Topographically and Mechanically Tunable PNIPAM Scaffold Chi Chen (
ABSTRACT) Poly(N-isopropyl-acrylamide) (PNIPAM) is a thermoresponsive polymer
with
a wide
range
of
biological
applications, including drug delivery
, biosensing,
and tissue
engineering.
The
tunability of the structural and mechanical properties of PNIPAM makes it particularly at- tractive in emulating cell environments and dynamic cytoskeletal deformations. In this the- sis, PNIPAM’s properties and application in different forms i.e., solution, brushes, hydrogels, and surface patterned hydrogels, are discussed.Moreover, my collaborators and I designed lithographically patterned substrates, coated with PNIPAM films, and investigated their structural and mechanical responses to thermally driven changes in the PNIPAM hydration states. Atomic force microscopy (AFM) measurements are essential for studying soft bio- logical materials. The problems of performing AFM measurements on super soft material is discussed. In the AFM measurement on the lithographically patterned substrates, data show that the substrate pattern and coating method enable the fabrication of scaffolds with different topographic and mechanical properties across a wide thermal range. Importantly, these scaffolds exhibit variations in both lateral topography and Young’s modulus, render- ing them well suited for investigations of differential mechanical stresses experienced by cells and cell membranes. Lastly, the future work can improve the applications of this scaffold is discussed Topographically and Mechanically Tunable PNIPAM Scaffold Chi Chen (GENERAL AUDIENCE ABSTRACT) Poly(N-isopropyl-acrylamide) (PNIPAM) is a polymer can absorb and expel water at dif- ferent temperature. Many studies have been done on PNIPAM and in different form. In solution, brush, hydrogel form, PNIPAM behavior differently , but shares the thermore- sponsive ability. All the forms of PNIPAM is discussied in this thesis. Controlling the topography of a structured surface coated with PNIPAM is tested by my collaborators and I.
Atomic force microscopy (AFM) measurements
are performed
to
measure
the
topographies and
mechanical properties. The
advantage and shortage of AFM measuring soft sample is discussed. It is believed that this lithographically patterned substrates can be use to support membranes, and bend the membrane at special curvature to mimic cell environment Dedication I dedicate this thesis to my beloved parents. iv Acknowledgments I need
to thank my
adviosr
Prof
. Ashkar
for guiding me
on
my research
and helping
me
finish this thesis. Prof. Toomey and Michael Kane prepared the smaple for me. Doc. Liam taught me how to use AFM, and measured samples for me during COVID, when I cannot perform it myself. I am glad to have Teshani Kumarage’s help on the experiments and data analyze,and Hadi Rahmaninejad’s help on coding Ting’s model. Lastly, I need to thank Prof. Efremov for modifying his published code for Ting’s model, when I contact him. v Chapter 1 Poly(N-isopropylacrylamide): Prop erties and Applications Poly(N-isopropylacrylamide) (PNIPAM) is
an excellent
candidate
for a
wide
range of bio- logical and biomedical applications
, including drug delivery, artificial cartilage, and tissue engineering. This thesis describes the use of PNIPAM coatings on nanostructured substrates for the design of novel scaffolds aimed at cell membrane studies. Starting from the basic properties and applications of PNIPAM, this thesis outlines the rationale for the use of PNIPAM by considering the implications of topography and mechanical properties in the anticipated applications of these scaffolds. 1 2 CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS 1.1 Introduction Figure 1.1: Chemical structure of poly(N-isopropylacrylamide). PNIPAM contains — CONH— amide groups, propyl —CH(CH3)2— moieties, and —(CH2Ch)n— polymerization parts. (Figure is adapted from Schild et al.)[1] PNIPAM is a member of the thermoresponsive polymer family. It was first synthesized in 1956 by Sprecht [2], but its significance was mostly realized in the 1990s (see Figure 1.2). The advent of new techniques (e.g. atomic force microscopy[3]
and small-angle neutron scattering
) made
it possible to
study
the
micron-scale,
and
even nano-scale, structural and mechanical properties of such polymers. Given its versatile applications, PNIPAM became the most studied polymer of its kind in recent years, and probably the most commonly used thermoresponsive polymer. This scientific focus on PNIPAM is driven not only by the adop- tion of new techniques, but also by the virtue of PNIPAM’s special properties. PNIPAM has
a lower critical solution temperature (LCST) above which the
polymer transforms into
a
collapsed state (
from
swollen state). In particular, the hydrophilic amide groups dissolve in water below the LCST and form a strong hydrogen bonding with water. This results in polymer chains being in a swollen state. Above the LCST, the hydrophobic propyl moieties oppress the amide groups, because the hydrogen bonds are weaker, thus resulting in a col- lapsed state. This collapse manifests in a volume change, and the temperature at which 1.1. INTRODUCTION 3 this occurs is
referred to as the volume phase transition temperature (VPTT
). Both
LCST
and VPTT
of
free
PNIPAM
chains have been measured by previous studies, with typical values of 32 °C [4] and 34 °C [5], respectively. These two crucial temperatures happen to be slightly lower than the normal body temperature, which afford PNIPAM a wide use in in vivo biomedical applications. Early studies of PNIPAM were focused on the solution state and phase separation behavior. [4, 6, 7] However, with the introduction of AFM and other techniques, the measurement of mechanical properties and topographies of PNIPAM hydro- gel became possible. But despite its potential, the poor mechanical strength of PNIPAM [8] posed as an obstacle in practical applications. This chapter introduces free and constrained PNIPAM chains in solution and the basic properties of PNIPAM hydrogels. Figure 1.2: This figure shows the citation number of papers involving PNIPAM since it was first reported in 1956. PNIPAM citation numbers had a sudden increase in the beginning of 1990s due to the introduction of new characterization techniques. (Figure is adapted from Schild et al.)[1] 1.2 Properties and Applications of PNIPAM solution Studies of solution-state PNIPAM began right after the first synthesis by Wooten et al. [9]. This chapter will introduce the early history of PNIPAM, and the properties of PNIPAM solutions. 1.2. PROPERTIES AND APPLICATIONS OF PNIPAM SOLUTION 1.2.1 Synthesis of PNIPAM Chains There are multiple ways to synthesize PNIPAM chains, and three of them are widely used by previous studies:
free radical initiation in organic solution, redox initiation in aqueous media
, and ionic polymerization. Free radical initiation for PNIPAM follows standard free radical polymerization methods. Two parameters in these processes affect the molecular weight: polydispersity (PD) and chain transfer. Hoffman et al.[10] were able to change the molecular weight from 2000 to 250,000 by manipulating chain transfer. They found that the molecular weight has a significant impact on PNIPAM chain collapse. Plunkett et al.[11] reported that chain collapse decreases with decreasing molecular weight and grafting density (brush) above LCST. On the other hand , redox initiation is the other most commonly used method. As early as 1957, Wooten et al.[9] found how to control polymerization rate and avoid much greater PD, with pH value controlled by different buffers. Finally, ionic polymerization is very different from the first two methods. In 1959, Shields and Coover[12] reported their studies on ionically polymerized PNIPAM. The PNIPAM synthesized by ionic polymerization has a very weak amide group. As a result, ionically polymerized PNIPAM is insoluble, has a very high density, and a high melting point. In other words, this form of PNIPAM does not exhibit an LCST. This result was confirmed by several later publications.[13-15] 1.2.2 Properties of PNIPAM Solution Figure 1.3: The solid line shows cloud point curve of the PNIPAM solution, and the solid line with dots shows the optical density of the PNIPAM solution. (Figure is adapted from Schild et al.)[1] The LCST of PNIPAm solution is driven by entropy, which breaks the hydrogen bonds be- tween water molecules and amide groups. When the concentration of the polymer is high enough, the LCST appears as phase separation. Cloud point is a common method to de- termine the LCST of PNIPAM solutions (Figure 1.3). Heskins and Guillet[4] reported their observation in 1968. They measured the LCST by simple visual observation of phase sep- aration, which is also known as cloud point method. They determined the LCST to be 32 °C. However, this method is only effective for pure PNIPAM solutions. In the presence of additives in the solution, visual assessments of the LCST become more challenging. Despite the simplicity of this approach, it has been effectively used to measure different cloud point curves for polymers with different molecular weights. Schild and Tirrell[7] measured PNI- PAM solution with different molecular weight, salt, surfactants and comonomers. Different peak shapes of cloud point curves show the influence of these elements on the LCST, as shown in Figure 1.4. Figure 1.4: The cloud point curve peaks of PNIPAM solution with different salt shows different shapes. (Figure is adapted from Schild et al.)[7] To measure more complicated solutions and acquire more accurate measurements, other methods such as differential scanning calorimetry (DSC) have been adapted. Heskins and Guillet[4] reported measurement of released heat from breaking hydrogen bonds, signifying the collapsing process. As expected, the amount of released heat was also related to molecular weight. Schild and Tirrel[7] improved this measurement with a better DSC instrument, and the result was in excellent agreement with the result from the cloud point method. Figure 1.5: Light scattering for swollen state PNIPAM in chloroform at 25 °C. (Figure is adapted from Heskins et al.)[4] To get further information on size, shape, and even dynamics of PNIPAM chains in solution quasi-elastic light scattering (QELS) has also been used .[16] Heskins and Guillet[4] reported a 4.5-fold increase of molecular weight with this technique. Later studies were able to improve Heskins and Guillet’s measurement by applying QELS on dilute solution to get more detailed information on the transition process. Hirotsu[2][17] and Fujishige[3][18] observed the chain collapse at (34 °C) well before the onset of chain aggregation (39 °C). 1.2.3 Applications of PNIPAM Solution Although the application of PNIPAM in solution is limited, its early use with hydrophobic surfactants led to enhanced viscosity that facilitated oil recovery processes.[19, 20] This controllable viscosity property was also utilized in PNIPAM applications in increasing gloss and ink receptivity when coated on paper. Hoffman et al.[21, 22] applied the PNIPAM solution for immunoassay. PNIPAM-N-acryloxysuccinimide (NASI) chains can bond to an antibody (immunoglobulin). They mixed antibody-attached-PNIPAM-NASI chains with antigen and a different fluorophore labeled antibody (antibody2). After several heating and cooling cycles, they found that if any antibody2 exists, a fluorophore labeled layer would form. This process is similar to antigen detection of COVID-19. Figure 1.6: Immunoassay scheme for PNIPAAM. PNIPAM-N-acryloxysuccinimide (NASI) chains can bond to an antibody (immunoglobulin). (Figure is adapted from Schild et al.)[1] 1.3 Properties and Applications of PNIPAM Brushes PNIPAM brush is a special structure of PNIPAM chains. Polymer brushes are structures with one end of polymer chains tethered on a surface or interface, and the density of polymer chains are high enough that there’s not much space for polymer chains to move parallel to the tethered surface, so that
polymer chains are forced to stretch away from surface to avoid overlapping
. Considering the wide applications of polymer brushes, such as new adhesive materials[24], protein-resistant bio-surfaces[25], and preventing flocculation[26], properties of PNIPAM brushes have been tested by AFM and Surface Plasmon Resonance (SPR) by many studies. A fine AFM measurement was reported by Jones et al.[27] in 2002. They choose atom transfer radical polymerization (ATRP) to graft PNIPAM brushes onto gold substrates. Surface-confined ATRP in aqueous solvents has been proven to have following advantages compared to conventional free-radical polymerization onto a surface: quicker, more controllable, and an order of magnitude thicker. Figure 1.7: AFM image of PNIPAM brushes on hexadecanethiol (HDT) sample, prepared with surface confined ATRP method. Between the PNIPAM brushes is the HDT layer on Au substrate. (Figure is adapted from Jones et al.)[27] In this experiment, two samples with the structure shown in figure 1.8 were measured in water. The brush height of the first sample decreased from 46 nm to 11 nm with a tem- perature increase from 25 °C to 35 °C, and the second sample decreased from 29 nm to 11 nm
when the temperature was
raised
from 25 °C to 40 °C
. This process was repeatable and reversible. Adhesion force was measured with force-curves of the whole surface. At 19 °C, no adhesion force was detected for PNIPAM, but there’s a 3-4 nN adhesion force when measured on hexadecanethiol (HDT) domains. At 40 °C, the 3-4 nN adhesion force of hexadecanethiol remained unchanged, and a new adhesion force of 8-9 nN appeared, which belongs to PNIPAM brushes. Figure 1.8: (a) The peak at 3-4nN, indicating the adhesion force of HDT at 19 °C. (b) At 40 °C, other than the old peak belongs to HDT, a new peak, representing adhesion force of PNIPAM, around 8-9nN appears. (c) Solid line is one of the unloading force-distance curves from AFM indentation at 19°C. The dash line is one of the unloading force curves at 40°C, where there’s a huge ditch. The huge ditch is a representation of adhesion force. (Figure is adapted from Jones et al.)[27] In a later study, Balamurugan et al.[28] used Surface Plasmon Resonance (SPR) to measure the topography change of PNIPAM brushes as a function of time. SPR is a technique that measures the adsorption of material on a planar surface by stimulating resonant oscillation of conduction electrons at the interface with incident light. Their measurement showed a nonlinear transformation, with a sharp transition at 32 °C, which matches the LCST of PNIPAM in solution. However, due to the rapid heating and cooling rate ( 4.5 °C/min) it is possible that the results in this study have a delay in topographic changes at the reported temperatures. Figure 1.9: The advancing water contact angle, related to wettability, shows a shape tran- sition at 32 °C. (Figure is adapted from Balamurugan et al.)[28] 1.4 Properties and Applications of PNIPAM Microgels PNIPAM hydrogels are crosslinked PNIPAM chains. Crosslinking gives PNIPAM hydrogels 3-D structures, which lead to measurable topographies and mechanical properties. These two properties can be tuned by crosslinking and other methods . With novel techniques, measurement on super soft hydrogel have become possible, and it caused a remarkable boom in PNIPAM focused studies. This section will introduce microgels, one of the simplest forms of PNIPAM hydrogels 1.4.1 Properties of Hydrogel Hydrogels have the following characteristics: loosely cross-linked, hydrophilic, usually vis- coelastic, and able to swell by a large ratio. Fundamentally, hydrogels can absorb a huge amount of water, while increasing their weight by several orders. One of the most important properties of hydrogels is that despite their highly absorbent nature, they are insoluble in water because of the 3D crosslinked structures. 1.4.2 Polymer Crosslinking Crosslinking of hydrogels has several effects on various properties of the hydrogel. Firstly, elasticity is affected by crosslink density. Hydrogels become more rigid with increasing crosslink density. Viscosity and elasticity are gradually lost during this process, and hydro- gels might become brittle. Secondly, high crosslink density results in a higher glass transition temperature, which leads to higher strength and toughness. Thirdly, the melting point of hy- drogel is inversely proportional to crosslink density. Finally, the crosslink density can impose strong constraints on shape changes of hydrogel structures; i.e. hydrogels with high crosslink density cannot swell and collapse as readily as their low crosslink density analogs. There’s two ways to synthesize crosslinked hydrogels: chemical crosslinking and physical crosslink- ing. Physical crosslinks are not permanent and reversible, which form by hydrogen bonding, hydrophobic interaction, entangled chains or crystallite formation. To synthesize physical crosslinks, ionic interactions could be one option, and it can be done under normal room con- ditions.[1] Stereo-complex formation, hydrophobic interactions, crystallization and protein interactions are several other ways to synthesize physical crosslinks. Chemical crosslinks are permanent, formed by covalent bonds. Chain-growth polymerization, specifically
free radical polymerization, is one of the most commonly used
chemical crosslink techniques
for
synthesizing
PNIPAM
microgels. Anionic and cationic polymerization can be employed in chain-growth polymerization. Addition and condensation polymerization can take place in water. Gamma and electron beam polymerization don’t need additional crosslinkers, EM irradiation replaces the function of traditional crosslinkers. The method used for crosslink- ing the samples in this thesis is UV beam crosslinking, which has a similar mechanism as gamma and electron beam polymerization.[1] Crosslinking agents, comonomers, and degree of crosslinking all have a huge impact on hydrogel mechanical strength, as well as other properties. For example, crosslinking agents affect the swelling ratio, and the degree of crosslinking.[1] For the hydrogels used in my study, MaBP is adapted as crosslinkers. 1.4.3 PNIPAM Microgels Microgels are droplets, formed by hydrogels, which have a radius around 1 micrometer. PNIPAM microgels are relatively easy to make and reproduce by spin coating on to substrate. The size of microgels is suitable for AFM and Dynamic-light scattering (DLS) measurements, and is applicable as drug delivery systems and novel optical materials. Tagit et al.[29] tested microgels with PNIPAM hydrogels synthesized using surfactant-free radical polymerization with crosslinker N,N’-methylenebis-acrylamide (BIS). The measurement started from 25 °C and ended at 40 °C. DLS and UV/Vis absorption measurements were adopted for measuring the size and shape of the microgels. AFM was used
to image the
whole
sample and acquire force-distance curves. The
microgels were observed transforming from almost flat ellipses to spheres with rising temperature. The young’s modulus obtained from Oliver-Pharr model[30] varied from 1.8 MPa at 25 °C to 12.8 MPa at 40 °C. Schmidt et al.[31] published a very similar study in which they used the same synthesis method, but controlled BIS at 10% and 2%, and spin coated it onto a flat silicon substrate. Their AFM measurements showed a similar transformation from almost flat ellipses to spheres. They further reported that microgels with higher BIS concentration maintained a more stereoscopic shape below LCST, indicating that the higher crosslink concentration resulted in a more rigid hydrogel. Figure 1.10: AFM images of the microgels obtained by spin-coating, with mostly uniform arrangement. (Figure is adapted from Schmidt et al.)[31] Scherzinger et al.[32] studied the influence of different solvents on PNIPAM microgels and PNIPAM solutions in swollen states. They found that both PNIPAM chains and PNIPAM microgels were swollen in both water and methanol below the LCST. However, both PNIPAM chains and PNIPAM microgels were in collapsed states even below the LCST in mixed water/methanol solvent, known as cononsolvency. They also reported that the crosslink density barely affected the cononsolvency of PNIPAM microgels, but significantly affected PNIPAM chains. Figure 1.11: PNIPAM microgel cononsolvency effect. PNIPAM microgels are in a swollen state under the LCST in both water and methanol, but they are in a collapsed state in water- methanol mixture under the LCST. (Figure is adapted from Scherzinger et al.)[33] 1.4.4 Applications of PNIPAM Microgels One major application of PNIPAM hydrogels is drug delivery.[34-36] The LCST of PNIPAM lies between room temperature and body temperature, so that swollen state PNIPAM hydro- gel drug carriers will start collapsing as soon as they enter the body. However, this system has some shortcomings. Pure PNIPAM hydrogels have very slow response rates, such that by the time PNIPAM hydrogel drug carriers fully collapse, they would have already exited the blood circulation stream. In order to improve targeted drug delivery, interpenetrating polymer networks (IPN) or simple crosslinking with other types of polymers can significantly increase the collapse rate. As mentioned above, several PNIPAM copolymers are able to fully collapse within 30 minutes. At this response rate, releasing drugs when the carriers reach the stomach and intestines become possible. Moreover, a combination of hydrophobic gels and ionic gel acrylic acids are usually used to form IPN or crosslink with PNIPAM to avoid drugs being released in low PH body parts (stomach).[37] However, there is a restriction on the cytotoxicity of the other polymers, which crosslink with PNIPAM. On the other hand, if the other polymer is biodegradable, it is possible to biodegrade the whole carrier. Oral delivery of insulin and delivery of calcitonin are examples of well-studied PNIPAM drug carriers.[37-40] Figure 1.12: (a) The process of PNIPAM hydrogel drug carrier release drug. (b) A biodegrad- able carrier biodegrade itself while releasing drugs. (Figure is adapted from Ashraf et al.)[41] Another application of PNIPAM hydrogels is its use in cell cultures. The traditional way of growing cell cultures is on a sheet of PNIPAM microgel. Schmidt et al.[31] were the first to introduce this method, which was later adopted by other groups.[44-46] The cells are cultured onto the microgel sheet above LCST. Cells are more adhesive at higher temperatures and the attachment of cells to the hydrophobic sheet surfaces is favored. The harvesting is realized in the swollen state with less adhesive interactions between the cells and the hydrophilic microgel sheet, and a whole cell sheet can be harvested. Their microgel is swelling from 10 kPa to 100kPa. However, microgel’s Young’s modulus varies by a large amount from the top of microgel to the side of microgel.[29] 1.5. CONCLUSIONS Figure 1.13: (Top) Cell culture on PNIPAM microgel film and harvest below the LCST. (Bottom) A special AFM method of measuring topographies of cells. (Figure is adapted from Kim et al.)[44] 1.5 Conclusions In this chapter, the basic properties and applications of PNIPAM are introduced. From the 32 °C LCST shared by most PNIPAM copolymers, the potential of PNIPAM on biological application is obvious. The applications presented in this chapter are confined within PNI- PAM microgel, but crosslinking with different polymers and supported by various substrates leaves huge probability to extend the application of PNIPAM. The following two chapters will provide a more detailed introduction on topographies and mechanical properties of novel applications of PNIPAM. 1.6 References 1. Schild, H.G., Poly (N-isopropylacrylamide): experiment, theory and application. Progress in polymer science, 1992. 17(2): p. 163-249. 2. Sprecht, E., A. Neuman, and H. Neher, US Pat. 2,773,063. Rohm Haas, 1956. 3. Binnig, G., C.F. Quate, and C. Gerber, Atomic force microscope. Physical review letters, 1986. 56(9): p. 930. 4. Heskins, M. and J.E. Guillet, Solution properties of poly (N-isopropylacrylamide). Journal of Macromolecular Science—Chemistry, 1968. 2(8): p. 1441-1455. 5. Kumar, A., et al., Smart polymers: Physical forms and bioengineering applications. Progress in Polymer Science, 2007. 32(10): p. 1205-1237. 6. Fujishige, S., K. Kubota, and I. Ando, Phase transition of aqueous solutions of poly (N-isopropylacrylamide) and poly (N-isopropylmethacrylamide). The Journal of Physical Chemistry, 1989. 93(8): p. 3311-3313. 7. Schild, H.G. and D.A. Tirrell, Microcalorimetric detection of lower critical solution tem- peratures in aqueous polymer solutions. Journal of Physical Chemistry, 1990. 94(10): p. 4352-4356. 8. Takigawa, T., et al., Change in Young’s modulus of poly (N-isopropylacrylamide) gels by volume phase transition. Polymer Gels and Networks, 1998. 5(6): p. 585-589. 9. Wooten, W., R. Blanton, and H. Coover Jr, Effect of pH on homopolymerization of N‐isopropylacrylamide. Journal of Polymer Science, 1957. 25(111): p. 403-412. 1.6. REFERENCES 10. Dong, L.C. and A.S. Hoffman, Thermally reversible hydrogels: III. Immobilization of enzymes for feedback reaction control. Journal of controlled release, 1986. 4(3): p. 223-227. 11. Plunkett, K.N., et al., PNIPAM chain collapse depends on the molecular weight and grafting density. Langmuir, 2006. 22(9): p. 4259-4266. 12. Coover, H., Crystalline poly (N-isopropylacrylamide). Journal of Polymer Science, 1959. 39(135): p. 532-533. 13. Kennedy, J. and T. Otsu, Hydrogen transfer polymerization with anionic catalysts and the problem of anionic isomerization polymerization. Journal of Macromolecular Science— Reviews in Macromolecular Chemistry, 1972. 6(2): p. 237-283. 14. Hirotsu, S., Y. Hirokawa, and T. Tanaka, Volume‐phase transitions of ionized N‐isopropylacrylamide gels. The Journal of chemical physics, 1987. 87(2): p. 1392-1395. 15. Hirotsu, S., Phase transition of a polymer gel in pure and mixed solvent media. Journal of the Physical Society of Japan, 1987. 56(1): p. 233-242. 16. Taylor, L.D. and L.D. Cerankowski, Preparation of films exhibiting a balanced temper- ature dependence to permeation by aqueous solutions—a study of lower consolute behavior. Journal of Polymer Science: Polymer Chemistry Edition, 1975. 13(11): p. 2551-2570. 17. Sato, K., et al., Effects of deformation induced phase transformation and twinning on the mechanical properties of austenitic Fe–Mn–Al alloys. ISIJ international, 1989. 29(10): p. 868-877. 18. Kubota, K., S. Fujishige, and I. Ando, Single-chain transition of poly (N-isopropylacrylamide) in water. Journal of Physical Chemistry, 1990. 94(12): p. 5154-5158. 19. Siano, D., et al., Thermodynamics and hydrodynamics of a nonionic microemulsion. Colloids and surfaces, 1987. 26: p. 171-190. 24 20. Blackwell, J. and M. Nagarajan, Conformational analysis of poly (MDI-butandiol) hard segment in polyurethane elastomers. Polymer, 1981. 22(2): p. 202-208. 21. Monji, N. and A.S. Hoffman, A novel immunoassay system and bioseparation process based on thermal phase separating polymers. Applied biochemistry and biotechnology, 1987. 14(2): p. 107-120. 22. Cole, C.-A., et al., N-Isopropylacrylamide and N-acryloxysuccinimide copolymer: a thermally reversible, water-soluble, activated polymer for protein conjugation. 1987, ACS Publications. 23. Zhao, B. and W.J. Brittain, Polymer brushes: surface-immobilized macromolecules. Progress in Polymer Science, 2000. 25(5): p. 677-710. 24. Raphael, E. and P. De Gennes, Rubber-rubber adhesion with connector molecules. The Journal of Physical Chemistry, 1992. 96(10): p. 4002-4007. 25. Amiji, M. and K. Park, Surface modification of polymeric biomaterials with poly (ethy- lene oxide), albumin, and heparin for reduced thrombogenicity. Journal of Biomaterials Science, Polymer Edition, 1993. 4(3): p. 217-234. 26. Clayfield, E., E. Lumb, and P. Mackey, Retarded dispersion forces in colloidal particles— exact integration of the Casimir and Polder equation. Journal of Colloid and Interface Science, 1971. 37(2): p. 382-389. 27. Jones, D.M., et al., Variable adhesion of micropatterned thermoresponsive polymer brushes: AFM investigations of poly (N‐isopropylacrylamide) brushes prepared by sur- face‐initiated polymerizations. Advanced Materials, 2002. 14(16): p. 1130-1134. 28. Balamurugan, S., et al., Thermal response of poly (N-isopropylacrylamide) brushes probed by surface plasmon resonance. Langmuir, 2003. 19(7): p. 2545-2549. 1.6. REFERENCES 29. Tagit, O., N. Tomczak, and G.J. Vancso, Probing the morphology and nanoscale mechan- ics of single poly(N-isopropylacrylamide) microgels across the lower-critical-solution temper- ature by atomic force microscopy. Small, 2008. 4(1): p. 119-26. 30. Oliver, W.C. and G.M. Pharr, An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. Journal of Materials Research, 1992. 7(6): p. 1564-1583. 31. Schmidt, S., et al., Adhesion and Mechanical Properties of PNIPAM Microgel Films and Their Potential Use as Switchable Cell Culture Substrates. Advanced Functional Materials, 2010. 20(19): p. 3235-3243. 32. Scherzinger, C., et al., Cononsolvency of poly-N-isopropyl acrylamide (PNIPAM): Mi- crogels versus linear chains and macrogels. Current opinion in colloid interface science, 2014. 19(2): p. 84-94. 33. Vafek, O., J.M. Murray, and V. Cvetkovic, Superconductivity on the brink of spin-charge order in a doped honeycomb bilayer. Physical review letters, 2014. 112(14): p. 147002. 34. Okuyama, Y., et al., Swelling controlled zero order and sigmoidal drug release from thermo-responsive poly(N-isopropylacrylamide-co-butyl methacrylate) hydrogel. Journal of Biomaterials Science, Polymer Edition, 1993. 4(5): p. 545-556. 35. Jones, D.S., et al., Characterization of the physicochemical, antimicrobial, and drug release properties of thermoresponsive hydrogel copolymers designed for medical device ap- plications. Journal of Biomedical Materials Research Part B: Applied Biomaterials: An Official Journal of The Society for Biomaterials, The Japanese Society for Biomaterials, and The Australian Society for Biomaterials and the Korean Society for Biomaterials, 2008. 85(2): p. 417-426. 26 36. Coughlan, D. and O. Corrigan, Drug–polymer interactions and their effect on ther- moresponsive poly (N-isopropylacrylamide) drug delivery systems. International Journal of Pharmaceutics, 2006. 313(1-2): p. 163-174. 37. Kim, S. and H. Jacobs, Self-regulated insulin delivery-Artificial pancreas. Drug devel- opment and industrial pharmacy, 1994. 20(4): p. 575-580. 38. Serres, A., M. Baudyš, and S.W. Kim, Temperature and pH-sensitive polymers for human calcitonin delivery. Pharmaceutical research, 1996. 13(2): p. 196-201. 39. Ramkissoon-Ganorkar, C., et al., Effect of molecular weight and polydispersity on kinetics of dissolution and release from pH/temperature-sensitive polymers. Journal of Bio- materials Science, Polymer Edition, 1999. 10(10): p. 1149-1161. 40. Schmaljohann, D., Thermo-and pH-responsive polymers in drug delivery. Advanced drug delivery reviews, 2006. 58(15): p. 1655-1670. 41. Ashraf, S., et al., Snapshot of phase transition in thermoresponsive hydrogel PNIPAM: Role in drug delivery and tissue engineering. Macromolecular Research, 2016. 24(4): p. 297-304. 42. Akimoto, J., et al., Thermally controlled intracellular uptake system of polymeric mi- celles possessing poly (N-isopropylacrylamide)-based outer coronas. Molecular pharmaceu- tics, 2010. 7(4): p. 926-935. 43. Loh, X.J., et al., Biodegradable thermogelling poly (ester urethane) s consisting of poly (lactic acid)–thermodynamics of micellization and hydrolytic degradation. Biomaterials, 2008. 29(14): p. 2164-2172. 44. Kim, Y.-J. and Y.T. Matsunaga, Thermo-responsive polymers and their application as smart biomaterials. Journal of Materials Chemistry B, 2017. 5(23): p. 4307-4321. 1.6. REFERENCES 27 45. Choi, A., et al., Rapid harvesting of stem cell sheets by thermoresponsive bulk poly (N-isopropylacrylamide)(PNIPAAm) nanotopography. Biomaterials Science, 2020. 8(19): p. 5260-5270. 46. Tsai, H.-Y., et al., Two-dimensional patterns of poly (N-isopropylacrylamide) microgels to spatially control fibroblast adhesion and temperature-responsive detachment. Langmuir, 2013. 29(39): p. 12183-12193. Chapter 2 Surface Top ography and Biological Applications of Hydrogels To explore further applications, various topographies of PNIPAM hydrogel have been de- veloped in recent years using different approaches. This chapter will discuss the different topographies of PNIPAM hydrogel and their applications. 2.1 Surface Patterning One of the uses of PNIPAM is in switchable surfaces in lab-on chip devices to control the flow through channels.[1-3] It has also been used for the retention and release of particles passing through the channel. A derivative of this function is PNIPAM hydrogel pillar patterns which exhibit “capture and release” function. Castellanos et al.[1] fabricated PNIPAM hydrogel monolith patterns onto silicon substrates with soft lithography. PDMS micro-mold with rectangular capillaries are filled with PNIPAM solution, then exposed under UV light for 8min. The PNIPAM hydrogel monoliths are the same size as the rectangular capillaries: 40um in height, 20um in width, 20um apart, and fluorescence is used to confirm no PNIPAM left in between (Figure 2.1). A sample with smaller monoliths is made, and it showed opposite behavior then the other one. The width of monoliths of the first sample swells in lateral direction from 12um to 20um (Figure 2.1), whereas the second sample doesn’t swell in this 28 2.1. SURFACE PATTERNING direction. The second sample swells in vertical direction from 1.3um to 2.3 um. The monolith pattern is proved to be very stable after it undergoes a temperature cycles for one week, but delamination on the edge is observed. Lastly, the response time is faster than unconfined PNIPAM hydrogels. Figure 2.1: (left) Fluorescent image of PNIPAM hydrogel monolith pattern on substrate. (right) Bright-field microscopy images of PNIPAM hydrogel monoliths at 25°C and 40°C, with monolith width 20.7um and 12.2um. (Figure is adapted from Castellanos et al.)[1] To test the gating behavior of the formed channels, 6 and 20 m diameter polystyrene micro- spheres were applied to the surface.
At 25°C, i.e. in the swollen state of
PNIPAM, none of
the
microspheres was able to enter the 3.2 um trenches; At 40°C, i.e. above the LCST, 6um microspheres were able to enter the 12um trenches. Moreover, the hydrophobicity increases the adhesion of the monoliths, so that 6um microspheres are more likely attaching to the monoliths. Lowering the temperature back to 25 °C, 6um microspheres were captured in the trenches, and their position was fixed. This “capture and release” mechanism can be used to sort particles with different sizes. However, the releasing of captured particles depends on diffusion, which takes a long time. CHAPTER 2. SURFACE TOPOGRAPHY AND BIOLOGICAL APPLICATIONS OF HYDROGELS Figure 2.2: (left) A diagram showing PNIPAM hydrogel monoliths swelling with tempera- ture. The expanded head part at 25°C will capture particles between the monoliths. (right) Bright-field microscopy image of PNIPAM hydrogel monoliths showing the capture of 6um microspheres at 25°C. The width of PNIPAM hydrogel monoliths is 20um .(Figure is adapted from Castellanos et al.)[1] In another study, DuPont Jr. et al. [2] reported on the instability of this structure. The preparation of the sample followed the same procedure as the previous study. They found the monoliths attached to the surface swell by a factor of 2.7 in height and 2.6 in width compared to unconstrained monoliths which swell by a factor of 2 in both dimensions. Edge buckling appears with the swelling behavior, which is expected to result in instabilities when exceeding a limit.[3]
The wavelength of the edge instability is 69
um, smaller than
the width of
monoliths (85um) in
dry
state. Fluorescence measurements show the instability is mostly towards the outer edge of the monoliths, opposite to the prediction from linear theory. 2.1. SURFACE PATTERNING Figure 2.3: (top) The edge buckling effect on the monoliths. (bottom) Localized fluorescence quenching test shows the instability is most towards the outer edge. (Figure is adapted from DuPont et al.)[2] For narrower monoliths, a bulk buckling instability replaced the edge buckling effect.[2] Unlike the edge buckling effect, bulk buckling instability is affected by the height of the monoliths. More et al.[4] predicted the wavelength of the instability in between swollen state and collapsed state. DuPont Jr. et al. suggested that this instability highly depended on crosslink density, where lower crosslink density can delay the instability in vertical direction. The edge buckling is also affected by the history of the monoliths. For example, if the monoliths were soaked in high concentration NaCl, the wavelength of the edge buckling would be shorter. Figure 2.4: Different view of the bulk buckling instability effect for narrower monoliths. (Figure is adapted from DuPont et al.)[2] 2.2 Cell Culture In the previous chapter, several publications using microgel sheets to culture cells have been introduced. Structured PNIPAM hydrogels, with features similar to periodic PNIPAM monoliths, can be used for cell culture, and they usually come with special functions. Tsai et al.[5] coated PNIPAM hydrogels on to PS surface with dip coating method, and formed peri- odic strips covered with PNIPAM microgels. Dip coating is able to uniformly coat hydrogel on to substrates and control the thickness of the coated layer by varying the retraction speed, as shown in Figure 2.5. Compared to spin coating, which is widely used for coating non- structured surfaces, dip coating generates more uniform films. Tsai et al. coated strips of 2.2. CELL CULTURE PNIPAM microgels at slow speed 50.9nm/min (Figure 2.5B), and the spacing between strips at a higher speed of 90um/min (Figure 2.5C). To achieve three combos with 50um/100um strips and spacings, the withdrawal distance (Figure 2.5A) is controlled to be 50um/100um at lower speed. An advantage of PNIPAM microgel is that the deformation led to greater contract area and van der Waals force, in comparison, the stiffer PS microgel could not be coated onto this surface. Figure 2.5: Different view of the bulk buckling instability effect for narrower monoliths. (Figure is adapted from Tsai et al.)[5] The diameter of the microgel particles swells across the LCST from 450nm at 25°C to 250nm at 40°C. AFM images confirmed that all samples are monolayers, and the height swells from 80nm to 120nm at 37°C. The degree of swelling is in the same range as PNIPAM microgels that are uniformly coated onto a surface.[6] The VTTP is at 31.3°C, and it is in good agreement with previous reported value.[7] The strip-spacing structures were measured with optical microscopy,. The 50um/50um (strip/spacing) sample showed a wider strip and a thinner spacing. However, this 50um/50um sample proved to be very stable after experiencing a cooling down cycle for three days. Figure 2.6: Differential interference contrast microscopy of 50um/50um sample (A, C)
at 37°C for 3 days, (B, D) after three temperature
circles
of
23h
at 37°C
and 1h
at 25°C
. The sample is very stable after swelling. (Figure is adapted from Tsai et al.)[5] These PNIPAM patterned substrates were then used to seed NIH3T3 fibroblast cells . No special behavior was observed for substrates with strip/spacing ratios of 50/100 and 100/100, but cells seeded onto the 50/50 substrate showed highly ordered alignment with elongated fibroblasts aligned along the strips.[5] The strips of cells are thinner than PNIPAM strips. When culturing cells on to 100/50 and 350/50 substrates, the cell strips are consistently close to 50um, which shows the cell strips are attached to the spacing, not PNIPAM strip.[5] This was further confirmed for cell spacings that are 100um and 350um. This phenomena is explained by hydrophobicity, which is correlated with cell adhesion, and it is responsible for the releasing mechanism when cells are seeded on PNIPAM. 2.2. CELL CULTURE Figure 2.7: (A, C) DIC image of 100/50 sample and cell strips. (B, D) DIC image of 200/50 sample and cell strips. The cell strips are both 50um, but the spacing is wider for the 200/50 sample. (Figure is adapted from Tsai et al.)[5] The cell alignments and cell strip patterns will start gradually disappear after 24h, and most features totally disappear after 72h. The cell spread on the surface uniformly after 72h. The detachment of the cells starts at 3 min after temperature is lowered to 25°C, and the complete detachment ends at 5 min. [5] The detachment speed is much faster than the response rate of normal PNIPAM hydrogels. [8] Lastly, the PNIPAM microgel strips remain unchanged after cell detachment. In comparison, Zhernenkov et al. [9] tested a scaffold with a similar structure using a lithographically patterned silicon substrate coated with a crosslinked film of PNIPAM hydrogel. The scaffolds were prepared by spin coating PNIPAM solution onto the substrate. The coated layer forms a continuous film that is then crosslinked with UV-light to form a cohesive network. Neutron scattering measurements and specular reflectivity fits are used to examine the structure and the PNIPAM hydrogel layer. D2O is used for clear background in neutron scattering measurements.[10] Figure 2.8 shows the thickness of each layer above and below LCST. At 300K, PNIPAM hydrogels fill the groove of the grating. At 315K, the thickness of PNIPAM hydrogels at the bottom of the groove is 57nm, and the PNIPAM layer on the top is 34nm.[9] The sample undergoes several temperature cycles and keeps a reproducible structure. Figure 2.8: The table shows the thickness, NSLD, and roughness of the bare grating, and the coated substrate above and below LCST. (a) is the NR data of bare grating, and the coated substrate above and below LCST. (b) shows the comparison of NSLD and the real structure. (Figure is adapted from Zhernenkov et al.)[9] This model assumes the PNIPAM hydrogels attach to the side wall. Zhernenkov et al. used Bragg scattering with dynamical theory to determine the PNIPAM layer on the side wall to complete the model. The dynamical theory calculation from the Bragg scattering agrees with NR measurement, showing that at 300K that PNIPAM hydrogels completely fill the groove. At 315K, the PNIPAM layer on the side wall is calculated to be 30nm this result is later confirmed by AFM measurement.[9] 2.2. CELL CULTURE Figure 2.9: (A) Detector image of the first order Bragg ridge. (B) Off-specular data of the first Bragg ridge. (C) AFM image of the coated substrate. (Figure is adapted from Zhernenkov et al.)[9] 3T3 mouse embryonic fibroblast cells are seeded onto the scaffold at 37°C. The cell align- ments are mostly along the groove direction, confirming the alignment capability of periodic structures observed earlier by Tsai et al. Figure 2.10: Cis–trans isomerisation of azobenzene used for the manufacturing of a periodic scaffold. The topography of this scaffold is controlled by UV light and visible light, and it is a stable structure under light circles. (Figure is adapted from Liu et al.)[11] Other stimuli polymers can also form periodic scaffolds for various applications. Liu et al.[11] tested a periodic scaffold coated with azobenzene crosslinked polymer. Polymer with azobenzene crosslinker experiences cis–trans isomerisation with UV light. 2.3. NANOPILLARS 2.3 Nanopillars Sanz et al.[12] synthesized PNIPAM nanopillar arrays in 2017. Both PNIPAM and PNIPAM nanocomposite with acrylamide monomers (AAm) were filled in an anodized aluminum oxide (AAO) nanoreactor, and synthesized by ATRP method (Figure w.11). The size of the nanopillars was confirmed by AFM to be the same as the AAO reactor. The
Young’s modulus of the
PNIPAM nanopillar arrays
was measured to be
0.88
MPa
below LCST and 3.0 MPa above LCST; PNIPAM-AAm nanopillar arrays had a Young’s modulus of 8.4 MPa below LCST and 3.0 MPa above LCST. The crosslinking enhanced the mechanical properties of the nanopillars below LCST, as expected, but it reversed the transition above LCST. With MD simulations and contact angles, they speculated that the hydrophilic AAm was being pushed to the surface of nanopillars, resulting in an increase in hydrophilicity and subsequent softening of the nanopillars. Figure 2.11: Schematic of the synthesis of PNIPAM nanopillars and the procedure to take PNIPAM nanopillars out of AAO nanoreactors and use them for surface patterning. (Figure is adapted from Sanz et al.)[12] Giussi et al.[13] modified the PNIPAM nanopillar by replacing PNIPAM-AAm with PNIPAM- Fe3O4 hydrogel. Most properties of PNIPAM-Fe3O4 nanopillars are similar to PNIPAM- AAm nanopillars. The difference shows in their mechanical properties; the PNIPAM-Fe3O4 nanopillars had a Young’s modulus of 1.2 MPa below LCST and 8.6 MPa above LCST, opposite to the behavior of PNIPAM-AAm nanopillars. More interestingly, the existence of Fe3O4 gives the nanopillars magnetization in the perpendicular direction proportional to the total number of nanoparticles in the gel. This novel structure hasn’t been applied to practical use yet. 2.4. CONCLUSIONS 2.4 Conclusions In this chapter, four different topographies and their applications were introduced: the swelling behavior of PNIPAM stripes and their use in “robotic arm” applications; the periodic coating of PNIPAM hydrogels and its cell culture applications;PNIPAM gating in lab-on- chip devices; and PNIPAM nanopillars formed in AAO reactors. The coating of PNIPAM hydrogels on periodic nanopatterned substrates will be introduced in detail in chapter 5, and its properties will be explored for its potential use as a scaffold for studies of model cell membranes that mimic the complex curvatures of living cells. 2.5 References 1. Castellanos, A., et al., Size-Exclusion “Capture and Release” Separations Using Surface- Patterned Poly(N-isopropylacrylamide) Hydrogels. Langmuir, 2007. 23(11): p. 6391-6395. 2. DuPont Jr, S.J., et al., Swelling-induced instabilities in microscale, surface-confined poly(N-isopropylacryamide) hydrogels. Soft Matter, 2010. 6(16): p. 3876-3882. 3. Biot, M., Applied Science Research. Series A, 1963. 12: p. 168-182. 4. Mora, T. and A. Boudaoud, Buckling of swelling gels. The European Physical Journal E, 2006. 20(2): p. 119-124. 5. Tsai, H.-Y., et al., Two-dimensional patterns of poly (N-isopropylacrylamide) microgels to spatially control fibroblast adhesion and temperature-responsive detachment. Langmuir, 2013. 29(39): p. 12183-12193. 6. Schmidt, S., et al., Adhesion and Mechanical Properties of PNIPAM Microgel Films and Their Potential Use as Switchable Cell Culture Substrates. Advanced Functional Materials, 42 2010. 20(19): p. 3235-3243. 7. Schild, H.G., Poly (N-isopropylacrylamide): experiment, theory and application. Progress in polymer science, 1992. 17(2): p. 163-249. 8. Nash, M.E., et al., Synthesis and characterization of a novel thermoresponsive copolymer series and their application in cell and cell sheet regeneration. Journal of Biomaterials Science, Polymer Edition, 2013. 24(3): p. 253-268. 9. Zhernenkov, M., et al., Thermoresponsive PNIPAM Coatings on Nanostructured Gratings for Cell Alignment and Release. ACS Applied Materials Interfaces, 2015. 7(22): p. 11857- 11862. 10. Milewska, A., J. Szydlowski, and L.P. Rebelo, Viscosity and ultrasonic studies of poly (N‐isopropylacrylamide)–water solutions. Journal of Polymer Science Part B: Poly- mer Physics, 2003. 41(11): p. 1219-1233. 11. Liu, D., et al., Photo‐switchable surface topologies in chiral nematic coatings. Ange- wandte Chemie International Edition, 2012. 51(4): p. 892-896. 12. Sanz, B., et al., Thermally-induced softening of PNIPAm-based nanopillar arrays. Soft Matter, 2017. 13(13): p. 2453-2464. 13. Giussi, J.M., et al., Thermo-responsive PNIPAm nanopillars displaying amplified respon- siveness through the incorporation of nanoparticles. Nanoscale, 2018. 10(3): p. 1189-1195. Chapter 3 Mechanical Prop erties of PNIPAM Hydrogels This chapter focuses on the mechanical properties of PNIPAM hydrogels and recent ap- proaches to improve these properties. Native PNIPAM hydrogels have very poor mechanical properties, which significantly limited
its application in tissue engineering, drug delivery
, and cell culture.
However, the
absolute value
of
Young’s modulus can vary dramatically de- pending on the measurement mode and the models used in the data analysis. Therefore, this chapter won’t focus on the direct comparison of the absolute value of the Young’s modulus, but will focus on introducing various strategies of improving mechanical properties. 3.1 Native PNIPAM Hydrogels Several factors affect the mechanical properties of PNIPAM, including the molecular weight, crosslink density, crosslink agent, polymerization, and measurement temperature. Crosslink- ing not only brings insolubility and great swelling ratio to hydrogels, but also gives hydrogels stronger mechanical properties than single polymer single chains. Despite that, the mechan- ical properties of native PNIPAM hydrogels are still very weak (below 10 kPa) , especially in the swollen state. Native PNIPAM hydrogels are so weak and fragile that it’s very hard to measure. AFM indentation may break the hydrogel sample and the signal may be coved 43 CHAPTER 3. MECHANICAL PROPERTIES OF PNIPAM HYDROGELS by background noise. Several studies reported their measurements on the Young’s modulus, and they are summarized in figure 3.1. Out of these examples, six report Young moduli below 10 kPa in the swollen state, and one of them is slightly above 10 kPa in the collapsed state. The PNIPAM hydrogels tested by Fei et al.[1], who reported 81 kPa, has a much higher crosslinker density than the others. The Young’s modulus of 180 kPa reported by Takigawa [2]et al. is probably due to their choice of models. Figure 3.1: Select examples of reported mechanical properties of native PNIPAM hydrogels in previous studies. Most of the Young’s modulus are below 10kPa. Those extremely high values are due to different measurements methods and different models applied. (Figure is adapted from Haq et al.[3]) 3.2 Interpenetrating Polymer Network Numerous efforts
to improve the mechanical strength of PNIPAM hydrogels
have been made.
The
concept
of
Interpenetrating polymer networks (IPN) is one of the basic strategies. The method requires synthesizing two polymers together, and the network cannot be separated without breaking the chemical bonds. Applying IPN to PNIPAM hydrogels brings a different improvement other than mechanical strength. As mentioned in the previous chapters, PNI- PAM hydrogels have very slow response rates. It takes PNIPAM hydrogel 2 hours to expel 3.2. INTERPENETRATING POLYMER NETWORK 50% of water.[1] Zhang et al.[4] synthesized IPN hydrogels with 10 wt% of poly(vinylalcohol) (PVA), and these hydrogels expelled 98% water in only 18 minutes. This is due to the IPN channels that increase the diffusion rate of water. A summary of mechanical strength mea- surements of IPN PNIPAM hydrogels is listed in Table 3.2. An average of one magnitude increase of Young’s modulus in a swollen state is recorded for the IPN method compared to native PNIPAM hydrogels. Ohya et al. tested a PNIPAM-grafted gelatin hydrogel. The gelatin fabrication technique is similar to IPN, but on a bigger scale. PNIPAM-gelatins ar- chitectures with different PNIPAM chain density have Young’s modulus varying from 50 kPa to 250 kPa under room temperature.[5] It also showed a concavo-convex structure, similar to the topographies used in our work. Figure 3.2: A collection of mechanical properties of IPN PNIPAM hydrogels reported in multiple studies. The absolute values of those IPN PNIPAM vary a lot due to the reasons explained in the beginning of this section. (Figure is adapted from Haq et al.[3]) Double network is a unique type of IPN. In double network hydrogels, there are two asym- metrical networks, where the first one has high crosslink density and the other one has low to none crosslink density. The first synthesis of double network hydrogel was reported by Gong et al. with high crosslink density
poly(2-acrylamido-2-methylpropanesulfonicacid) (PAMPS) as
the
1st network and
low crosslink density
polymer polyacrylamide (PAAM) as
the
2nd network
. Fei et al. synthesized double network hydrogels with two PNIPAM hydrogels with different crosslink density, which result in a factor of 2 increase in the Young’s modulus.[6] By adding AMPS in the first network,a 4-fold increase of Young’s modulus was obtained.[1] It is due to AMPS-PNIPAM copolymer, which have bigger pores, can swell into the second PNIPAM network better than pure PNIPAM hydrogel.[7] Both of the DN hydrogels synthe- sized by Fei et al. have the same LCST, VTTP, and swelling behavior as native PNIPAM hydrogels. Li et al. synthesized DN hydrogels with PNIPAM as high crosslink density as one of the networks and PH responsive ionic polymer polyacrylic acid (PAA) as the low crosslink density network. The Young’s modulus they obtained was 3.6 MPa, and it came with a lower VTTP at 29 °C.[8] 3.2. INTERPENETRATING POLYMER NETWORK Figure 3.3: The double network of high crosslink density PNIPAM-AMPS and low crosslink density PNIPAM. (Figure adapted from Fei et al.[1]) Slide ring hydrogel is a structure in which cyclic molecules (host) are trapped on polymer chains (guest), and they can only slide along polymer chains. This mechanism is specifically referred to as polyrotaxane if the guest chains are polymer chains. PNIPAM is too thick and heavy to be threaded by the small cyclic molecules ( -cyclodextrin and -cyclodextrin). However, it is possible to use slide rings modified PEG as crosslinkers of PNIPAM chains. Bin Imran et al. performed this synthesis with multiple modified cyclic molecules, for example, 2-acryloyloxyethyl isocyanate and ionic monomer sodium acrylic acid. Unfortunately, the formed networks showed little improvement in the Young’s modulus ( 40 kPa), but the stretchability increased significantly.[9-11] 3.3. NANOCOMPOSITE PNIPAM HYDROGELS Figure 3.4: Schematic of Slide ring hydrogel with cyclic molecules (host) trapped on polymer chains (guest). (Figure adapted from Fleury et al.[12]) 3.3 Nanocomposite PNIPAM Hydrogels Adding fillers into polymers can significantly enhance the mechanical properties of the result- ing composite[13]; Adding nano-scale fillers can further enhance the mechanical properties, because smaller filler size results in larger interfaces between the filler and the polymer. These hydrogels, called nanocomposites, have the best capability of increasing mechanical properties of PNIPAM hydrogel. Several examples of nanocomposite PNIPAM hydrogels will be introduced. Figure 3.5: A collection of mechanical properties of nanocomposite PNIPAM hydrogel. (Fig- ure adapted from Haq et al.[3]) Wang et al.[14] not only incorporated nanocomposite, but also grafted their samples layer by layer. Specifically, NIPAM and DEOP (crosslinking agent) are filled between clay layers, then cured with UV light (Figure 3.6). Transmission electron microscopy confirmed the aligned structure with soft hydrogel layer in between hard clay laminar layers. AFM mea- surements on their samples showed a Young modulus of 1.54-43.2 MPa, varying with clay content, and indicated that the samples still maintained 740%-1200% deformation ability. The lowest elongation 740% was obtained with highest clay content of 23.2 wt% and highest Young’s modulus was 43.2 MPa.The nanocomposite also showed hardening effect when the deformation reached a threshold. 3.3. NANOCOMPOSITE PNIPAM HYDROGELS Figure 3.6: Schematic of how layered nanocomposite films were coated and synthesized. (Figure is adapted from Wang et al.[14]) Forney et al.[15] published their nanostructured PNIPAM-PDMS hydrogels generated through photopolymerization in lyotropic liquid crystal templates (LLC). The complicated structure, shown in Figure 3.4, was confirmed by SAXS. They claimed that in the swollen state the amount of water the structure can up take increased significantly. The increase of swelling ratio can be from factor of 2
up to an order of magnitude
higher,
depending on
the
PDMS concentration . The
mechanical strength
of
the nano-structured hydrogels is between 40 kPa to 80 kPa. They also reported that in the swollen state the rate of deswelling sped up to several minutes. Figure 3.7: The scheme of
LLC templating process to generate hexagonal PDMS–PNIPAM hydrogels
. (Figure adapted
from
Forney et al.[15]) Xia et al. reported a novel approach to synthesizehydrogel nanocomposites. With a high 25:1 NIPAM: MBA ratio, and 10-40 min short polymerization time, they showed that the reaction of unsaturated bonds in BIS formed chemical crosslinking with nanoparticles.[16] The resultant nanocomposite was highly stretchable (1430%) and had high tensile strength (
200 kPa in swollen
state
and 1
.5MPa
in collapsed state). The
mechanical properties are listed in Table 3.3. Moreover, the response rate also increased, such that the entire swelling process occurred within only 10 min. 3.3. NANOCOMPOSITE PNIPAM HYDROGELS Figure 3.8: Special polymerization of PNIPAM-MBA, and the architecture of nano- structured hydrogel with nano-structured architecture in both states formed by the un- saturated double bonds shown in (c). (Figure adapted from Xia et al.[16]) Lian et al.[17] reported strain hardening effect of PNIPAm−Laponite nanocomposite gels. Large deformation caused self-reinforcement with high clay content. AFM measurements of the Young’s modulus showed that the nanocomposite is stronger after the recovery from large deformation, and which is believed to be caused by the high clay content. The limit of this self-reinforcement effect is the distance between clay platelets
smaller than the particle diameter, when the clay concentration
exceeds
6% w/v
. For
the
sample with
the
highest
clay
content, AFM indentation shows that the sample is 358 kPa before stretching, 229 kPa as a swollen gel, 518 kPa as a stretched gel, 523 kPa as a fracture surface, and 67 kPa as a swollen fracture surface. Figure 3.9: Schematic of PNIPAm−Laponite nanocomposite gels before, after stretching, and recovered. When L>d (low clay content), there isn’t self-reinforcement effect after recovery. When L<d (high clay content), there is strong self-reinforcement effect after re- covery. (Figure adapted from Lian et al.[17]) 3.4 References 1. Fei, R., et al., Ultra-strong thermoresponsive double network hydrogels. Soft Matter, 2013. 9(10): p. 2912-2919. 2. Takigawa, T., et al., Change in Young’s modulus of poly (N-isopropylacrylamide) gels by volume phase transition. Polymer Gels and Networks, 1998. 5(6): p. 585-589. 3. Haq, M.A., Y. Su, and D. Wang, Mechanical properties of PNIPAM based hydrogels: A 3.4. REFERENCES review. Materials Science and Engineering: C, 2017. 70: p. 842-855. 4. Zhang, X.Z. and C.C. Chu, Synthesis and properties of the semi‐interpenetrating polymer network–like, thermosensitive poly (N‐isopropylacrylamide) hydrogel. Journal of applied polymer science, 2003. 89(7): p. 1935-1941. 5. Ohya, S., S. Kidoaki, and T. Matsuda, Poly(N-isopropylacrylamide) (PNIPAM)-grafted gelatin hydrogel surfaces: interrelationship between microscopic structure and mechanical property of surface regions and cell adhesiveness. Biomaterials, 2005. 26(16): p. 3105-3111. 6. Fei, R., et al., Thermoresponsive nanocomposite double network hydrogels. Soft Matter, 2012. 8(2): p. 481-487. 7. Nakajima, T., et al., A universal molecular stent method to toughen any hydrogels based on double network concept. Advanced functional materials, 2012. 22(21): p. 4426-4432. 8. Li, Z., et al., Preparation and characterization of pH-and temperature-responsive nanocom- posite double network hydrogels. Materials Science and Engineering: C, 2013. 33(4): p. 1951-1957. 9. Imran, A.B., et al., Fabrication of mechanically improved hydrogels using a movable cross-linker based on vinyl modified polyrotaxane. Chemical communications, 2008(41): p. 5227-5229. 10. Imran, A.B., et al., Poly (N-isopropylacrylamide) gel prepared using a hydrophilic polyrotaxane-based movable cross-linker. Macromolecules, 2010. 43(4): p. 1975-1980. 11. Bin Imran, A., et al., Extremely stretchable thermosensitive hydrogels by introducing slide-ring polyrotaxane cross-linkers and ionic groups into the polymer network. Nature communications, 2014. 5(1): p. 1-8. 12. Fleury, G., et al., From high molecular weight precursor polyrotaxanes to supramolecular 56 sliding networks. The ‘sliding gels’. Polymer, 2005. 46(19): p. 8494-8501. 13. Okada, A. and A. Usuki, Twenty years of polymer‐clay nanocomposites. Macromolecular materials and Engineering, 2006. 291(12): p. 1449-1476. 14. Wang, J., et al., A Strong Bio-Inspired Layered PNIPAM–Clay Nanocomposite Hydrogel. Angewandte Chemie International Edition, 2012. 51(19): p. 4676-4680. 15. Forney, B.S., C. Baguenard, and C.A. Guymon, Improved stimuli-response and mechani- cal properties of nanostructured poly(N-isopropylacrylamide-co-dimethylsiloxane) hydrogels generated through photopolymerization in lyotropic liquid crystal templates. Soft Matter, 2013. 9(31): p. 7458-7467. 16. Xia, L.-W., et al., Nano-structured smart hydrogels with rapid response and high elas- ticity. Nature communications, 2013. 4(1): p. 1-11. 17. Lian, C., et al., Self-reinforcement of PNIPAm–Laponite nanocomposite gels investigated by atom force microscopy nanoindentation. Macromolecules, 2012. 45(17): p. 7220-7227. Chapter 4 Atomic force microscopy (AFM) AFM has been mentioned several times in previous chapters for measuring both topogra- phies and mechanical properties. This chapter will introduce the theory and usages of this technique. In chapter 3, one of the major flaws of AFM that has been discussed is , that the Young’s modulus acquired from the force-distance curve is not accurate for super soft material, and it varies a lot with choice of models. This chapter will analyze the causes and introduce potential solutions to this flaw. 4.1 Introduction AFM was invented in 1985 by G. Binning, C. F. Quate, and Ch. Gerber.[1] It can scan a sample surface with a tip at atomic level. The geometry of the setup is a cantilever mounting on a chip, which is held still on a base with a tip
at the end of the cantilever
as shown
in figure 4.1
. The interaction
between the tip and the surface
causes
the cantilever to bend, and
a laser reflects from
the back side of the
end of the
cantilever onto a
detector.
The
detector records
the
displacement
of the
laser point which corresponds to the
bending of the cantilever
, and
the
geometry
of the sample
surface.
The bending of the cantilever
provides not only the information of topography, but also the information of mechanical properties, simply by
Hooke’s Law
and
the spring constant of the cantilever. The spring constant of the cantilever
is usually
in
between 5 mN/m and 40 N/m. The choice of the cantilever is based 57 CHAPTER 4. ATOMIC FORCE MICROSCOPY (AFM) on the mode of contact required and sample types. Figure 4.1: (top)
Detector measures the
displacement
of the laser
reflected
by the
end
of the
cantilever, caused by
the
bend of the cantilever, which reveals the surface topography of the sample surface. (bottom) The force applied on the cantilever
can be calculated
from
Hooke’s Law
with
spring constant of the cantilever
. (Figure adapted from NanoWizard AFM handbook) There are two basic contact modes of AFM: tapping
mode and
contact
mode. In contact
4.1. INTRODUCTION
mode, tip is
always
in contact with the
sample
surface
and, and drags along the sample surface. Both force (bending angle) and height can be used as fixed parameters in this mode to satisfy different measurements. Contact mode usually is used to discuss large lateral force, but it has a high probability to damage the sample. For tapping mode, There are several derivatives where they all share one common character, the tip is oscillating in all those derivate modes but differs in their lowest oscillation point. Intermittent contact mode is the most used mode for topography measurement. The lowest oscillation points of intermittent contact mode are at the surface of the sample, so that it is unlikely to damage the sample surface. The second type is non-contact mode, whose lowest oscillation point is just above the surface. This mode is not widely used, because of the large number of forces involved between sample surface and the tip to control the distance between them. Lastly, for force modulation mode, where the tip never leaves the sample surface. Similar to contact modes, there will be a large lateral force, and a high possibility to damage the surface. This mode is widely used for acquiring force-distance curves. Figure 4.2: (top) Intermittent contact and non-contact mode are both used for topography measurement, because they don’t damage the sample surface. (bottom) Contact mode and force modulation mode can also be used for topography measurement, but they have a high possibility of damaging the sample. However, they are the best option for acquiring force- distance curves. (Figure adapted from NanoWizard AFM handbook) 4.2 Models used to analyze Force-Distance Curve AFM measures mechanical properties of samples by acquiring a force-distance curve from contact mode and force modulation mode. To get Young’s Modulus, adhesion force, and viscosity from force-distance curve, many models have been proposed. H. Hertz[2] proposed a model in 1882, and ever since the invention of AFM, Hertz model has been widely used for analyzing force-distance curves. However, the Hertz model is not well defined for different tip shapes, adhesive surfaces, viscoelastic surfaces, plastic deformation, and super soft bio- materials. As a result, different models are proposed for different situations. 4.2. MODELS USED TO ANALYZE FORCE-DISTANCE CURVE 61 4.2.1 Hertz Model Heinrich Hertz’s graduate work in 1882 is considered the fundamental theory of contact mechanism. [3] He studied the mechanical response when two glass lenses are pressed against each other and determined a relation between the contact of a flat surface and a parabolic surface:[2] In this equation, P is the applied force between two surfaces;
E and are the
Young’s
modulus and Poisson’s ratio of the
flat surface; R
is the
radius of
the
parabolic surface;
and
h
is the
indentation depth.
The
Hertz model is restricted by several assumptions it made. To name, the sample (the flat surface) needs to be homogeneous, isotropic and present a linear elastic response. These condition require the sample to be uniform everywhere; the sample need to experience the same force in all direction; stress need to be proportional to strain, and all deformation of the sample need to be able to completely recover after indentation.[4] 4.2.2 Sneddon’s Extension The Hertz model only considered the contact between sample and a parabolic indenter (h R). However, one of the most used tips in AFM indentation is conical shape. Based on Hertz model, Ian N Sneddon gave a solution to conical indenter and an accurate solution for spherical indenters, since the original model derived by Hertz can only approximate to be spherical indenters when h R.[5] The Sneddon’s equation for spherical indenters is:[5] All the parameters in this equation represent the same meaning as in Hertz model, except for rc as the contact radius. The contact radius relates to the contact depth by the equation: [6, 7] For parabolic indenter, which is Hertz model, the relation between contact radius and the contact depth becomes:[5] The Sneddon’s extension for Hertz model for conical indenter is:[5] where θ is the half-angle of cone, the hmax is maximum indentation depth. 4.2.3 JKR Model and DMT Model To address another limitation of Hertz model, which doesn’t consider any other interaction between surface and indenter.With the existence of adhesive force, Hertz model cannot accu- rately predict the contact area when the indenter is pulled down by the surface. To address this issue,
in 1971, Johnson, Kendall and Roberts (JKR) proposed a new theory between two elastic
surface with adhesive force.[8] Derjaguin, Muller and Toporov derived an equation in 1975 based on Hertz model which also accounted adhesive force.[9] JKR observed higher contact area at low load, comparing to Hertz model, and they also observed nonzero contact area at zero load. They derived a relation of contact radius between a sphere and a plane:[8] where γ is the work done by adhesive force. They defined a force, that is the minimum force which the probe is still attached to the surface, to be pull-off force, and in JKR model this force is given by:[8] Derjaguin, Muller and Toporov kept the contact area and deformation same as predicted in Hertz model, but added an adhesive force to the load. Their contact area relation is given by:[9] with a pull force of:[9] JKR model and DMT model give different results, but no one is better than the other. JKR model is more accurate for bigger radius probes, stronger and short-range adhesive force, and softer sample.[10] Whereas DMT model is more accurate for the opposite conditions. A factor, Tabor’s parameter µT is always used to quantitatively describe the application range of these two models whichis defined as:[11] where z0 is the equilibrium separation of the surfaces. JKR is more accurate when Tabor’s parameter is bigger than 3. Maugis defined a different parameter to represent the boundary of these two models.[12] For λ>5, JKR model can give an accurate result, and when λ<0.1, DMT model can give an accurate result.[12] In between 0.1 and 5, it is called the transition regime. Figure 4.3: (top) This figure compared Hertz model, JKR model, DMT model to the actual indentation with adhesive force. Hertz model doesn’t consider any adhesive force; JKR model expects a short-range big adhesive force; DMT model considers a long-range medium adhesive force. (Figure adopted from Grierson et al.)[10] 4.2.4 Oliver-Pharr Model Oliver and Pharr introduced another model in 1992[13] and made some improvements on the model in 2004[14]. They proved the Young’s modulus can be directly calculate by[13] where
A is the area of the indenter
projected on sample
at contact depth
, and S is the stiffness of the sample,
calculated from the slope of the unloading curve at maximum
indentation.[13]
The
Young’s modulus
of
sample
is
related by the equation:[13] where i represents indenter, and the s represents sample. Oliver-Pharr model can be applied on any axisymmetric indenter, but projected areas of the indenter at contact depth is different for different indenters. For spherical indenters, the relation is Figure 4.4: The stiffness
is the slope of the unloading curve at
maximum indentation. hf
is the
displacement of
a
non-recovered sample,
the
plastic deformation. (Figure adapted from Oliver et al.)[14] For paraboloidal indenters, the relation is For conical indenters, the relation is Comparing to Hertz, Kontomaris noticed Hertz model always results in a smaller value than Oliver-Pharr model for biological samples, even though these two models should have same result for purely elastic deformation.[15] Oliver-Pharr model is the only model one can apply to analyze elastic-plastic deformation between these two models. Therefore Oliver- Pharr model provide a more accurate result than Hertz model in this situation. As a result, many reported Oliver-Pharr model is better for fitting soft material. [16-18] However, there are many publications points out that Oliver-Pharr model is poor at fitting viscoelastic materials.[19, 20] Tranchida used a correction factor on Oliver-Pharr Model when apply it on viscoelastic materials which is defined as:[20] where p is the penetration depth. a and b are factors acquired by calibrating on the polymer with highest modulus. The correction factor is a ratio between Oliver-Pharr model value and the real value: The correction factor is 1 for a stiff surface. Figure 4.5: The parameters in Oliver-Pharr model during unloading. (Figure adapted from Oliver et al.)[14] 4.2.5 Ting’s Model There are many efforts to fit force-distance curve for viscoelastic samples, because of its existence in majority biological samples.[21-24] However, it is very difficult due to its time sensitive properties, even with 2 m/s indentation speed, the viscoelastic behavior is still observable.[22] Ting’s model is one of the models that is designed for viscoelastic samples, which has been widely used.[25] Here, tm is the time of loading phase; tind is the complete time; is a dummy time variable; t1 is determined by the equation:[22] Ting’s model is usually with relaxation modulus,
such as standard linear solid (SLS
) and
power-law rheology
(PLR). PLR
is
a decay function: and, SLS is: Efremov et al. published a numerical solution to Ting’s model and they fitted force- indentation curve with an algorithm base on their numerical solution to Ting’s model.[22] In their study, they also reported that different indentation speeds result in different force- indentation curves for viscoelastic samples, which I also experienced in my experiment data.[22] Figure 4.6: The experiential and theoretical force-indentation curves of viscoelastic materials differ a lot by indentation speeds. (Figure adopted from Efremov et al.)[22] 4.3 Indentation on Super Soft Materials AFM is a powerful technique that can image biological samples and measure the mechanical strength of biological tissues. However, biological samples are ultra-soft and as a result, it is very easy to be damaged, especially with shape probes. Moreover, surface biological samples usually show adhesion and viscoelastic properties. Finally, there’s molecular interaction between probe and sample surface. Besides the biological samples itself, the AFM indentation on biological samples usually requires an aquatic environment. The indenter in the aquatic 4.3. INDENTATION ON SUPER SOFT MATERIALS environment experiences more drag force than in air, which make the signal very noisy. Figure 4.7: Length, time, and strength scales of biological materials. (Figure adopted from Moeendarbary et al.)[26] Many groups have tried to measure mechanical properties of soft and ultra-soft biological samples (E<1 MPa). Constantinides et al. modified probe to indentation in horizontal direction to minimize the noise in liquid.[27] They used both Oliver-Pharr model and linear viscoelastic Kelvin–Voigt model[28] to fit the force-indentation curve. They reported the bottom effect from a stiff substrate which will be discussed in detail in chapter 4.4 when the sample is not thick enough, which was also reported by Zhou et al.[29] They still measured liver and skin both under 1MPa. Kohn et al. reported on eliminating adhesion errors in Oliver-Pharr model at low indentation depth.[30] They think the error can be from 60% to 300%, so they modified Oliver-Pharr model to lower Young’s modulus. The measurement on PDMS and PEG using the corrected Oliver-Pharr model mostly matches the values from the JKR model. Kauman et al. points out the importance of z-displacement of surface position can significantly affect the Young’s modulus of most models, since the surface position is not always at zero z-displacement for ultra-soft materials.[31] Galluzzi et al. showed the capability of measuring a heterogeneous surface.[32] They inserted hard microspheres into PNIPAM gel, and successfully measured both regions in one indentation. Figure 4.8: The blue line is a force curve of an indentation with tip starts below the surface, and the green line is a force curve of an indentation with tip starts above the surface. It caused a 43% error in Young’s modulus. (Figure adapted from Kauman et al.)[31] 4.4. BOTTOM EFFECT 4.4 Bottom Effect There’s wide interest in studying surface properties of thin film with AFM. All the models introduced above assume the sample as an infinity half space, however, when measurement is done on a thin polymer film, the sample is not an infinity half space anymore, and the stiff substrate supporting the thin film is causing great overestimation of the Young’s modulus. It is believed that indentation depth exceeds 10% of sample thickness will lead to big substrate effect. [33-35] Figure 4.9: The (a) force curve and (b) Young’s modulus shows the effect from the stiff substrate in an 85 nm indentation. (c) The experiment data subtract theoretical Young’s modulus of neat substrate. (Figure adapted from Watcharotone et al.)[33] There are many efforts to eliminate the bottom effect. Garcia et al.[36] develop a bottom effect viscoelastic theory to minimize the bottom effect influence on the viscoelastic response of a cell cultured on a stiff substrate. Nguyen et al. reported the probe with large radius effects the Young’s modulus, especially at low film thickness.[37] Their result did not agree with any theory model on bottom effect, but their result wellagrees with Selby et al.’s exper- imentally results.[38] Selby et al. measured hydrogels (>500 nm p(HEMA+MA)) with 100 nm and 50 nm radius probes.[38] Figure 4.10: The Young’s modulus vs film thickness with shape (black) and dull (red) probe. The doll probe experienced more substrate effect than thin probe. (Figure adopted from Nguyen et al.)[37] Garcia et al. applied a bottom effect correction on to viscoelastic model, which is very useful for bio-material measurement.[36] The bottom effect correction they applied is: 4.4. BOTTOM EFFECT where I is the indentation, and both α and β depend on tip geometry. Cheng et al. studied the bottom effect in a vertical direction.[39] They put the sample vertical to AFM, with the sample and substrate both under measurement. They reported the Young’s modulus at the interphase between sample and substrate. Figure 4.11: (a) The force experienced by the sample and the substrate. (b) The modulus differs by probe diameter. (Figure adopted from Cheng et al.)[39] 4.5 Reference 1. Binnig, G., C.F. Quate, and C. Gerber, Atomic force microscope. Physical review letters, 1986. 56(9): p. 930. 2. Hertz, H., Ueber die Berührung fester elastischer Körper. Journal fur die Reine und Angewandte Mathematik, 1882. 1882(92): p. 156-171. 3. Johnson, K.L. and K.L. Johnson, Contact mechanics. 1987: Cambridge university press. 4. Kontomaris, S., et al., Determination of the linear elastic regime in AFM nanoindentation experiments on cells. Materials Research Express, 2019. 6(11): p. 115410. 5. Sneddon, I.N., The relation between load and penetration in the axisymmetric boussinesq problem for a punch of arbitrary profile. International Journal of Engineering Science, 1965. 3(1): p. 47-57. 6. Wenger, M.P., et al., Mechanical properties of collagen fibrils. Biophysical journal, 2007. 93(4): p. 1255-1263. 7. Kontomaris, S. and A. Stylianou, Atomic force microscopy for university students: appli- cations in biomaterials. European Journal of Physics, 2017. 38(3): p. 033003. 8. Johnson, K.L., K. Kendall, and a. Roberts, Surface energy and the contact of elastic solids. Proceedings of the royal society of London. A. mathematical and physical sciences, 1971. 324(1558): p. 301-313. 9. Derjaguin, B.V., V.M. Muller, and Y.P. Toporov, Effect of contact deformations on the adhesion of particles. Journal of Colloid and interface science, 1975. 53(2): p. 314-326. 10. Grierson, D., E. Flater, and R. Carpick, Accounting for the JKR–DMT transition in adhesion and friction measurements with atomic force microscopy. Journal of adhesion 4.5. REFERENCE 77 science and technology, 2005. 19(3-5): p. 291-311. 11. Greenwood, J., Adhesion of elastic spheres. Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, 1997. 453(1961): p. 1277-1297. 12. Maugis, D., Adhesion of spheres: the JKR-DMT transition using a Dugdale model. Journal of colloid and interface science, 1992. 150(1): p. 243-269. 13. Oliver, W.C. and G.M. Pharr, An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. Journal of Materials Research, 1992. 7(6): p. 1564-1583. 14. Oliver, W.C. and G.M. Pharr, Measurement of hardness and elastic modulus by instru- mented indentation: Advances in understanding and refinements to methodology. Journal of materials research, 2004. 19(1): p. 3-20. 15. Kontomaris, S. and A. Malamou, Hertz model or Oliver Pharr analysis? Tutorial regarding AFM nanoindentation experiments on biological samples. Materials Research Express, 2020. 7(3): p. 033001. 16. Drira, Z. and V.K. Yadavalli, Nanomechanical measurements of polyethylene glycol hydrogels using atomic force microscopy. Journal of the Mechanical Behavior of Biomedical Materials, 2013. 18: p. 20-28. 17. Blum, M.M. and T.C. Ovaert, Low friction hydrogel for articular cartilage repair: eval- uation of mechanical and tribological properties in comparison with natural cartilage tissue. Materials Science and Engineering: C, 2013. 33(7): p. 4377-4383. 18. Ye, D., et al., Deformation drives alignment of nanofibers in framework for inducing anisotropic cellulose hydrogels with high toughness. ACS applied materials interfaces, 2017. 9(49): p. 43154-43162. 78 19. Kaufman, J.D., et al., Time-dependent mechanical characterization of poly (2-hydroxyethyl methacrylate) hydrogels using nanoindentation and unconfined compression. Journal of ma- terials research, 2008. 23(5): p. 1472-1481. 20. Tranchida, D., et al., Accurately evaluating Young’s modulus of polymers through nanoindentations: A phenomenological correction factor to the Oliver and Pharr procedure. Applied Physics Letters, 2006. 89(17): p. 171905. 21. Hecht, F.M., et al., Imaging viscoelastic properties of live cells by AFM: power-law rheology on the nanoscale. Soft matter, 2015. 11(23): p. 4584-4591. 22. Efremov, Y.M., et al., Measuring nanoscale viscoelastic parameters of cells directly from AFM force-displacement curves. Scientific reports, 2017. 7(1): p. 1-14. 23. De Sousa, J., et al., Double power-law viscoelastic relaxation of living cells encodes motility trends. Scientific reports, 2020. 10(1): p. 1-10. 24. Garcia, P.D., C.R. Guerrero, and R. Garcia, Nanorheology of living cells measured by AFM-based force–distance curves. Nanoscale, 2020. 12(16): p. 9133-9143. 25. Ting, T.C.T., The Contact Stresses Between a Rigid Indenter and a Viscoelastic Half- Space. Journal of Applied Mechanics, 1966. 33(4): p. 845-854. 26. Moeendarbary, E. and A.R. Harris, Cell mechanics: principles, practices, and prospects. Wiley Interdisciplinary Reviews: Systems Biology and Medicine, 2014. 6(5): p. 371-388. 27. Constantinides, G., et al., Probing mechanical properties of fully hydrated gels and biological tissues. Journal of biomechanics, 2008. 41(15): p. 3285-3289. 28. Cheng, L., et al., Spherical-tip indentation of viscoelastic material. Mechanics of mate- rials, 2005. 37(1): p. 213-226. 29. Zhou, G., et al., Cells nanomechanics by atomic force microscopy: focus on interactions 4.5. REFERENCE at nanoscale. Advances in Physics: X, 2021. 6(1): p. 1866668. 30. Kohn, J.C. and D.M. Ebenstein, Eliminating adhesion errors in nanoindentation of com- pliant polymers and hydrogels. Journal of the mechanical behavior of biomedical materials, 2013. 20: p. 316-326. 31. Kaufman, J.D. and C.M. Klapperich, Surface detection errors cause overestimation of the modulus in nanoindentation on soft materials. Journal of the mechanical behavior of biomedical materials, 2009. 2(4): p. 312-317. 32. Galluzzi, M., et al., Space-resolved quantitative mechanical measurements of soft and supersoft materials by atomic force microscopy. NPG Asia Materials, 2016. 8(11): p. e327- e327. 33. Watcharotone, S., et al., Interfacial and substrate effects on local elastic properties of polymers using coupled experiments and modeling of nanoindentation. Advanced Engineer- ing Materials, 2011. 13(5): p. 400-404. 34. Vanlandingham, M., et al., Nanoscale indentation of polymer systems using the atomic force microscope. The Journal of adhesion, 1997. 64(1-4): p. 31-59. 35. Du, B., et al., Study of elastic modulus and yield strength of polymer thin films using atomic force microscopy. Langmuir, 2001. 17(11): p. 3286-3291. 36. Garcia, P.D. and R. Garcia, Determination of the viscoelastic properties of a single cell cultured on a rigid support by force microscopy. Nanoscale, 2018. 10(42): p. 19799-19809. 37. Nguyen, H.K., S. Fujinami, and K. Nakajima, Elastic modulus of ultrathin polymer films characterized by atomic force microscopy: The role of probe radius. Polymer, 2016. 87: p. 114-122. 38. Selby, A., C. Maldonado-Codina, and B. Derby, Influence of specimen thickness on the 80 nanoindentation of hydrogels: measuring the mechanical properties of soft contact lenses. journal of the mechanical behavior of biomedical materials, 2014. 35: p. 144-156. 39. Cheng, X., et al., Characterization of local elastic modulus in confined polymer films via AFM indentation. Macromolecular rapid communications, 2015. 36(4): p. 391-397. Chapter 5 Nanostructured PNIPAM Coatings with Tunable Topography and Mechanical Properties 5.1 Introduction Stimuli-responsive polymers have sparked significant
attention in the
polymer
community due to their potential in
bio-
applications, including
drug delivery [1-5], cell culture [5-7], biosensing [3], and tissue engineering [4, 8]. These polymers inherently undergo conforma- tional changes with change in temperature [9], pH [10], light[11], electric fields [12], or ionic strength [13]. Due to its low toxicity, biocompatibility, and tunable response under physiolog- ical conditions,
poly(N-isopropylacrylamide) (PNIPAM
), one
of the stimuli-responsive
poly- mers, has
been used in
multiple applications
PNIPAM is a
well-studied
thermo-responsive polymer with LCST
of
32 °C. When the temperature is
lower than LCST,
the
hydrophilic group dissolves
in
water,
and
the hydrogel undergoes swelling. Whereas above LCST, the hydrophobic group excretes water out, and the hydrogel changes into a collapsed state[14]. The 32 °C LCST, which is close to the body temperature, gives PNIPAM a huge advantage in realizing several bio-related functions, including loading drugs by breath-in method[15] and lowering the rate of drug delivery to a non-poisonous level[3, 16-18]. Another advantage 81 CHAPTER 5. NANOSTRUCTURED PNIPAM COATINGS WITH TUNABLE TOPOGRAPHY AND 82 MECHANICAL PROPERTIES of PNIPAM is that itis soft enough to simulate the mechanical properties of body tissues typ- ically desired in tissue engineering and cell culturing. The mechanical properties of PNIPAM have been studied by many groups earlier [19-21]. It has been reported that the mechanical properties of PNIPAM-based hydrogels across LCST show a low kPa range using different methods and fitting models. Pure PNIPAM hydrogel’s mechanical properties are considered poor to mimic harder body tissues, such as cartilage whose young’s modulus is in the low MPa range. As a result, improving the mechanical properties of PNIPAM hydrogel is vital to extending its applications[14, 22]. To improve the mechanical properties of PNIPAM -in particular, to enhance the Young’s modulus - various methods such as, grafted with gelatin[7], nanopillar in AAO nanoreactor[23, 24], and PNIPAM–Clay Nanocomposites [25], have been tried over the past few years. Composite PNIPAM with other molecules has been reported to increase the Young’s modulus significantly, although methods of measurements and the fitting models greatly influence the results. For example, p(NIPAM-co-AMPS) [26], calcium alginate [27], and poly(acrylamide) [28] show Young’s moduli as high as 100 kPa. Burmistrova and coworkers showed that higher cross-linker density can also improve the strength of PNIPAM hydrogels.[29] In this paper, we focus on discovering the behavior and properties of PNIPAM coated onto patterned silicon substrates. A similar structure of PNI- PAM has been studied before, with patterned PNIPAM monoliths [30, 31]. The dip-coating is performed in two directions, parallel and perpendicular to the substrate channels at a con- stant speed. In this work, the surface structure and mechanical properties are both measured by AFM.
5.2. MATERIALS AND METHODS 5.2 Materials and Methods 5.2.1 Fabrication of
PNIPAM scaffolds A PNIPAM was prepared by mixing 0.05 g of PNIPAM copolymerized with 3 wt% MaBP in 10 mL of isopropanol resulting in 0.63 wt% of PNIPAM. The solution was either spin-coated or dip-coated onto lithographically patterned silicon substrates purchased from LightSmyth Technologies. The substrates used in this study are in the form of linear gratings, i.e. periodic channels with well-defined period (p), channel height (h), and line width (w) obtained by scanning electron microscope (SEM). The substrates have the following characteristic (p, h, w) parameters: Scaffold 1 (855 nm, 200 nm, 428 nm), Scaffold 2 (675 nm, 170 nm, 280 nm), Scaffold 3 (606 nm, 190 nm, 303 nm).PNIPAM was coated onto Scaffolds 1-3 by dip-coating. Prior to coating, the substrates were
thoroughly cleaned by alternating rinses of chloroform, toluene, methanol, and high purity deionized water
. After
the rinse, the
substrates were plasma-etched, followed by dipping in a 2% 3-aminopropyltriethoxysilane 98% acetone solution, and were then oven-dried at 100 ℃ for at least 5 mins. The substrates were then clipped to a dip-coater arm and immersed in the PNIPAM solution, such that the channels were along the vertical coating direction for Scaffolds 1 and 2 and along the horizontal direction (i.e. perpendicular to the dipping direction) for Scaffold 3. The dip- coating process was performed using a HARVARD Apparatus PHD 2000 syringe pump turned 90° on its side to allow for controlled speed (8.6 mm/min) in the vertical direction. After coating, the substrates were left to air-dry for 30 mins and were then cross-polymerized for 15 mins using a BLAK-RAY ultra-violet lamp. MECHANICAL PROPERTIES 5.2.2 Topographic Studies. The topographic changes in the PNIPAM scaffolds were studied by
atomic force microscopy (AFM) performed on
the
MFP-3D AFM instrument (Asylum Research
)
at the Center for Nanophase Materials Sciences at Oak Ridge
Nation Lab (CNMS-ORNL). Studies of the surface topography
were performed in tapping mode to avoid damage to the
PNIPAM films during measurements. The experiments were done using a silicon cantilever (NSC38/NoAl Mikromasch), with a trihedral tip,
resonance frequency of
20
kHz, and a
force
constant of 0
.08
N/m
All measurements
were
performed in water and the temperature was controlled using a heating stage. The scaffolds were immersed in water for 30 mins before starting the experiment to fully hydrate the PNIPAM films. Temperature variation studies were performed in small temperature steps 1-2 ºC, followed by a 15 min equilibration time. 5.2.3 Mechanical Measurements Measurements of the Young’s modulus were performed using force curves or force maps. Data were collected during tip approach (loading) and tip retraction (unloading). The curves were analyzed using different available models including the Hertz model modified to the Sneddon’s equation for conical indenter. [32] The Oliver-Pharr Model was used to fit the unloading force-indentation curves
to determine the elastic modulus of the
polymer
film. This model
is typically applied in determining the Young’s modular for super soft materials [33]. To further assess the feasibility of these models in measuring the mechanical properties of the scaffolds, we also analyzed the force curves using Ting’s model, developed for for materials that have viscoelastic properties [34].
5.3. RESULTS AND DISCUSSION 5.3 Results and Discussion 5.3.1
Topography AFM measurements were performed on the hydrated scaffolds in 1-2 °C temperature steps over a thermal range between 15 °C and
35 °C, i.e
. far
below and above the lower critical solution temperature
(LCST)
of PNIPAM. For
comparison, control measurements were con- ducted on the dry samples prior to hydration. These measurements show that, in the dry state, the coated PNIPAM films conform to the underlying substrate periodicity but exhibit a smaller height difference, indicating that PNIPAM forms a thicker layer in the bottom of the channels compared to the tops (Figure 5.1). Figure 5.1: Surface profiles of the same grating before and after coating with PNIPAM. MECHANICAL PROPERTIES Figure 5.2: Surface topography of Scaffolds 1 and 2, dip-coated along the channel direction, for temperature variation from 35 °C to 22 °C. The average surface profiles using line cuts perpendicular to the channel direction are shown in stacked format in panels (a) and (c) for Scaffold 1 and 2, respectively. The same profiles are plotted in overlaid form starting from the same baseline in panels (b) and (d) for direct comparison. (Top) The topography changes in sample 1. (Bottom) The topography changes in sample 2. In the hydrated state, the hydrogel film showed varying topography as a function of temper- ature. In each measurement, we performed area scans of 5um×5um and considered multiple line cuts perpendicular to the channel direction. The surface profiles shown in Figure 2 are the averages of 10 lines cuts taken over different parts of the scaffold. Figure 2 shows the topography of Scaffolds 1 and 2 as the temperature was raised from 22 °C to 35 °C. We observed that as the PNIPAM transitions into a swollen state, the height difference be- tween the lines and channels decreases, indicating that the PNIPAM in the channels swells at a larger rate than that on the line tops. More interestingly, measurements around the transition temperature show that the PNIPAM film along the channel walls experiences a delayed transition, resulting in protrusions in the surface profile at the position of the chan- nel walls. This is possibly due to the larger volume of PNIPAM expanding along the channel walls in the vertical direction. However, this intermediate state sheds important light on the transition mechanism of PNIPAM films coating structured substrates. The intermedi- ate topography resulting from such transitions can significantly influence the use of these scaffolds in tissue engineering applications and in the studies of dynamic deformations in supported model cell membranes. Further inspection of the height difference between line tops and channel depths shows a trend of slowly decreasing height difference between 35 °C and 31 °C and a more rapid decrease between 31 °C to 21 °C. These results are similar to earlier observations by Vidyasagar et al. on surface tethered PNIPAM. Their study reported a slow change in film thickness away from the transition temperature and a rapid change between 31 °C and 30 °C, indicating a fast swelling/deswelling transition in that narrow temperature range.[37]. While our results follow a similar trend, we do not observe the same sharp change in height difference. This could be due to confinement effects that limit the degree of polymer swelling within the submicron channels used in this work. To fully inspect the behavior of the PNIPAM films along the substrate features, Zhernenkov et al. prepared similar substrates by spin-coating method and investigated them using neutron reflectometry and solvent deuteration. Their results showed that in the collapsed state, the PNIPAM film assumes a thickness of 34 nm along the line tops and 57 nm in the bottom of the channel. [6] However, in the swollen state, the film expands by multiple folds, completely filling the channels. These results were later corroborated by AFM images. MECHANICAL PROPERTIES Figure 5.3: Line cuts perpendicular to the channel direction show the topographic changes in PNIPAM-coated scaffolds prepared on a substrate with (d, h, w) of (855 nm, 200 nm, 428 nm); i.e. Scaffold 3. A similar transition (Figure 5.3) is observed on Scaffold 3, which was dip-coated perpendic- ular to the channel direction, resulting in a thicker polymer coating within the channels. At temperatures above the LCST of the PNIPAM film, the film surface shows a similar profile to that of the underlying substrate, i.e. same period of 855 nm but shallower height ( 80 nm) compared to that of the substrate ( 200 nm). As the film transitions to a swollen state, the polymer along the channel walls swells significantly more in the vertical direction compared to the polymer on the line tops. However, due to the large volume of polymer within the channels, the intermediate protrusions forming as a response of the film swelling along the channel walls get kinetically trapped and result in a doubling of the periodicity of the surface structures compared to the underlying substrate. In this case, the newly formed protrusions are more pronounced than the ridges due to the underlying substrate lines, technically shift- ing the peaks and wells in the scaffold topography. Notably, confined PNIPAAm-co-MaBP shows different degrees of volume expansion in parallel direction, and normal direction to the substrate has been reported, which is related to the behavior of sample 3.[38] To inspect the transition behavior of the polymer films, we used the height difference as a metric for film swelling/deswelling. Figure 4 shows the changes in the height difference of the surface (i.e. peak to trough) across the probed temperature range. Sigmoidal fits of the thermal de- pendence of the height difference indicate an LCST of 27 °C for all scaffolds. This measured LCST is 5 °C lower than the LCST of unconfined PNIPAM hydrogels. Figure 5.4: The height differences of three scaffolds at various temperatures across the phase transition are fitted with a sigmoidal function. All three samples show a transition temperature of 27 °C. MECHANICAL PROPERTIES 5.
3.2 Mechanical Properties
The
mechanical properties of the
scaffolds
were measured
by AFM force curves. This mea- surement mode requires the indentation of the surface and measure the approach and retrac- tion forces between the AFM tip and the measured surface. Force curve measurements, or force maps, were conducted following the topography studies at each temperature. Consid- ering the softness of the PNIPAM in water and the trihedral geometry of the used AFM tip, we chose to fit the obtained force curves to the Oliver-Pharr model which is readily available in the Asylum software. In addition, we also applied the Hertz model to fit the loading curves. Here, we should point out that considering the nanoscale thickness of the PNIPAM film and the hard nature of the silicon substrate we expect the
measurements of the Young’s modulus of the
polymer
film
to be affected
by
the close proximity to the substrate, especially in the top region. Figure 5.5: (a) Plot of the loading force curves of PNIPAM Scaffold 1 in the groove region. (b) The obtained elastic modulus is around 1 MPa in the swollen state at 24°C, and it increases to 15 MPa in the collapsed state at 30°C. The force curves obtained at different temperatures indicate that the mechanical properties of Scaffolds 1 and 2 significantly change with temperature (Figure 5b). Nine selected loading force curves from groove region of Scaffold 2 recorded from 22 °C to 35 °C are presented in Fig 5. The slopes of the force curves, which indicate the measured stiffness, decrease with the drop in temperature. Stiffness relates to the effective Young’s modulus by [33]: where, S is the stiffness; β is the correction factor and is equal to- 1.05 [40]; Eeff
is the effective Young’s modulus
; and A is
the
contact area. This
effective Young’s modulus
relates to the Young’s modulus of the sample, E, as[33]: where
Ei is the Young’s modulus of the indenter; v is the Poisson’s ratio of the
sample;
and vi is the Poisson’s ratio of the indenter. The
Young’s modulus for
the
channel region of both Scaffolds at various temperatures is shown in Figure 5.5. The results show an increase in the Young’s modulus with increasing temperature, with a trend that is very similar to that of the height difference. The Young’s modulus at 22 °C for both samples is just below 1 MP, which is similar
to the Young’s modulus of
PNIPAM
hydrogel in water
. However,
the
fits show a dramatic increase to 15 MPa at 30 °C. This is most likely due to the thinness of the film, which could result in the measured modulus becoming convoluted with the
Young’s modulus of the
silicon substrate (on
the
order
of
GPa).
This
phenomenon
is
exacerbated along
the
top regions of the scaffolds, which have relatively thinner layers compared to the channel regions. In those top regions, neither the Hertz model nor the Oliver-Pharr model gave reasonable fits for the force curves. MECHANICAL PROPERTIES Figure 5.6: Comparison of the fit
values of the Young’s modulus, obtained from
fits of
the
same loading
force
curves to the Hertz model and the Oliver-Pharr model measured on (a) Scaffold 1 and (b) Scaffold 2. The Hertz model shows the same increase iin the Young’s modulus with rising temperature, but with a much higher values than those obtained from the Oliver-Pharr model. The unloading force curve end at zero indentation, shows that the process of indentation is purely elastic. On the other hand, no adhesive force is observed on the unloading force curve, which is expected in the liquid environment [22]. Despite this, a previous study reported adhesive forces for PNIPAM in water.[42] and developed a model [43], which is based on the Hertz model but is better suited for conical indenters. This is given by where θ is the half-angle of the conical probe. Here it is worth pointing out that the Oliver- Pharr model assumes the sample to be purely elastic. However, considering the softness of PNIPAM in the swollen state, the samples are likely to show viscoelastic behavior, which will delay the recovery of deformation due to indentation, causing a slope difference at the early stages of the unloading curve as reported by Patra et al..[44] Here, we followed a similar approach as an additional comparison to the Oliver-Pharr model. An early model taking viscoelastic effects into account was derived by Lee and Radok [45]. Here, we use the model developed by Ting [34], that was further modified by Bastian et al. for cone-shaped probes [46], such that: The equation above is applied to the loading curve, where 0is the indentation speed; E0 is the young’s modulus; C represents the geometry of the probe; t0 is an arbitrary timescale parameter- chosen to be 1; and β is the fluidity which varies from 0 to 1, where 0 is liquid and 1 is solid.[46] For the unloading curves, where tm is the time of maximum indentation, and t1 (t) is given by:[30] The solution to the two above equations is [46]: MECHANICAL PROPERTIES Figure 5.7: Comparison of the Young’s modulus of (a) Scaffold 1 and (b) Scaffold 2 as obtained from fits to the Oliver-Pharr model (red) and Ting’s model (black). The fits to the Ting model show the same behavior across LCST as Oliver-Pharr Model (Figure 7). However, the Young’s modulus
obtained from the
Ting
model
fits has
higher
values
than that
extracted
from the
Oliver-Pharr
model
due to the viscoelastic property assumption. For Scaffold 1, the Young’s modulus varies from 100 KPa at 20 °C to 1.2 MPa at 31 °C; for Scaffold 2, the Young’s modulus varies from 100 KPa at 20 °C to 10 MPa at 30 °C. These values seem to depend on the film thickness and in this study vary with temperature. This follows finding reported in earlier studies which
have shown that
if
indentation depth
exceeds
10% of the film thickness
, there will be a significant substrate effect on the Young’s modulus.[47, 48] At higher temperatures (i.e. collapsed film) the top region where the film thickness is lower, the substrate effect results in a significant increase in the Young’s modulus beyond reasonable values.[49-52] It is reported that for PMMA, PS, and PC the Young’s modulus will increase when the film thickness is smaller than 300 nm -200 nm. [50, 53] At the edge of the groove, it will have a different type of bottom effect from a different direction, which has been reported by Cheng et. al.[54] On the other hand, the trihedral shape probes we used in this study can reduce this effect by a certain amount.[55] Account the bottom effect, the Young’s modulus values from both Oliver-Pharr 5.4. CONCLUSION model and Ting’s model are within the range of living cells to some harder body tissue. (reference) Those mechanical properties render these novel structured scaffolds as potential platforms for advanced cell culture applications and future studies on topographically tunable supported cell membranes. Jablin et al. verified the use of PNIPAM coatings on flat silicon substrates in studies of supported model membranes formed of DPPC and DPPE lipids. They reported that the mechanical properties of the lipid membranes can influence the swelling and hydration of the polymer film. It will be interesting to measure how this behavior affects the topographic and mechanical properties of the current scaffolds.[38] 5.4 Conclusion This work shows that controllable bio-compatible scaffolds can be designed by coating ther- moresponsive hydrogels onto nanostructured substrates with different feature sizes. Unex- pected surface topographies can be obtained by tuning the film thickness or coating method. By controlling the rate of temperature change, we can achieve different topographies that can be kinetically trapped into otherwise unlikely configurations. By further changing the coating direction, PNIPAM films can realize different topographies in a swollen state and re- verse back to same collapsed state. These topography changes can serve as a potential agent for controlled capture and release of cultured cells. These tunable nanostructured scaffolds can also serve as an excellent platform for the studies of dynamic deformations in supported model cell membranes. The tunability and features of the obtained surface topography can be effectively used to simulate the dynamic curvature variations and mechanical stresses experienced by living cells. 96 MECHANICAL PROPERTIES 5.5 References 1. Alexander, A., et al., Polyethylene glycol (PEG)–Poly(N-isopropylacrylamide) (PNI- PAAm) based thermosensitive injectable hydrogels for biomedical applications. European Journal of Pharmaceutics and Biopharmaceutics, 2014. 88(3): p. 575-585. 2. Bawa, P., et al., Stimuli-responsive polymers and their applications in drug delivery. Biomed Mater, 2009. 4(2): p. 022001. 3. Guan, Y. and Y. Zhang, PNIPAM microgels for biomedical applications: from dispersed particles to 3D assemblies. Soft Matter, 2011. 7(14): p. 6375-6384. 4. Ward, M.A. and T.K. Georgiou, Thermoresponsive Polymers for Biomedical Applications. Polymers, 2011. 3(3): p. 1215-1242. 5. Kim, Y.-J. and Y.T. Matsunaga, Thermo-responsive polymers and their application as smart biomaterials. Journal of Materials Chemistry B, 2017. 5(23): p. 4307-4321. 6. Zhernenkov, M., et al., Thermoresponsive PNIPAM Coatings on Nanostructured Gratings for Cell Alignment and Release. ACS Applied Materials Interfaces, 2015. 7(22): p. 11857- 11862. 7. Ohya, S., S. Kidoaki, and T. Matsuda, Poly(N-isopropylacrylamide) (PNIPAM)-grafted gelatin hydrogel surfaces: interrelationship between microscopic structure and mechanical property of surface regions and cell adhesiveness. Biomaterials, 2005. 26(16): p. 3105-3111. 8. Koetting, M.C., et al., Stimulus-responsive hydrogels: Theory, modern advances, and applications. Materials Science and Engineering: R: Reports, 2015. 93: p. 1-49. 9. Li, Y. and T. Tanaka, Kinetics of swelling and shrinking of gels. The Journal of Chemical Physics, 1990. 92(2): p. 1365-1371. 5.5. REFERENCES 10. Moselhy, J., et al., In vitro studies of the interaction of poly(NIPAm/MAA) nanoparticles with proteins and cells. Journal of Biomaterials Science, Polymer Edition, 2000. 11(2): p. 123-147. 11. Sershen, S.R., et al., Temperature-sensitive polymer–nanoshell composites for photother- mally modulated drug delivery. Journal of Biomedical Materials Research, 2000. 51(3): p. 293-298. 12. TANAKA, T., et al., Collapse of Gels in an Electric Field. Science, 1982. 218(4571): p. 467-469. 13. Duracher, D., et al., Cationic amino-containing N-isopropyl- acrylamide–styrene copoly- mer latex particles: 1-Particle size and morphology vs. polymerization process. Colloid and Polymer Science, 1998. 276(3): p. 219-231. 14. Haq, M.A., Y. Su, and D. Wang, Mechanical properties of PNIPAM based hydrogels: A review. Materials Science and Engineering: C, 2017. 70: p. 842-855. 15. Blackburn, W.H., et al., Peptide-functionalized nanogels for targeted siRNA delivery. Bioconjugate chemistry, 2009. 20(5): p. 960-968. 16. Das, M., et al., Biofunctionalized pH‐responsive microgels for cancer cell targeting: rational design. Advanced Materials, 2006. 18(1): p. 80-83. 17. Zhang, J., Z. Qian, and Y. Gu, In vivo anti-tumor efficacy of docetaxel-loaded thermally responsive nanohydrogel. Nanotechnology, 2009. 20(32): p. 325102. 18. Nayak, S., et al., Folate-mediated cell targeting and cytotoxicity using thermoresponsive microgels. Journal of the American Chemical Society, 2004. 126(33): p. 10258-10259. 19. Takigawa, T., et al., Change in Young’s modulus of poly (N-isopropylacrylamide) gels by volume phase transition. Polymer Gels and Networks, 1998. 5(6): p. 585-589. 98 MECHANICAL PROPERTIES 20. Puleo, G., et al., Mechanical and rheological behavior of pNIPAAM crosslinked macro- hydrogel. Reactive and Functional Polymers, 2013. 73(9): p. 1306-1318. 21. Gundogan, N., D. Melekaslan, and O. Okay, Rubber elasticity of poly (N-isopropylacrylamide) gels at various charge densities. Macromolecules, 2002. 35(14): p. 5616-5622. 22. Galluzzi, M., et al., Space-resolved quantitative mechanical measurements of soft and supersoft materials by atomic force microscopy. NPG Asia Materials, 2016. 8(11): p. e327- e327. 23. Sanz, B., et al., Thermally-induced softening of PNIPAm-based nanopillar arrays. Soft Matter, 2017. 13(13): p. 2453-2464. 24. Giussi, J.M., et al., Thermo-responsive PNIPAm nanopillars displaying amplified respon- siveness through the incorporation of nanoparticles. Nanoscale, 2018. 10(3): p. 1189-1195. 25. Wang, J., et al., A Strong Bio-Inspired Layered PNIPAM–Clay Nanocomposite Hydrogel. Angewandte Chemie International Edition, 2012. 51(19): p. 4676-4680. 26. Fei, R., et al., Ultra-strong thermoresponsive double network hydrogels. Soft Matter, 2013. 9(10): p. 2912-2919. 27. Petrusic, S., et al., Development and characterization of thermosensitive hydrogels based on poly(N-isopropylacrylamide) and calcium alginate. Journal of Applied Polymer Science, 2012. 124(2): p. 890-903. 28. Muniz, E.C. and G. Geuskens, Compressive elastic modulus of polyacrylamide hydrogels and semi-IPNs with poly(N-isopropylacrylamide). Macromolecules, 2001. 34(13): p. 4480- 4484. 29. Burmistrova, A., et al., Effect of cross-linker density of P(NIPAM-co-AAc) microgels at solid surfaces on the swelling/shrinking behaviour and the Young’s modulus. Colloid and 5.5. REFERENCES 99 Polymer Science, 2011. 289(5): p. 613-624. 30. Castellanos, A., et al., Size-Exclusion “Capture and Release” Separations Using Surface- Patterned Poly(N-isopropylacrylamide) Hydrogels. Langmuir, 2007. 23(11): p. 6391-6395. 31. DuPont Jr, S.J., et al., Swelling-induced instabilities in microscale, surface-confined poly(N-isopropylacryamide) hydrogels. Soft Matter, 2010. 6(16): p. 3876-3882. 32. Hertz, H., Ueber die Berührung fester elastischer Körper. Journal fur die Reine und Angewandte Mathematik, 1882. 1882(92): p. 156-171. 33. Oliver, W.C. and G.M. Pharr, An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. Journal of Materials Research, 1992. 7(6): p. 1564-1583. 34. Ting, T.C.T., The Contact Stresses Between a Rigid Indenter and a Viscoelastic Half- Space. Journal of Applied Mechanics, 1966. 33(4): p. 845-854. 35. Du, B., et al., Study of elastic modulus and yield strength of polymer thin films using atomic force microscopy. Langmuir, 2001. 17(11): p. 3286-3291. 36. Carias, V., J. Wang, and R. Toomey, Poly (N-isopropylacrylamide) cross-linked coatings with phototunable swelling. Langmuir, 2014. 30(14): p. 4105-4110. 37. Vidyasagar, A., J. Majewski, and R. Toomey, Temperature induced volume-phase tran- sitions in surface-tethered poly (N-isopropylacrylamide) networks. Macromolecules, 2008. 41(3): p. 919-924. 38. Jablin, M.S., et al., Influence of lipid membrane rigidity on properties of supporting polymer. Biophysical journal, 2011. 101(1): p. 128-133. 39. Jablin, M.S., et al., In-plane correlations in a polymer-supported lipid membrane mea- sured by off-specular neutron scattering. Physical review letters, 2011. 106(13): p. 138101. 100 MECHANICAL PROPERTIES 40. Cheng, Y.-T. and C.-M. Cheng, Relationships between hardness, elastic modulus, and the work of indentation. Applied Physics Letters, 1998. 73(5): p. 614-616. 41. Vidyasagar, A., et al., Continuous and discontinuous volume-phase transitions in surface- tethered, photo-crosslinked poly (N-isopropylacrylamide) networks. Soft Matter, 2009. 5(23): p. 4733-4738. 42. Rezende, C.A., L.-T. Lee, and F. Galembeck, Surface mechanical properties of thin polymer films investigated by AFM in pulsed force mode. Langmuir, 2009. 25(17): p. 9938-9946. 43. Harding, J.W. and I.N. Sneddon, The elastic stresses produced by the indentation of the plane surface of a semi-infinite elastic solid by a rigid punch. Mathematical Proceedings of the Cambridge Philosophical Society, 1945. 41(1): p. 16-26. 44. Patra, L. and R. Toomey, Viscoelastic response of photo-cross-linked poly (n-isopropylacrylamide) coatings by QCM-D. Langmuir, 2010. 26(7): p. 5202-5207. 45. Lee, E.H. and J.R.M. Radok, The Contact Problem for Viscoelastic Bodies. Journal of Applied Mechanics, 1960. 27(3): p. 438-444. 46. Brückner, B.R., H. Nöding, and A. Janshoff, Viscoelastic Properties of Confluent MDCK II Cells Obtained from Force Cycle Experiments. Biophysical Journal, 2017. 112(4): p. 724- 735. 47. Fischer-Cripps, A.C., Examples of nanoindentation testing, in Nanoindentation. 2002, Springer. p. 159-173. 48. Vanlandingham, M., et al., Nanoscale indentation of polymer systems using the atomic force microscope. The Journal of adhesion, 1997. 64(1-4): p. 31-59. 49. Garcia, P.D., C.R. Guerrero, and R. Garcia, Nanorheology of living cells measured by 5.5. REFERENCES AFM-based force–distance curves. Nanoscale, 2020. 12(16): p. 9133-9143. 50. Chung, P.C., E. Glynos, and P.F. Green, The Elastic Mechanical Response of Supported Thin Polymer Films. Langmuir, 2014. 30(50): p. 15200-15205. 51. Silbernagl, D. and B. Cappella, Mechanical properties of thin polymer films on stiff substrates. Scanning, 2010. 32(5): p. 282-293. 52. Kratz, K., T. Hellweg, and W. Eimer, Structural changes in PNIPAM microgel particles as seen by SANS, DLS, and EM techniques. Polymer, 2001. 42(15): p. 6631-6639. 53. Watcharotone, S., et al., Interfacial and substrate effects on local elastic properties of polymers using coupled experiments and modeling of nanoindentation. Advanced Engineer- ing Materials, 2011. 13(5): p. 400-404. 54. Cheng, X., et al., Characterization of local elastic modulus in confined polymer films via AFM indentation. Macromolecular rapid communications, 2015. 36(4): p. 391-397. 55. Nguyen, H.K., S. Fujinami, and K. Nakajima, Elastic modulus of ultrathin polymer films characterized by atomic force microscopy: The role of probe radius. Polymer, 2016. 87: p. 114-122. Chapter 6 Future Work This chapter is focused about the interaction between membrane and PNIPAM scaffold and the importance of the curvature that the new scaffold can provide. The membrane curvatures, which exist in real cells and cell organelles, have unique functions and they can be caused due to various reasons such as membrane asymmetry, . How the novel scaffold can mimic those characters will also be discussed in this chapter. 6.1 Membrane PNIPAM Scaffold Interaction One of the major forces, which drives me to study this novel scaffold, is its possibility to mimic cellular curvature of membranes. However, culturing membrane onto PNIPAM scaf- fold will cause interactions between membrane and PNIPAM scaffold. There are several studies looked into the interactions between membrane and normal PNIPAM scaffold. [1- 3] Those studies can help the future work of studying the interactions between membrane and PNIPAM scaffold with special topographies. Smith et al. studied single bilayer mem- branes (DPPC and DPPE) on PNIPAM-co-MaBP surface-tethered polymer network.[1] This surface-tethered polymer network can swell from 17 nm to 90 nm, between 25 °C and 37 °C. They observed both in- and out-of-plane fluctuations of the bilayer membranes, causing by inhomogeneous swelling of the polymer scaffold. The sample went through several temper- ature cycles from 37 °C to 25 °C in 16 days, and it remained intact, proving great stability 102 6.1. MEMBRANE PNIPAM SCAFFOLD INTERACTION of the sample. CHAPTER 6. FUTURE WORK Figure 6.1: (a) and (c) shows the neutron reflectivity (NR) profile at 37 °C and 25 °C respectively. (b) and (d) shows the scattering length density (SLD) profiles in fully collapsed and fully swollen state, contrast with the real sample profiles. The polymer swells from 17 nm to 90 nm, across the LCST, and a much higher fluctuation of the bilayer membranes is observed at 25 °C. (bottom left) The SLD profiles at different temperatures show the transition of fluctuation magnitude. (bottom right) The gray data points are at 37 °C after several temperature cycles to 25 °C in 16 days, and the black data points are before that. It shows great stability of the sample. (Figure adapted from Smith et al.)[1] 6.1. MEMBRANE PNIPAM SCAFFOLD INTERACTION Jablin et al. studied the fluctuations of membrane on PNIPAM support further.[2] They used off-specular scattering (OSS) to measure the in-plane structure and the
in-plane height-height correlation lengths of the
membrane and
the
fluctuations.[4] The
bilayer
thickness is 3.8 nm, while the correlation is
30 m at 37 °C
and
11 m at 25 °C
. They reported
the membrane
is
almost
flat at 37 °C, and
much
more distorted at 25 °C
, but the bending rigidity remains unchanged.[2] Jablin et al. published their work on the interaction between membrane and PNIPAM scaffold and they reported DPPC and DPPE significantly affects the swelling behavior of PNIPAM scaffolds. At 37 °C, two samples didn’t show much difference, only the sample supported DPPE is slightly more hydrated. At 25 °C, the sample supported DPPC swells 10% more than DPPE, but both samples are more uniform than at 37 °C. After removing the membrane cap, both PNIPAM scaffolds reestablished. They believed the higher bending rigidity of DPPE restricts the swelling of the PNIPAM scaffold.[3] This result is rather surprising, because the membrane is not bound to the scaffold, only weak Van de Waal force is between the interaction. For PINPAM scaffolds with special topographies, the cultured membrane is expected to have influence on the swelling of PNIPAM, which can result in different topographies and different Young’s modulus. The future work can investigate this effect to have better control over the desired curvature of cultured membrane. CHAPTER 6. FUTURE WORK Figure 6.2: These are SLD profiles of the samples with real structure as background. The DPPC sample is
on the left and the
DPPE
sample
is
on the right. The
top two are
at
37 °C, and the bottom two are at 25 °C. Both samples are more uniform at 25 °C. The DPPE sample is more hydrated at 37 °C, compared to the DPPC sample. The scaffold with DPPC swells more at 25 °C, because of the lower bending rigidity of DPPC. (Figure adapted from Jablin et al.)[3] 6.2. CELLULAR CURVATURES 6.2 Cellular Curvatures Curvatures are everywhere in cellular organelles. Only a few organelles such as lysosomes and peroxisomes, are almost spherical. Endoplasmic reticulum (ER), Golgi apparatus, mi- tochondrion, centrosome all has high curvature in the outer or inner surface. The surfaces of the cells also have curvature caused by proteins. Figure 6.3: An image of an electron micrograph of mitochondria. It shows the high curvature in the inner surface of micrography. (Figure adopted from Fawcett et al.)[5] There are many causes of curvature in cells, such as asymmetric monolayers, proteins insert in the membranes, and scaffolds. PNIPAM scaffolds with surface topography can mimic the scaffolding mechanism. In vivo scaffolding mechanisms require the protein scaffold to have stronger bending rigidity than the membrane to overcome the energy cost to bend the membrane.[6] The PNIPAM scaffold is tested to have a relatively high Young’s modulus to bend the membranes. Scaffolding mechanism is involved in generating curvature during vesicle buddings, generating the tubular ER network, generating the membrane sheet in ER, and generating the curvatures in mitochondrion and caveolae. It is very important to mimic CHAPTER 6. FUTURE WORK scaffolding mechanisms to understand the formation of cellular organelles and mechanism of cell functions. Figure 6.4: The scaffolding mechanism causes membrane curvature by the interaction be- tween scaffolding proteins and the membrane head groups. (Figure adopted from Shibata)[6] The membrane curvatures in the cell are very dynamic. They change with biological functions such as movement, division, and growth. The membranes are going to fusion and fission to form a new shape.[7] To mimic the fusion and fission process of membrane remodeling can be a future application of PNIPAM scaffold with surface topography, by varying the temperature of the sample at a constant speed or varied speed. 6.3 Conclusion This novel PNIPAM scaffold can stay with different curvature, as the previous chapter showed, at different temperature, which allows this scaffold to mimic different cellular or- ganelles for membranes. Considering the relatively slow response time of the scaffold, it is still not ideal to mimic the curvature change during a dynamic process, such as division. This can be improved by using different crosslinkers or polymerizing with different polymers. 6.4. REFERENCES PNIPAM copolymers usually can form tunnels between polymer strips which can expel wa- ter much quicker than pure PNIPAM, and the effect depends on the copolymer proportion. [8, 9] Crosslinking has similar effects, and the effect is depending on the crosslinker density. [9, 10] To precisely control the topography of the scaffold with a bilayer membrane on the top might need future studies on how much would different membranes affect the swelling of PNIPAM scaffold. 6.4 References 1. Smith, H.L., et al., Model lipid membranes on a tunable polymer cushion. Physical review letters, 2009. 102(22): p. 228102. 2. Jablin, M.S., et al., In-plane correlations in a polymer-supported lipid membrane measured by off-specular neutron scattering. Physical review letters, 2011. 106(13): p. 138101. 3. Jablin, M.S., et al., Influence of lipid membrane rigidity on properties of supporting polymer. Biophysical journal, 2011. 101(1): p. 128-133. 4. Pynn, R., Neutron scattering by rough surfaces at grazing incidence. Physical Review B, 1992. 45(2): p. 602. 5. Fawcett, D.W., S. Doxsey, and G. Büscher, Salivary gland of the tick vector (R. appen- diculatus) of East Coast fever. I. Ultrastructure of the type III acinus. Tissue and Cell, 1981. 13(2): p. 209-230. 6. Shibata, Y., et al., Mechanisms shaping the membranes of cellular organelles. Annual review of cell and developmental biology, 2009. 25(1): p. 329-354. 7. Chernomordik, L.V. and M.M. Kozlov, Mechanics of membrane fusion. Nature structural 4 6 8 10 12 14 16 18 20 22 32 34 36 38 40 44 46 48 50 52 54 58 60 62 64 66 68 70 72 74 76 84 86 88 90 92 94 104 106 108 CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS 1.2. PROPERTIES AND APPLICATIONS OF PNIPAM SOLUTION CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS 1.2. PROPERTIES AND APPLICATIONS OF PNIPAM SOLUTION CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS 1.3. PROPERTIES AND APPLICATIONS OF PNIPAM BRUSHES CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS 1.3. PROPERTIES AND APPLICATIONS OF PNIPAM BRUSHES CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS 1.4. PROPERTIES AND APPLICATIONS OF PNIPAM MICROGELS CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS 1.4. PROPERTIES AND APPLICATIONS OF PNIPAM MICROGELS CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS 1.4. PROPERTIES AND APPLICATIONS OF PNIPAM MICROGELS CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS CHAPTER 1. POLY(N-ISOPROPYLACRYLAMIDE): PROPERTIES AND APPLICATIONS 30 CHAPTER 2. SURFACE TOPOGRAPHY AND BIOLOGICAL APPLICATIONS OF HYDROGELS CHAPTER 2. SURFACE TOPOGRAPHY AND BIOLOGICAL APPLICATIONS OF HYDROGELS CHAPTER 2. SURFACE TOPOGRAPHY AND BIOLOGICAL APPLICATIONS OF HYDROGELS CHAPTER 2. SURFACE TOPOGRAPHY AND BIOLOGICAL APPLICATIONS OF HYDROGELS CHAPTER 2. SURFACE TOPOGRAPHY AND BIOLOGICAL APPLICATIONS OF HYDROGELS CHAPTER 2. SURFACE TOPOGRAPHY AND BIOLOGICAL APPLICATIONS OF HYDROGELS CHAPTER 3.
MECHANICAL PROPERTIES OF PNIPAM HYDROGELS
CHAPTER 3.
MECHANICAL PROPERTIES OF PNIPAM HYDROGELS
CHAPTER 3.
MECHANICAL PROPERTIES OF PNIPAM HYDROGELS
CHAPTER 3.
MECHANICAL PROPERTIES OF PNIPAM HYDROGELS
CHAPTER 3. MECHANICAL PROPERTIES OF PNIPAM HYDROGELS CHAPTER 3. MECHANICAL PROPERTIES OF PNIPAM HYDROGELS CHAPTER
4. ATOMIC FORCE MICROSCOPY (AFM
) CHAPTER
4. ATOMIC FORCE MICROSCOPY (AFM
) 4.2. MODELS USED TO ANALYZE FORCE-DISTANCE CURVE CHAPTER 4. ATOMIC FORCE MICROSCOPY (AFM) 4.2. MODELS USED TO ANALYZE FORCE-DISTANCE CURVE CHAPTER 4. ATOMIC FORCE MICROSCOPY (AFM) 4.2. MODELS USED TO ANALYZE FORCE-DISTANCE CURVE CHAPTER 4. ATOMIC FORCE MICROSCOPY (AFM) 4.2. MODELS USED TO ANALYZE FORCE-DISTANCE CURVE CHAPTER
4. ATOMIC FORCE MICROSCOPY (AFM
) CHAPTER
4. ATOMIC FORCE MICROSCOPY (AFM
) CHAPTER
4. ATOMIC FORCE MICROSCOPY (AFM
) CHAPTER
4. ATOMIC FORCE MICROSCOPY (AFM
) CHAPTER 4. ATOMIC FORCE MICROSCOPY (AFM) CHAPTER 4. ATOMIC FORCE MICROSCOPY (AFM) CHAPTER 5. NANOSTRUCTURED PNIPAM COATINGS WITH TUNABLE TOPOGRAPHY AND CHAPTER 5. NANOSTRUCTURED PNIPAM COATINGS WITH TUNABLE TOPOGRAPHY AND 5.3. RESULTS AND DISCUSSION CHAPTER 5. NANOSTRUCTURED PNIPAM COATINGS WITH TUNABLE TOPOGRAPHY AND 5.3. RESULTS AND DISCUSSION CHAPTER 5. NANOSTRUCTURED PNIPAM COATINGS WITH TUNABLE TOPOGRAPHY AND 5.3. RESULTS AND DISCUSSION CHAPTER 5. NANOSTRUCTURED PNIPAM COATINGS WITH TUNABLE TOPOGRAPHY AND 5.3. RESULTS AND DISCUSSION CHAPTER 5. NANOSTRUCTURED PNIPAM COATINGS WITH TUNABLE TOPOGRAPHY AND CHAPTER 5. NANOSTRUCTURED PNIPAM COATINGS WITH TUNABLE TOPOGRAPHY AND CHAPTER 5. NANOSTRUCTURED PNIPAM COATINGS WITH TUNABLE TOPOGRAPHY AND CHAPTER 5. NANOSTRUCTURED PNIPAM COATINGS WITH TUNABLE TOPOGRAPHY AND 101 103 5 7 9 11 13 15 17 19 21 23 25 29 31 33 35 37 39 41 45 47 49 51 53 55 59 63 65 67 69 71 73 75 79 83 85 87 89 91 93 95 97 105 107 109
