Facile Fabrication of Bio‐ and Dual‐Functional Poly(2‐oxazoline) Bottle‐Brush Brush Surfaces

Abstract Poly(2‐oxazoline)s (POx) bottle‐brush brushes have excellent biocompatible and lubricious properties, which are promising for the functionalization of surfaces for biomedical devices. Herein, a facile synthesis of POx is reported which is based bottle‐brush brushes (BBBs) on solid substrates. Initially, backbone brushes of poly(2‐isopropenyl‐2‐oxazoline) (PIPOx) were fabricated via surface initiated Cu0 plate‐mediated controlled radical polymerization (SI‐Cu0CRP). Poly(2‐methyl‐2‐oxazoline) (PMeOx) side chains were subsequently grafted from the PIPOx backbone via living cationic ring opening polymerization (LCROP), which result in ≈100 % increase in brush thickness (from 58 to 110 nm). The resultant BBBs shows tunable thickness up to 300 nm and high grafting density (σ) with 0.42 chains nm−2. The synthetic procedure of POx BBBs can be further simplified by using SI‐Cu0CRP with POx molecular brush as macromonomer (M n=536 g mol−1, PDI=1.10), which results in BBBs surface up to 60 nm with well‐defined molecular structure. Both procedures are significantly superior to the state‐of‐art approaches for the synthesis of POx BBBs, which are promising to design bio‐functional surfaces.


Experiment details
Self-assembled monolayer (SAM) of APTES-BiBB initiator: Silicon substrates were cleaned with oxygen plasma cleaner for 15 min and washed extensively with distilled water, and dried by dry nitrogen flow. The freshly cleaned substrates were aminefunctionalized by immersed into a 5 % (v/v) aminopropyltrimethoxysilane (APTES) solution in dry acetone and ultrasonicated for 45 min under argon. After the SAM formation, the substrates were rinsed with acetone and dried by dry nitrogen flow. The substrate was then immersed in dry DCM (10 mL) under argon atmosphere. Afterwards, 0.2 mL triethylamine (TEA) was added before dropwise addition of 0.2 mL of 2-bromoisobutyryl bromide (BiBB) at 0 °C under argon. The mixture was then stirred under room temperature (RT) overnight. After the reaction the substrate was rinsed with DCM, water, ethanol, and acetone, and then dried with a dry nitrogen flow. The immobilization of the initiator SAM was finished for subsequent SI-CuCRP. Water contact angle after SAM immobilization: θ = 69 ± 2 °. SAM thickness: ~ 2 nm measured by ellipsometry.

Synthesis of 2-iso-propenyl-2-oxazoline (IPOx) monomer:
The IPOx monomer was prepared according to Seeliger and coworkers. [1] 10.0 g formaldehyde (HCHO, 0.33 mol, 1.1 eq.) and 29.73 g 2-ethyl-2-oxazoline (EtOx, 0.3 mol, 1 eq.) was mixed in a 500 mL round-bottomed flask with reflux-condenser. After addition of 0.759 g triethylamine (TEA, 75 mmol), the mixture was heated to 50 °C under stirring for 20 min. The mixture was then heated to 120 °C and stirred for 3 h under nitrogen atmosphere. The middle product hydox was distilled at 90 °C under reducing pressure 0.04 mbar. The obtained hydrox was then mixed with 3.0 g NaOH (75 mmol). After addition of few amount hydrochinon, the mixture was then heated to 120 °C and stirred for 4 h under nitrogen. The IPOx product was distilled at 120 mbar and 80 °C, which is colorless liquid. Yield: 12.1 g (36 %)

Synthesis of P(MeOx)7-MAA macromonomer:
The P(MeOx)7-MAA macromonomer was prepared according to Kobayashi et al. [2] 3.7926 g (23.1 mmol, 1 eq.) of MeOTf were weighed into a Schlenck-tube and dissolved in 30 mL ACN under argon atmosphere in glovebox. Then 10.6556 g (125.2 mmol, 5.4 eq.) of MeOx were added. The mixture was discharged from the glovebox and heated in an oil bath at 90 °C for 30 minutes. Afterwards the reaction was allowed to cool down to room temperature. 4.9776 g (57.8 mmol, 2.5 eq.) of MAA and 5.84 g (57.7 mmol, 2.5 eq.) of Et3N were added under argon protection. The mixture was heated up to 70 °C and stirred overnight. Then, 6 g of K2CO3 were added and the reaction was stirred at room temperature overnight. The solution was decanted and filtrated. ACN was removed under vacuum and the remaining solid was dissolved in 20-25 mL methanol. The product

SUPPORTING INFORMATION
3 was then precipitated into 750 mL diethyl ether. The solid was separated from the solvent and dried under vacuum. The product was purified by 2 more time precipitating and the solid was dried in freeze dryer to obtain the product as light sticky yellow powder. Yield: 9.19 g (70 %). SIPGP of PIPOx on silicon oxide: The silicon oxide substrate was cleaned with plasma cleaner for 15 min. Then the substrate was rinsed with Millipore-Q water and submerged into the solution of dry acetone and 5% (v/v) APTES. After 45 min ultrasonification the substrate was rinsed with acetone, ethanol and Millipore-Q water thoroughly. In a test tube put in 1 mL IPOx and degassed with nitrogen for 30 min to remove the oxygen, the cleaned substrate was submerged into the monomer and put under UV lamp (wavelength ~ 350 nm). The SIPGP was carried out under nitrogen for certain time. After the SIPGP, the substrate was taken out and rinsed with water and ethanol thoroughly.

Surfaced initiated LCROP to synthesis PIPOx-g-PMeOx BBBs:
In glove box an initiating-substrate was submerged in a solution of 2 mL ACN with an excess amount of MeOTf (30 mg) at approximately -35 °C under a dry argon atmosphere. After stirring for 5 h at 0 °C, the mixture was allowed to equilibrate to RT and stirred overnight. 1 g monomer MeOx (0.01mol) was added to the reaction flask afterwards. The reaction solution was heated at 80 •C and stirred for 4 h. Then, the solution was cooled down to 0 °C, and 1 mL N-tert-butyloxycarbonyl-piperazine dissolved in ACN solution (200 mg/mL) was added under argon atmosphere and the solution was stirred overnight at RT. Afterwards, an excess amount of potassium carbonate was added to the solution and stirred overnight to neutralize excess cations. The substrate was removed from the reaction solution and cleaned by ultrasonications in Milli-Q water, ethanol, and ethyl acetate for 1 min each. [4] The thickness data are summarized in Table S1. In order to estimate the grafting density of the BBBs, we employed AFM in liquid media to investigate their swollen thickness. The polymerization degree N and grafting density of the BBBs can be calculated according to equation (2) and (3). [5] The swollen ratio and grafting density are shown in Table   S2.

SUPPORTING INFORMATION 4
In order to estimate the side chain length as well as molecular weight. The reaction solution after LCROP was purified and precipitated, the obtained polymers were tested by GPC using DMAc as eluent. The data are shown in Figure S4, the change of the ratio between side chain monomer and initiator does not influence the polymer brush heights on surface obviously. The measured molecular weights of the polymers in solution are in good agreement with the theoretical molar mass. The dispersity of the polymer varies from 1.1 to 1.3 ( Table S3).

Deprotection of the Boc group:
The silicon substrate with PIPOx-g-PMeOx-Boc BBB was submerged in a solution of 1 mL trifluoroacetic acid (TFA) and 1 mL chloroform. The mixture was stirred at RT for 3 h. Then, the substrate was neutralized in a 5 % NaHCO3 water solution for 2 h. Finally, the substrate was washed by ultrasonification in water, ethanol and ethyl acetate for 1 min each. [4b] SI-CuCRP for the synthesis of polymer brush gradients: A silicon wafer piece modified with an APTES-BiBB-SAM was sandwiched with a copper plate with one side in direct contact and the other spaced at a distance D = 1 mm using a 1 mm thick stripe of glass (sample length L = 10 mm). The setup was fixed with clamps and emerged into respective reaction solutions (eg. Figure 5a).
The thickness data and surface wettability are summarized in Table S3.     Figure S4: Left: Thickness data of PIPOx-g-PMeOx-Boc BBB applying different monomer/initiator ratio; Right: Molecular weight and polydispersity index (PDI) of the polymers in reaction solution measured by GPC.

SUPPORTING INFORMATION
11  We performed AFM scans on different positions along Y-direction of both PIPOx brush and POx BBB gradients ( Figure S7). We found that the polymer layers at edge (P1 and P3) are 11 -28% thicker than that of middle positions (P2 and P4). One possible reason is that the concentration of monomer along Y-direction is different due to lateral diffusion. The unique experimental set-up of SI-CuCRP (Figure 1 b) creates a confined polymerization/reaction "chamber" between the copper plate and initiating-substrate. Once the monomer in the confined "chamber" is consumed by polymerization, the outside monomers diffuse into the "chamber", which therefore creates gradient concentrations of monomer along Y-direction from edge to middle and results in varied polymerization rate and brush thickness.

SUPPORTING INFORMATION
15   To further study the POx BBB surface via one-step approach, AFM scans were employed to investigate the grafting density and surface morphologies. The grafting density calculated from swelling ratios (146%, Figure S9e) is ca. 0.19 chains/nm 2 , which is considerably lower than that via two-step approach (0.42 chains/nm 2 ). This is because the monomer used in one-step approach is macromomer (Mn = 536 g/mol, PDI = 1.10), which is difficult to polymerize on surface due to larger steric hindrance and therefore leads to lower grafting density.
The POx BBB surface via one-step approach showed lower thickness (hdry ~ 37 nm) and higher roughness (Rms = 4.4 nm) in comparison with that of two-step approach (hdry ~ 100 nm, Rms = 1.2 nm). The water contact angel (θ) difference of two POx BBB surfaces are mainly attributed to the termination reagent used in the LCROP process in two-step approach (Figure 1). The N-tertbutoxycarbonyl piperazine (N-Boc-piperazine) has a hydrophobic Boc end-group, which contributes to the higher water contact angle (θ = 49 ~ 58°) of POx BBB from two-step approach. The POx BBB resulted from one-step approach has dual-functionalities as well, because the backbone brush and side chain of POx BBB can be functionalized, respectively.