limit the conditions of operation. Additives can be an option as 22 well with glycerol being a good choice. It has three hydroxyl groups on a three-carbon backbone. As a small, polar, water soluble chemicals, it is fully compatible with water.

It also is served as an anti-frozen as well as moisturizer ingredient in many applications. Adding glycerol into chemically crosslinked hydrogel systems could enhance the hydrogel operational temperature range and keep hydrogel moist still with desirable properties.

The properties of PAA-glycerol gels were studied with a variety of techniques, including thermogravimetric analysis can obtain the degradation temperature.

Differential scanning calorimetry, tensile testing and rheological testing.

1.10 Summary

It is interesting to speculate that, if ATP can be shown to be an effective chelator of 퐶푎2+ bound in PAA hydrogels or electrospun tubes, a PAA actuator based on reversible dimensional changes driven by interactions that are biologically-relevant could be designed. Toward that end, it was of interest to study some rheological and mechanical properties of PAA gels. Hydrogel rheology provides a panorama of gel modulus at different frequencies. Also, we were interested in strain sweep to determine the breaking point of such hydrogel.

2. Methodology

2.1 PAA Hydrogel Synthesis

The chemicals used to prepare a typical 10 mL sample of the PAA gel includes

2.1 mL of 3 M acrylic acid, 0.4 mL of 130 mM N,N’-methylene bisacrylamide, 0.3

mL of 90 mM potassium persulfate, 1.0 mL of 10x HEPES buffered saline, and 7.2

mL of water. The calculated quantity of water was first added to a conical tube,

followed by the measured amounts of acrylic acid, N-N’ methylene bisacrylamide,

and potassium persulfate. It is important to add the potassium persulfate last, as it is

the initiator. Once all the materials were in the conical tube, the solution was mixed

vigorously using the vortex mixer. After that, the solution was degassed using the

sonicaction bath in order to minimize the bubbles.

Ingredients (Unit: mL) PAA Thread Shape Hydrogel

3M Acrylic Acid 2.1

130 mM N,N’Methylene Bisacrylamide 0.4

90 mM Potassium Persulfate 0.3

10X HEPES Buffered Saline 1.0

Water to 10 ml 7.2 Table 1, Configuration of PAA gel thread.

Once the mixture described above was prepared and placed in desirable vessels

(for example, Teflon tubes or glass petri dishes), it was placed in an oven at 80 ℃ and

left there for ninety minutes. Then, the polymer sample was removed from the oven

and left to cool for a period of time. The polymer was then transferred to 25 mM

HEPES buffered saline (HBS) solution, which is slightly basic, turning the pH to 7.3.

Eventually, the polymer sample reached an equilibrium swelling level in the HBS solution upon soaking overnight.

HEPES Buffered Saline Ingredients Molarity (mM)

NaCl 140

KCl 5

Na2HPO4 1

Dextrose 6

HEPES 25

Table 2, Ingredient of HBS

2.2 Calcium Ion Titration and Chelation

Polymer gels were then placed in fresh HBS solution, and six pieces of cylindrical gel threads (6 mm diameter) approximately three centimeters in length were cut. Then, the gels were placed in beakers, and titrated with 5 M calcium chloride solution. A 0.5 ml CaCl2 solution was added to each beaker, and the beaker was left to sit on the spinner for five minutes and then the length was measured. Then, the process was repeated over again. Although this method was effective, it took a rather long time to complete. An alternative used was to transfer the gel immediately

25 to a 2 M calcium chloride solution to soak overnight once it had swelled in the

HEPES solution.

In a similar fashion, PAA hydrogels were removed from the solution and six pieces of gel around three centimeters in length were cut. Once the gels were cut, they were measured once more to accurately record their lengths. Each piece of gel was placed in a separate beaker of 5 ml of 25 mM HBS solution. The beakers containing the gels and the solutions were placed on the magnetic mixer for approximately 15 minutes. After fifteen minutes, the lengths of the gels and the diameters were measured once more.

Next, 0.5 ml of the desired chelator solution was added to the existing HBS solution for each one of the gel pieces. The chelators used in this experiment include

0.1 M sodium citrate, 0.1 M sodium EDTA, 0.1 M sodium triphosphate and 0.1 M sodium ATP. Once the desired chelator was added to each of the six beakers, the beakers were once again placed on the magnetic mixer. After a duration of 15 minutes, the diameters and the lengths of the gels were once again measured. Then, an additional 0.5 ml of the chelator was added to each of the beakers. The process continued of letting the solution sit, measuring the length and diameter of the gel, and then adding more chelator. It was expected that the gels would continue to grow in size every time an additional 0.5 ml of chelator was added and the solution. Thus, the process was repeated over and over until it was found that the gels were no longer expanding further, after which the measurements were stopped.

2.3 PAA thread lengths as a function of pH

The pH titration experiments were carried out in a beaker with 20 ml HBS starting from pH 1. The lengths were measured after 2 hours of each NaOH addition. The pH were recorded right after the length measurement. Around 0.5 of pH value was increased, but if overdosed, there were no reverse titration. Magnetic stirs were used throughout the process.

2.4 Electrospun PAA Tubes

Electrospun PAA tubes were prepared in a series of several steps: solution preparation, spinning and post treatment.

Solution preparation: A 4% wt% commercial PAA (DuPont, Mw 450,000) in ethanol and ethylene glycol (EG) was added (16% wt relative to PAA, in this case

0.64%), to ethanol. The solution was degassed with a sonicator until the appearance of bubbles ceased. Prior to spinning, a drop of 1 M sulfuric acid was added and mixed well before the mixture was transferred to a syringe having metal needle. Also 10 % wt solution of poly(vinylpyrrolidone) (PVP, Dupont, 100 K) in ethanol was also prepased.

Spinning the PAA: A thin layer of PVP was first coated onto the rotating mandrel by dipping in PVP solution. Next, a syringe loaded with the

PAA/EG/Ethanol/H2SO4 solution was mounted on a syringe pump and connected with a voltage source (0–30 kV; CZE1000R, Spellman High Voltage Electronics

Corporation) at 15 kV vs ground to the metal needle of the syringe. The distance was

27 set around 15 cm between the needle to rotating mandrel collector. The pump speed was at 1.10 ml/hr, and spinning required about 3.5 hours for each tube.

Post spinning treatment: After electrospinning, the tube still on the mandrel was placed in an oven of 130 ℃ for 1 hr to convert the to crosslinked sodium polyacrylate away in water. Then, the mandrel was placed into a 1 M NaOH/1 M

NaCl solution for 1 hr. The underlying PVP was dissolved in the solution, and the

PAA tube rather slightly expand so that removal of the PAA tube from the mandrel was easy. After removal, the electrospun PAA tube was washed in deionized (DI) water, and the tube was soaked in 25 mM HBS solution for storage.

2.5 Differential Scanning Calorimetry

Measurements of the glass transition temperature of the glycerol plasticized

PAA gels were conducted by differential scanning calorimetry (TA DSC Q100). The calorimetric measurement procedure was similar for all samples regardless of composition. The scanning was performed from room temperature to 120 ℃. The cooling went from 120℃ to -70℃ at a rate of -10℃/min in nitrogen gas with hermetic pan to minimize water evaporation.

2.6 Thermogravimetric Analysis

A TA Instruments Q500 TGA was employed for the thermogravimetric tests.

0.1 μg of weight change can be detected by Q500 TGA. Prior to testing, samples were

dried in oven in 130 ℃ for 1 hour. After drying, a sample at around 10 mg mas was

loaded into the hermetic pan. Starting from room temperature, the temperature

increased to 700℃ by 20℃/min. Pure nitrogen was purged continuously at a flow rate

of 60 ml/min.

2.7 Rheology

(1) Sample preparation: different percentages of glycerol to water content

hydrogel were made. Glycerol was added before synthesis with the ratio of 0, 25 and

50% to rest of the water other than monomer, initiator and crosslinker solution. For

example, there is 1.8 ml of glycerol added inside 10 ml of 25 % of glycerol hydrogel

sample. 7.2 ml was the total amount minus monomer, initiator and crosslinker

concentration. A table above was provided for all ingredients. Synthesis was carried

out in 80 ℃ for 30 mins. After synthesis, samples were carefully remove from the

vessels, and cut with an 11 mm diameter circular cutter and soaked in the same

percentage glycerol/HBS solution overnight for equilibrium at pH 7.3. At equilibrium,

the hydrogel normally swells around 10 times in volume. Before mounting on

rheometer, a 16 mm diameter cutter was used to cut the swollen gel again.

Ingredients (Unit: mL) 0 % Glycerol 25% Glycerol 50% Glycerol PAA Hydrogel PAA Hydrogel PAA Hydrogel

3M Acrylic Acid 2.1 2.1 2.1

130 mM N,N’ Methylene 0.4 0.4 0.4 Bisacrylamide

90 mM Potassium Persulfate 0.3 0.3 0.3 Glycerol 0 1.8 3.6

5 M NaOH (Neutralization) 1.39 1.39 1.39

10X HEPES Buffered Saline 1.0 1.0 1.0

Water to 10 ml 4.81 3.01 1.21

Table 3, Configuration of PAA with different percentage glycerol.

(2). Post synthesis treatment: 0% hydrogels after synthesis were soaked in

different percentage solution to reach equilibrium. As to 25% and 50% of glycerol

hydrogel sample, they were treated with 25% glycerol HBS and 50 % glycerol HBS,

respectively for the exchange of glycerol to water happened in solution.

(3). Rheology: the hydrogel was loaded on a TA ARG2 with 25 mm parallel

upper plate and flat bottom plates. The gel was pressed down to gap between 1-3 mm,

and 1 ml HBS in the same percentage glycerol solution was added around the

hydrogel sample. Time sweep was carried out at 1 Hz, 1 hour oscillation. Frequency

sweep was conducted from 50 Hz to 0.01 Hz at 0.1% strain. The last test was set at

strain sweep test from 1% to 10000%, 1 Hz for knowing the breakpoint of gel.

2.8 Mechanical (Tensile) Testing

The tensile tests were conducted by Zwick/Roell Z0.5. Tensile test sample is

prepared according to the 2.4.1 part, but the hydrogel was cut into strips after

synthesis without soaking in solution. Before loading hydrogel strips, both ends of gel 30 strips are cured by epoxy for preventing slipping. Young’s moduli were obtained by integrating from 0.005N to 0.025N strain, and the breaking points are also recorded.

3. Results and Discussion

3.1 PAA Hydrogel with Shrinkage and Expansion Cycle

A key goal of this research was to demonstrate reversible dimensional changes

of PAA gels upon exposure of the sodium form to CaCl2 followed by treatment with a

chelator. Instead, such reversibility was demonstrated (Figure 2). Two cycles of

shrinkage and expansion are plotted. The Y axis is the normalized length, (measured

length/ original length) and the X-axis is the 5M of CaCl2/0.1M EDTA volume of

solution added into the vessel. The solid dots in the graph represent the shrinking

Reversible Titration Curve (n = 4)

First Cycle Shrinking First cycle swelling Second Cycle Shrinking Second Cycle Swelling Ca2+ EDTA Ca2+ EDTA 1.600 1.400 1.200 1.000 0.800 0.600

0.400 Normalized Length Normalized 0.200 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 mL of 5M CaCl2 or 0.1M EDTA Added In

process, and the hollow dots are the swelling process.

Figure 2, CaCl2 and EDTA Reversible Titration Curve (n = 4). This graph contains two cycles. The solid dots are for shrinking, and the hollow dots are for swelling. For the swelling part, 5 M CaCl2 was added by 0.5 ml or 1 ml increment, and

for the shrinking part, 0.1 M EDTA (at pH 7.3) was added in 0.5 ml increment. The

32 overlapped point on the graph was the transition of chelation and titration. More specifically, during the transition, the Ca2+ treated PAA hydrogel thread was taken out and placed into a new beaker for 15 minutes to reach equilibrium. Technically, there were no liquids added into the new beaker, so there are two dots plotted on the same X axis. The reason that the gel thread was transferred into another beaker was because there was an excess amount of Ca2+ inside the liquid surroundings. If the low concentration chelator solution was added directly, then the initial chelators added would bind with Ca2+ in the solution. Although there was a driving force to release the depleted Ca2+inside the gel thread to the solution once the surrounding environment is short of Ca2+, it takes time and the changes will not be evident.

Another reason to change the vessel was that the amount of Ca2+ in solution is comparatively high compared to the chelator concentration (0.1M). A high volume of chelators was required before an apparent impact of the length of the gel thread could be observed. Changing the solution made the experiment more time and cost effective.

As is proposed that the Ca2+ titration step, counterion condensation [31-33] occurred with to PAA hydrogel in 150 mM of NaCl solution via adsorption Na+ onto the negatively-charged PAA network. When Ca2+ was added, competitive counterion adsorption occurred with 2 Na+ replaced by Ca2+ within the gel. It was observed that during the titration process, the gels were opaque Ca2+ was bound.

Moreover, the Debye length may have an effect the ion exchange. Meng et al. demonstrated that the electrospun tubes with 150 mM NaCl shrank faster than the ones without salt, and above 150 mM there was no significant difference [8]. The

Debye lengths of 150 mM and 0 mM NaCl were 0.8 nm and 3.1 nm, respectively, but

33 considering the size of ions, which are 231 pm for Ca2+ and 190 nm for Na+, that is neglectable. Given that the coulomb force is inversely proportional to the distance between two charges, and if other effects were neglected, the force and acceleration attracted Ca2+ under 150 mM NaCl, according to Newton’s Second Laws of Motion,

2 푉푝표푙푦푚푒푟푉푖표푛푠푒 F = ma, would be 15X (퐹퐶표푢푙표푚푏 = 2 = 푚푎) faster than under 0 mM 4휋휀휀0푟

1 1 3.12 NaCl. (퐹 ∶ 퐹 = ∶ , thus = 15) 퐶 푢푛푑푒푟 150 푚푀 푁푎퐶푙 퐶 푢푛푑푒푟 0 푚푀 푁푎퐶푙 0.82 3.12 0.82

It was observed that the length of the gel dropped from 60% to 40% of its original length right after the first CaCl2-EDTA transition, transporting the gel to a new vessel. The shrinkage upon CaCl2titration from 100% to 60% length was due to electro-neutralization of the charges on PAA chains backbones. Without the electro repulsion forces in the network, the gels shrank to a smaller distance. For the 60% to

40%, the gel was transported to a new vessel with fresh HBS, thus the CaCl2 concentration in the surrounding became from 2.2 M Ca2+to 0 M. There would be a thermos-driven force grabbing Ca2+ away and also absorbed a little bit more liquid, where this liquid would cause an increase of entropy in the PAA hydrogel chains system for the chains to be moved more freely and have more flexibility. With these flexibilities, PAA hydrogel chains collapsed to form Gaussian random coils total conformational change. As a result, the length shrank rapidly. There was another example about this phenomena, which was protein folding. Proteins are stable with certain physiology salts concentration. If proteins were transferred from a salt-rich condition to a salt-free condition like the gel was transported to a fresh solution, they tend to denature in the salt-free condition. In a physical chemistry point of view, when absorbed salts are dragged out and solvent comes in, entropy of polypeptide increases,

34 and that induces an energy favored refolding that is called “collapse” for a polymer chain.

In the second cycle, the PAA hydrogels reached almost to 140 % original length.

When the chelator titration started, the polymer chains were packed tightly and well due to collapses. When the chelator (in the graph, EDTA) was added in, PAA chains were unwound, because the absorbed Ca2+ were chelated by EDTA gradually from the outside to inside. When the chains were free from Ca2+, they were inclined to be mixed well in solvent. Little by little, the PAA hydrogels were free from Ca2+, and the polymer chains were fully expanded, even more than the starting point.

At the starting points of the second Ca2+chelation process, due to full extension, the gels were looser than the original starting points. However, as Ca2+ was added again, the gels shrank quickly back to the packing stage due to counterion condensation. Although there was not a third cycle done, it was expected that the curve would look like the second cycle. Even though a third cycle was not completed it was expected that the curve would look like the curve of the second cycle.

The changing lengths of gels are a result of many factors: pH value, salt/ions concentration and packing history.

Figure 3, PAA hydrogel thread length variation with pH. The X axis is pH value, and the Y axis is normalized length.

The pH value has a substantial influence on PAA hydrogels. Various literatures reported that the pK푎 of PAA is around 4.2 – 4.5 depending on the molecular weight

[34, 35]. In our results, the length started to change from pH 3 to pH 6.5 at the maximum, but the rapid changes occurred between pH 4 to 5, which means charges on the PAA’s backbones increased dramatically in this range. This part is consistent with literature pK푎 values.

However, from the graph of Figure 3, we observed the decrease in length starting from pH 6.5, which was after 2 unit of pH value beyond the pK푎. Above pH 7, where the carboxylic acids in the gels are almost fully dissociated, counterion condensation prevailed for the concentration of Na+which increased due to

36 neutralization by NaOH. Counterions, in this case Na+, condensed on the negatively charged PAA backbones and cancelled the charges on backbone so that the length of the gels were decreased.

There were a few points where the lengths increased but the pH values were lowered. They were not measurement errors since each gel were measured 3 times and averaged all. These points were evident from pH 3.5 to pH 6.5. The random coil concept mentioned above could be helpful for explaining this phenomenon. With the random coil packing in the gel in the acidic surrounding, the small amount of NaOH added in would only neutralize the 퐻+ in the solution. At this point, the pH was slightly higher than the pKa, and the

PAA monomer units from the outside layers from random coil but not yet releasing hydrogen ions would proceed to dissociate. Once dissociated, those PAA monomer units would carry negative charges, and coulomb repulsion force would unwind the chains gradually. At this point, more monomers units were unwound, and more H+ were released, and the gel was extended. There is a paper supporting this [37]. Skouri et al. claimed there are a two-phase medium formed of flexible chains connecting dense regions. In these regions, some of the chains are trapped in between fixed cross-links, while others are held together. Although they argued that the interaction of hydrophobic backbones were responsible for that in my opinion. Hydrogen bonds are the cause instead. (their results are similar to mine) 37

3.2 PAA Shrinkage Ability

In past research, electrospun PAA tubes were shown to shrink to around 60% of original length upon treatment with CaCl2. The PAA hydrogel in this research, at pH

7.3, could shrink to almost the same degree as well.

2+ Figure 5, The CaCl2 Titration Curve. The hydrogel shrank as it uptakes Ca .

In Figure 5, thicker gels, with a 6 mm diameter, were used because larger gels had a higher capacity for CaCl2. 5 M CaCl2 was added by 0.5 ml increment at each point. Every time the CaCl2 solution was added, the shrinkage was noticeable even to the naked eye. When Ca2+ was absorbed, the transparent gel gradually turned progressively opaque. The cloudiness accumulated in the gel, and when the

Ca2+concentration was over 1.5 M in the solution, the entire gels turned white.

On the other hand, according to Muthukumar’s paper, the overcharge happened in multivalent ions. Microscopically, the ion bridging effect and intramolecular collapse likely occurred when the gels shrank. After a few additions of

CaCl2, the calcium stoichiometry charges were equal to the PAA, collapse happened heterogeneously, and the gel curled up a little bit. The whiteness inside the gel was the CaCl2 recharge effect in the gel. Although re-expansion was not observed, the whiteness was caused by crystallization of calcium chloride forming ordered structures inside the gel.

For data processing, each data point was categorized into twelve original groups, and those data points were normalized by its original length. The biggest and smallest values out of twelve were rejected, and the rest of the ten points were calculated into the averages and standard deviations (p < 0.05). The average are the dots plotted on the graph, and the standard deviations are the error bars.

3.3 Chelation Abilities

Figure 6, The Chelators Titration Plot.

Figure 6 shows the ability to re-swell a CaCl2-treated PAA gel for four different chelators. All four were prepared at 0.1 M for easy comparability. At room temperature, EDTA is slowly soluble in water up to 0.26 M. ATP is soluble in water to around 50 mg/ml in acidic solution and slightly higher in neutral condition [37].

Since ATP would be rapidly hydrolyzed at even higher pH with divalent ions appearance, ATP was freshly made before the experiments, and the ATP data were done under 4 ℃, and the other groups were carried out at room temperature. 0.5 ml of other chelators were added each time like Ca2+ chelation.

The results show the ranking of the chelation abilities from the best to the worst: sodium triphosphate, sodium ATP, sodium EDTA and sodium citrate. Previous research agrees with these results [41]. Changa described and compared the calcium ion binding power and binding mechanism of organic chelating agents, phosphates, organic oligomers and polymers. Sodium triphosphate, EDTA and sodium citrate

40 were also chosen in that paper. The calcium binding agent in molarity against bound calcium ions molarity were plotted, and the binding ability order was sodium triphosphate > EDTA > sodium citrate as well. Other than this, the calcium binding power (CBP), which was usually used in industry, was introduced.

bound CaCO (mg) Calcium binding power = 3 total calcium binding agent (g)

For those having high stability constants, the experimental calcium binding power should approach the theoretical data. However, for the PAA (Good-rite K-702) used in that research, have the highest CBP over all [41]. Counterion condensation, ion bridging and Gaussian random coil ideas would provide an explanation of the high PAA CBP. Ions tend to be condensed on the charged polyelectrolytes. Then, ion bridging would provide more stable associations. Furthermore, calcium ions were wrapped into the random coils of the local charges from PAA were equal to the local

Ca2+ concentration. The heterogeneous distributed Ca2+ PAA complexes were provided a better CBP.

Besides, a hypothesis for judging the chelators’ abilities in heterogeneous competitive conditions was proposed. Firstly, charges matter the most. Under physiological conditions, sodium citrate and sodium EDTA have 3 negative charges

[37], and the sodium triphosphate has 4.5 [38,39] while sodium ATP has 3.5 [37]. It was reported that the dissociation constants of five acid group of sodium triphosphate are 1.0 ; 2.2 ; 2.3 ; 5.7 ; 8.5[38] or 1.0 ; 2.2 ; 2.3 ; 3.7 ; 8.5[39]. Both sets of results gave the same conclusion of 4.5 charges.

Secondly, lone pair electrons stabilize the metal ions. For sodium EDTA, there are two secondary amine groups, and each amine group has a lone pair of electrons.

Lone pair electrons can associate with the empty orbits of metal ions and decrease the association energy.

Lastly, although sodium ATP has the most number of amine groups, the adenosine group is too massive and stiff for twisting and associated with cations. In other words, the adenosine group in ATP offers steric hindrance for chelation, which blocked the charges collaborated with other odd number charged ATP molecules

(ATP carries either 3 or 4 charges under physiological condition). Sodium triphosphate is clear and free to interact with other molecules.

3.4 Rheology

PAA hydrogel samples discussed earlier containing 0% glycerol, 25% glycerol and 50% glycerol were prepared by the procedure in the methodology part. The percentage was based on the total water in the formula, which is 7.2 of every 10 ml. In other words, 25% and 50% glycerol of samples contain 1.8 and 3.6 ml of pure glycerol in 10 ml typical sample. Glycerol is added before initiation. From figure 5, moduli were decreased as glycerol content was increased. Similar patterns could be observed on both storage and loss moduli curves. In the high frequencies region, the gel had higher moduli. On the other hand, the low frequency region, moduli decrease a little.

Frequency sweep: Storage Modulus 1000.000

100.000

10.000 Modulus (Pa) Modulus

1.000 0.010 0.100 1.000 10.000 100.000 Frequency (Hz)

0% glycerol 25% glycerol 50% glycerol

Frequency Sweep: Loss Modulus

1000

100

10 Modulus ( Pa ) Pa ( Modulus

1 0.010 0.100 1.000 10.000 100.000

0.1 Frequency ( Hz)

0% glycerol 25% glycelor 50% glycerol

Figure 7-1, PAA hydrogel frequency sweep storage modulus. Figure 5-2, PAA hydrogel frequency sweep loss modulus. Blue one is 0% glycerol content. Orange one is 25% glycerol content, and grey one is 50% glycerol content.

One of the major differences from the gels mentioned before is that the monomer acrylic acid, was neutralized before synthesis. This would be expected to affect the morphology of the gel, average radius of gyration and counterion condensation. Since there are salts from HBS, counterion condensation would discharge some of the monomers. Still there would be smaller random coils formed.

However, these random coils were important to the moduli of gels. The more random coils these gels contained, the higher the moduli. During deformation, these random coils could absorb deformation energy and dissipate the absorbed energy. That is to say, this gels were very elastic since the storage moduli were an order of magnitude higher than loss moduli.

4. Conclusion

Chemically-crosslinked PAA hydrogels are versatile materials. Firstly, the gel can shrink and swell reversibly whenever charges are neutralized. More specifically, when charges from each repeating unit repel each other, the gel swells. On the other hand, when multiply charged ions neutralize charges, the gel shrinks. That is to say, the neutralized, chemically-crosslinked PAA hydrogel has a capacity for exchanging monovalent and divalent ions with reversible dimensional changes.

ATP is a surprisingly good chelator in general under physiological conditions, even better than EDTA, although different chelators can possess different affinities for different target molecules. However, charge appears to be the most important component for judging chelation ability of Ca2+ in PAA gels. This work contributes to a growing set of design rules for the development of artificial muscles and neurons through fundamental considerations of macromolecular science.

From studies of PAA gel rheology, glycerol was identified as an effective plasticizer, as it softens PAA gel. At 50 % glycerol by weight, the modulus drops by almost an order of magnitude. PAA hydrogel is a good candidate for bio applications.

The modulus matches that of many human tissues such as brain, the elastic modulus of which is around 1k Pa [42, 43]. Moreover, PAA hydrogel can shrink and swell periodically, which makes it a good candidate for an artificial muscle. Last but not least, PAA hydrogel can quickly exchange sodium with calcium ions, making it good candidate for an artificial neuron. In short, PAA hydrogels can serve many purposes.

5. Future Work

For the chelation parts, electrospun PAA tube is a good comparison for shrinkage and expansion. The transport barrier is large for hydrogel but not for electrospun PAA tube. Besides, electrospun tube contain less acrylic acid units in their structure compared to hydrogel. With the faster reaction speed and more sensitive reaction, it is worthy for exploring electrospun PAA tube.

For the rheology part, tensile test, differential scanning calorimetry and thermogravimetric analysis are crucial for the plasticizer effect. With tensile test, the strength stress relation can be understood. With differential scanning calorimetry, the glass transition temperatures of different samples can be determined. With thermogravimetric analysis, degradation process can be known. With all the experiments above, the fact of glycerol as a PAA hydrogel plasticizer is clean and clear.

6. Bibliography: [1] Manning GS. Limiting Laws and Counterion Condensation in Polyelectrolyte Solutions. III. An Analysis Based on the Mayer Ionic Solution Theory. J Chem Phys. 1969;51(8):3249. doi:10.1063/1.1672502. [2] Manning GS. Limiting Laws and Counterion Condensation in Polyelectrolyte Solutions. III. An Analysis Based on the Mayer Ionic Solution Theory. J Chem Phys. 1969;51(8):3249. doi:10.1063/1.1672502. [3] Manning GS. Limiting Laws and Counterion Condensation in Polyelectrolyte Solutions. III. An Analysis Based on the Mayer Ionic Solution Theory. J Chem Phys. 1969;51(8):3249. doi:10.1063/1.1672502. [4] Falk M, Klann M, Reuss M, Ertl T. Visualization of signal transduction processes in the crowded environment of the cell. IEEE Pacific Vis Symp PacificVis 2009 - Proc. 2009:169-176. doi:10.1109/PACIFICVIS.2009.4906853 [5] G. E. Wnek, “Perspective: Do Macromolecules Play a Role in Nerve Stimulation and Nervous Transmission?” J. Polym. Sci. Part B: Polym. Phys., 54, 7 (2016) [6] Tjipangandjara KF, Somasundaran P. Effects of the conformation of polyacrylic acid on the dispersion-flocculation of alumina and kaolinite fines. Adv Powder Technol. 1992;3(2):119-127. doi:10.1016/S0921-8831(08)60678-0. [7] Russel, William B.; Saville, D. A.; Schowalter, William R. (1989). Colloidal Dispersions. New York: Cambridge University Press. [8] Meng L, Klinkajon W, K-hasuwan PR, Harkin S, Supaphol P, Wnek GE. Electrospun crosslinked poly(acrylic acid) fiber constructs: Towards a synthetic model of the cortical layer of nerve. Polym Int. 2015;64(1):42-48. doi:10.1002/pi.4793. [9] Tasaki I. Physiology and Electrochemistry of Nerve Fibers.; 1982. doi:10.1016/B978-0-12-683780-3.50001-3. [10] Tasaki I. On the Reversible Abrupt Structural Changes in Nerve Fibers Underlying Their Excitation. Phase Transit Cell Biol. 2008:1-21. [11] Tasaki I. Evidence for phase transition in nerve fibers, cells and synapses. Ferroelectrics. 1999;220:305-316. doi:10.1080/00150199908216221. [12] Khampieng T, Wnek GE, Supaphol P. Electrospun DOXY-h loaded-poly(acrylic acid) nanofiber mats: in vitro drug release and antibacterial properties investigation. J Biomater Sci Polym Ed. 2014;25(12):1292-1305. doi:10.1080/09205063.2014.929431. [13] Simons TJB. Calcium and neuronal function. Neurosurg Rev. 1988;11(2):119- 129. doi:10.1007/BF01794675. [14] Wilson JE, Chin A. Chelation of divalent cations by ATP, studied by titration calorimetry. Anal Biochem. 1991;193(1):16-19. doi:10.1016/0003-2697(91)90036-S. [15] Priori SG, Chen SRW. The online version of this article, along with updated information and services, is located on the World Wide Web at: 2013:871-883. doi:10.1161/CIRCRESAHA.110.226845. [16] Lodish H, Berk A, Zipursky SL, et al. Molecular Cell Biology. 4th edition. New York: W. H. Freeman; 2000. Section 18.4, Muscle: A Specialized Contractile Machine [17] Nelson, D. L., Cox, M. M., & Lehninger, A. L. (2013). Lehninger principles of biochemistry. New York: W.H. Freeman.

[18] M. Cohn, “Structural and Chemical Properties of ATP and Its Metal Complexes in Solution,” Ann. NY Acad. Sci, 603, no. 1 Biologic, pp. 151-164 (1990) [19] Tice L W. National E, Pathology I. Lead-adenosine triphosphate triphosphatase. 1969. [20] S. D. Ehrlich, Ethylenediaminetetraacetic acid, Pheonix: A.D.A.M., Inc, 2015. [21] Seely DMR, Wu P, Mills EJ. EDTA chelation therapy for cardiovascular disease: a systematic review. BMC Cardiovasc Disord. 2005;5:32. doi:10.1186/1471-2261-5- 32. [22] Bournazos S, Rennie J, Hart SP, Dransfield I. Choice of anticoagulant critically affects measurement of circulating platelet-leukocyte complexes [2]. Arterioscler Thromb Vasc Biol. 2008;28(1):2-4. doi:10.1161/ATVBAHA.107.153387. [23] Bronner F. Extracellular and intracellular regulation of calcium homeostasis. ScientificWorldJournal. 2001;1:919-925. doi:10.1100/tsw.2001.489. [24] Shahinpoor M, Kim KJ. Ionic polymer – metal composites : I . 2001;819(1):3-5. [25] Kim KJ, Shahinpoor M. Ionic polymer metal composites: II. Manufacturing techniques. Smart Mater Struct. 2003;12(1):65-79. doi:10.1088/0964-1726/12/1/308. [26] Shahinpoor M, Kim KJ. Ionic polymer–metal composites: III. Modeling and simulation as biomimetic sensors, actuators, transducers, and artificial muscles. Smart Mater Struct. 2004;13(6):1362-1388. doi:10.1088/0964-1726/13/6/009. [27]Shahinpoor M, Kim KJ. Ionic polymer–metal composites: IV. Industrial and medical applications. Smart Mater Struct. 2004;14(1):197-214. doi:10.1088/0964- 1726/14/1/020. [28] Alan Yueh-Luen Lee#, Yu-Da Chen#, Yu-Yung Chang#, Yu-Ching Lin, Chi-Fon Chang, Shing-Jong Huang, Shih-Hsiung Wu*, and Chun-Hua Hsu*. Structural Basis for DNA-mediated Allosteric Regulation Facilitated by AAA+ Module of Lon Protease. Acta Crystallogr. Sect. D, 2014 Feb;70(Pt 2):218-30. [29] Gunby RH, Ahmed S, Sottocornola R, et al. Structural Insights into the ATP Binding Pocket of the Anaplastic Lymphoma Kinase by Site-Directed Mutagenesis , Inhibitor Binding Analysis , and Homology Modeling. 2006:5759-5768. [30 ]Huang D, Zhou T, Lafleur K, Nevado C, Caflisch A. Kinase selectivity potential for inhibitors targeting the ATP binding site : a network analysis. 2010;26(2):198-204. doi:10.1093/bioinformatics/btp650. [31] Ripoll C, Norris V, Thellier M. Ion condensation and signal transduction. BioEssays. 2004;26(5):549-557. doi:10.1002/bies.20019. [32]Muthukumar M. Theory of counter-ion condensation on flexible polyelectrolytes: Adsorption mechanism. J Chem Phys. 2004;120(19):9343-9350. doi:10.1063/1.1701839. [33]Kundagrami A, Muthukumar M. Theory of competitive counterion adsorption on flexible polyelectrolytes: Divalent salts. J Chem Phys. 2008;128(24):1-16. doi:10.1063/1.2940199. [34]Cranford SW, Ortiz C, Buehler MJ. Mechanomutable properties of a PAA/PAH polyelectrolyte complex: rate dependence and ionization effects on tunable adhesion strength. Soft Matter. 2010;6(17):4175. doi:10.1039/c0sm00095g. [35] Wiśniewska M, Urban T, Grządka E, Zarko VI, Gun’ko VM. Comparison of adsorption affinity of polyacrylic acid for surfaces of mixed silica-alumina. Colloid Polym Sci. 2014;292(3):699-705. doi:10.1007/s00396-013-3103-x.

[36] Jones RE, Templeton DH. The crystal structure of acetic acid. Acta Crystallogr. 1958;11(7):484-487. doi:10.1107/S0365110X58001341. [37] Dawson, R. M. C., et al. Data for Biochemical Research, 3rd ed. Oxford University Press (New York, NY: 1986) [38] Corbridge, D. (1995). "Chapter 3: Phosphates". Studies in inorganic Chemistry vol. 20. Elsevier Science B.V. pp. 169–305. ISBN 0-444-89307-5. Retrieved January 30, 2015. – via ScienceDirect (Subscription may be required or content may be available in libraries.) [39] Holleman, A. F.; Wiberg, E. (2001), Inorganic Chemistry, San Diego: Academic Press, p. 729, ISBN 0-12-352651-5 [40] Journal I, Review CP. Review on Natural Polysaccharide Based Particulate Drug Delivery Systems : An Inimitable Tactic in Novel Drug Delivery Systems Review on Natural Polysaccharide Based Particulate Drug Delivery Systems : An Inimitable Tactic in Novel Drug Delivery. 2014;(July 2016). [41] Changa D. The binding of free calcium ions in aqueous solution using chelating agents, phosphates and poly (acrylic acid). J Am Oil Chem Soc. 1983;80(3):618-622. http://link.springer.com/article/10.1007/BF02679800. [42] Leipzig ND, Shoichet MS. The effect of substrate stiffness on adult neural stem cell behavior. Biomaterials. 2009 Dec30(36):6867-78 p.6867 right column 2nd paragraph [43] Taylor Z, Miller K. Reassessment of brain elasticity for analysis of biomechanisms of hydrocephalus. J Biomech. 2004 Aug37(8):1263-9. AND Gefen A, Margulies SS. Are in vivo and in situ brain tissues mechanically similar? J Biomech. 2004 Sep37(9):1339-52.