Precise Synthesis and Thermoresponsive Property of Poly(ethyl glycidyl ether) and Its Block and Statistic Copolymers with Poly(glycidol)

In this paper, we describe a comprehensive study of the thermoresponsive properties of statistic copolymers and multiblock copolymers synthesized by poly(glycidol)s (PG) and poly(ethyl glycidyl ether) (PEGE) with different copolymerization methods. These copolymers were first synthesized by ring-opening polymerization (ROP), which was initiated by tert-butylbenzyl alcohol (tBBA) and 1-tert-butyl-4,4,4-tris(dimethylamino)-2,2-bis[tris(dimethylamino)phosphoranylidenamino]-2Λ5,4Λ5-catenadi(phosphazene) (t-Bu-P4) as the catalyst, and then the inherent protective groups were removed to obtain the copolymers without any specific chain end groups. The thermoresponsive property of the statistic copolymer PGx-stat-PEGEy was compared with the diblock copolymer PGx-b-PEGEy, and the triblock copolymers were compared with the pentablock copolymers. Among them, PG-stat-PEGE, PG-b-PEGE-b-PG-b-PEGE-b-PG, and PEGE-b-PG-b-PEGE-b-PG-b-PEGE, and even the specific ratio of PEGE-b-PG-b-PEGE, exhibited LCST-type phase transitions in water, which were characterized by cloud point (Tcp). Although the ratio of x to y affected the value of the Tcp of PGx-stat-PEGEy, we found that the disorder of the copolymer has a decisive effect on the phase-transition behavior. The phase-transition behaviors of PG-b-PEGE, part of PEGE-b-PG-b-PEGE, and PG-b-PEGE-b-PG copolymers in water present a two-stage phase transition, that is, firstly LCST-type and then the upper critical solution temperature (UCST)-like phase transition. In addition, we have extended the research on the thermoresponsive properties of EGE homopolymers without specific α-chain ends.


Introduction
Aqueous solutions of some polymers show reversible phase-transition behavior, such as white turbidity when heated above the lower critical solution temperature (LCST) and dissolving again to clear when cooled below the LCST [1][2][3]. Such polymers include poly (N-substituted acrylamide)), poly (N-substituted methacrylamide), polyethers, and methyl cellulose [4][5][6][7][8]. On the other hand, sulfobetaine polymers, which are bipolar polymers, are known as polymers showing the upper critical solution temperature (UCST) after phase separation at low temperatures [9][10][11]. These polymers have been actively studied not only from the purely basic interest in clarifying phase-transition phenomena of aqueous solutions of polymers but also from the application aspect as intelligent polymers. The repeated unit applicable, and the LCST can be induced by hydrophobic modification of the hydroxyl groups and provide a blueprint to utilize them in biomedical applications.
In this study, we designed the copolymerization system of BnGE and EGE to extend to the synthesis of block copolymers and statistic copolymers without any specific end groups, that is, the multiple copolymers using PG as the hydrophilic chain segment and PEGE as the temperature-sensitive segment, as shown in Scheme 1. This study reports: (1) the living controlled ROP was catalyzed by t-Bu-P 4 using tBBA as initiator through the statistic and block copolymerization methods and then deprotected to obtain PG xstat-PEGE y and PG x -b-PEGE y , and the thermoresponsive properties of these copolymers with different x/y ratios were comprehensively studied; (2) the synthesis of multiblock copolymers composed of PG and PEGE chains, PG 15 20 , to explore the effect of the disorder to the copolymers on the thermoresponsive property; (3) we not only designed and synthesized the homopolymer PEGE without specific α-chain ends but also used dynamic light-scattering (DLS) measurement to explore the thermoresponsive properties of polymers in aqueous solutions.

Synthesis of PBnGE-stat-PEGE
A typical procedure for the statistic copolymerization of BnGE and EGE is described as follows: To a solution of tBBA (6.57 mg, 40 µmol) in 1.41 mL dry toluene, 0.05 mL of t-Bu-P 4 (40 µmol as 0.8 M solution in n-hexane), EGE (0.20 g, 2 mmol), and BnGE (0.33 g, 2 mmol) were added, in this order. The color of the solution changed from light yellow to dark yellow, indicating that the polymerization was successfully initiated. The polymerization was then quenched by adding a small amount of benzoic acid mixture to the polymerization solution. Aliquots were removed from the polymerization mixture to determine the conversion of BnGE and EGE by 1 H NMR measurements. The obtained polymer was purified through neutralized aluminum oxide to obtain PBnGE 50 -stat-PEGE 50 as a colorless and viscous liquid. Yield, 92.9%; M n, NMR , 13.5 kg mol −1 ; M w /M n , 1.07.

Synthesis of PG-stat-PEGE
A typical deprotection procedure is described as follows: Palladium on carbon (10% Pd/C; 100 mg) was added to a solution of PBnGE 50 -stat-PEGE 50 (100 mg; 7.41 µmol) in a mixture of methanol and CH 2 Cl 2 (1/1, v/v; 10.00 mL), and the whole mixture was stirred under a hydrogen atmosphere (3-5 MPa) for 24 h. After removing the Pd/C catalyst by filtration using celite, the solution was evaporated to obtain PG 50 -stat-PEGE 50 as a colorless and viscous liquid. Yield, 93.4%; M w,MALS , 8.8 kg mol −1 ; M w /M n , 1.10.

Synthesis of PBnGE-b-PEGE
A typical procedure for the diblock copolymerization of BnGE and EGE is described as follows: To a solution of tBBA (6.57 mg, 40 µmol) in 1.41 mL dry toluene, 0.05 mL of t-Bu-P 4 (40 µmol as 0.8 M solution in n-hexane) and BnGE (0.33 g, 2 mmol) were added, in this order. The color of the solution changed from light yellow to dark yellow, indicating that the polymerization was successfully initiated. After completion of BnGE polymerization (20 h, determined by the conversion of BnGE by 1 H NMR measurements), EGE (0.20 g, 2 mmol) was further added to the polymerization mixture to continue the block copolymerization for 20 h. The polymerization was then quenched by adding a small amount of benzoic acid mixture to the polymerization solution. Aliquots were removed from the polymerization mixture to determine the conversion of BnGE and EGE by 1 H NMR measurements. The obtained polymer was purified through neutralized aluminum oxide to obtain PBnGE 50 -b-PEGE 50 as a colorless and viscous liquid. Yield, 93.6%; M n, NMR , 13.5 kg mol −1 ; M w /M n , 1.05.

Synthesis of PG-b-PEGE
A typical deprotection procedure is described as follows. Palladium on carbon (10% Pd/C; 100 mg) was added to a solution of PBnGE 50 -b-PEGE 50 (100 mg; 7.41 µmol) in a mixture of methanol and CH 2 Cl 2 (1/1, v/v; 10.00 mL), and the whole mixture was stirred under a hydrogen atmosphere (3-5 MPa) for 24 h. After removing the Pd/C catalyst by filtration using celite, the solution was evaporated to obtain PG 50 -b-PEGE 50 as a colorless and viscous liquid. Yield, 90.5%; M w,MALS , 8.8 kg mol −1 ; M w /M n , 1.13.

Characterization
The 1 H and 13 C NMR spectra were recorded using a Bruker AVANCE III HD 500 (Bruker, Billerica, MA, USA) test in Changchun, China.
The polymerization solution was prepared in a MIKROUNA glove box (Changchun, China) equipped with a gas-purification system (molecular sieves and copper catalyst) and a dry argon atmosphere (H 2 O, O 2 < 1 ppm). The moisture and oxygen contents in the glove box were monitored by a MK-XTR-100 and a MK-OX-SEN-1, respectively.
The hydrodynamic diameter (D h s) of the obtained polymer was analyzed using a dynamic light-scattering (DLS) detector (Wyatt Technology, Dyna Pro Nanostar ® ) test in Changchun, China.
The cloud-point measurements were performed based on the ultraviolet-visible (UVvis) spectra by passing through a 10-mm path-length cell using a Jasco V-770 spectrophotometer (Tokyo, CA, Japan) equipped with a temperature-controller (Jasco CTU-100) test in Changchun, China. The temperature was increased at the rate of 1 • C min −1 , and the transmittance change with temperature was recorded using a spectrophotometer at the wavelength of 500 nm.
The matrix-assisted laser desorption/ionization time-of-flight mass spectrometry Bruker Autoflex III (MALDI-TOF MS) (Bruker, Billerica, MA, USA) measurements were performed using an Applied Biosystems Voyager-DE STR-H mass spectrometer with a 25-kV acceleration voltage test in Changchun, China. The positive ions were detected in reflector mode (25 kV). A nitrogen laser (337 nm, 3 ns pulse width, 106-107 W cm −2 ) operating at 3 Hz was used to produce the laser desorption, and the 200 shots were summed. The spectra were externally calibrated using a sample prepared from narrow-dispersed polystyrene. Samples for the MALDI-TOF MS were prepared by mixing the polymer (1.5 mg mL −1 , 10 µL), the matrix (trans-3-indoleacrylic acid, 10 mg mL −1 , 90 µL), and the actionizing agent (sodium trifluoroacetate, 10 mg mL −1 , 10 µL) in THF.

Synthesis of Statistic Copolymers and Block Copolymers
For preliminary research, we studied a series of copolymers of BnGE and EGE obtained by statistic/block copolymerization methods, i.e., PBnGE-stat-PEGE, PBnGE-b-PEGE. In accordance with Scheme 1, PBnGE was first prepared by the t-Bu-P 4 -catalyzed anionic ROP of the monomer BnGE using 4-tert-butylbenzyl alcohol (tBBA) as the initiator, followed by EGE as the M 2nd monomer for statistic/block copolymerization. Table 1 shows the list of the copolymerization results. Here, the total number with the ratio of monomers to initiator ([M 1st ] 0 /[tBBA]) and ([M 2nd ] 0 /[tBBA]) in the two stages is controlled at 100 to produce the copolymers with different molecular weights. All polymerization reactions are carried out with the monomer conversion rate (conversion rate >99.9%) to produce the corresponding copolymers, which required molecular weight ranges from 10.8 to 15.6 kg mol −1 . We prepared copolymers through the random copolymerization of BnGE with EGE, which are termed as, statistical copolymers. All of the size exclusion chromatography (SEC) traces of the PBnGE-stat-PEGE ( Figure 1) [10][11][12][13][14], respectively. In addition, the SEC trace showed a symmetrical monomodal peak with the M w /M n value of 1.05-1.08, which implied the absence of the side reactions like chain transfer during the block.  As shown in Figure 2, the structures of PBnGE 50 -stat-PEGE 50 (entry 5, Table 1) were verified by the 1 H NMR and 13 C NMR spectra. The 1 H NMR test revealed the comonomer ratio of BnGE to EGE and proved the presence of tBBA, which initiated the polymer. Comparing the integrals of the characteristic ethyl group (f ) peaks of BnGE at 4.45 ppm and the characteristic methyl group (k) peaks of EGE at 1.16 ppm revealed results very consistent with the comonomer ratio, with the targeted 1 to 1 ratio. The characteristic allyl (e) peaks at 1.36 ppm, assigned to the tert-butyl group at the α-chain end in PBnGE-stat-PEGE, confirmed that the presence of tBBA initiated the copolymerization. The peaks of the methylene protons (a, c, g, and i) and the methine protons (b and h) in the polymer main chains at 3.38-3.74 ppm, as well as those assigned to the methylene group protons (j) of EGE peak and the methylene proton (d) peaks of tBBA, obviously verified the successful copolymerization of PBnGE-stat-PEGE. Combined NMR and SEC measurements revealed the final composition of the statistic copolymer as PBnGE-stat-PEGE. Technically, 1 H NMR measurements still provided reliable information on the average degree of polymerization of the units generated for BnGE and EGE polymerizations (x and y, respectively). Additionally, the 1 H NMR spectra can obtain the number average molecular weights (M n,NMRs ) of the copolymer, which is consistent with the calculated molecular weight (M n,calcd ) determined by the monomer conversion and even the [  We aimed to investigate the thermoresponsive behaviors of the PG-b-PEGE consisting of PG and PEGE in order to gain insight into the effect of the monomer sequence. Thus, we conducted the sequential block copolymerization of PBnGE, followed by EGE, using the t-Bu-P 4 /tBBA system with [BnGE] 0 /[EGE] 0 /[tBBA] 0 ratios of 10/90/1, 20/80/1, 30/70/1, 40/60/1, and 50/50/1 to afford a series of PBnGE x -b-PEGE y with various x and y values (Scheme 1 and Table 1). As the first step of block copolymerization, t-Bu-P 4 with a [BnGE] 0 /[tBBA] 0 /[t-Bu-P 4 ] 0 ratio of 50/1/1 catalyzed ROP of BnGE for 20 h. 1 H NMR and SEC analysis of aliquots of the polymer showed that the monomer conversion rate reached 99%, and it was verified that active PBnGE oxyanions were formed, with an M n, NMR value of 8.3 kg mol −1 and M w /M n of 1.04. Then, the block copolymerization second stage was started by adding 50 equivalents of EGE relative to the tBBA used for the first polymerization. While monitoring the polymerization by 1 H NMR, the quantitative consumption of EGE after 24 h can be observed. What is more, the 1 H NMR and 13 C NMR of the obtained copolymer PBnGE 50 -b-PEGE 50 can capture the proton signal and carbon atom signal corresponding to PBnGE and PEGE, with calculated M w /M n and M n, NMR of 1.05 and 13.5 kg mol −1 , respectively. As shown in Figure 3, the SEC trace of the final product, PBnGE 50 -b-PEGE 50 , has shifted to a higher molecular-weight region compared with the PBnGE 50 obtained in the first stage of polymerization. It is not difficult to find that there is a small shoulder in the SEC curve in the higher molecular-weight region, but the M w /M n value of 1.05, including the shoulder, is still very narrow. The only shoulder showed in this trace is in the high molecular-weight region, which could be from the termination of the polymerization. These results confirmed the successful block copolymerization to produce PBnGE 50 -b-PEGE 50 with a predictable molecular weight and narrower M w /M n . PBnGE 10 -b-PEGE 90 , PBnGE 20 -b-PEGE 80 , PBnGE 30 -b-PEGE 70 , and PBnGE 40 -b-PEGE 60 were also successfully prepared in the same manner. In addition, we also prepared the triblock and pentablock polymerizations (Table S1 and Table S3, Supplementary Material).

Synthesis and Characterizations of PG-stat-PEGE and PG-b-PEGE
In order to obtain the designed copolymers PG-stat-PEGE and PG-b-PEGE, we used Pd/C as the catalyst to hydrogenate and crack a total of 14 samples of PBnGE-stat-PEGE and PBnGE-b-PEGE, and further removed the benzyl (Bn) group and 4-tert-butylbenzyl (tBBn) in the PBnGE segment. The deprotection reaction was performed under a hydrogen atmosphere, and methanol/CH 2 Cl 2 (1:1) was used as a mixed solvent at room temperature. The 1 H NMR measurement was performed to further confirm the quantitative progress of each reaction, and then the samples were purified to obtain pure deprotected polymers in a viscous liquid state (entries 15-28 in Table 2). The SEC traces of the PG-stat-PEGE after the debenzylation reaction are monomodal peaks, as shown in Figure S1 Table 2, ranging from 1.10 to 1.15, with a separation yield ranging from 85.3% to 93.5%. More importantly, by analyzing the values of M w /M n in Tables 1 and 2, it is concluded that the debenzylation reaction of the polymer did not lead to a significant broadening of the molecular-weight distribution.
As shown in Figure 4, for the 1 H NMR spectrum and the 13 C NMR spectrum of PG 50stat-PEGE 50 , PG as the main chain segment protons (a, b, and c) appeared at 3.48-3.82 ppm, and the characteristic peaks of Bn and tBBn groups were not observed in the spectrum. However, the PG signals through the 13 C NMR spectrum can be confirmed again, which appeared at 62.0, 80.4, and 70.8 ppm (a, b, and c, respectively). It is worth noting that although the characteristic peaks of the PEGE segment did not directly show in 1 H NMR, the peaks, g and h, of the ethyl group, respectively, appeared at 66.5 and 14.6 ppm in 13 C NMR. However, the above results indicated that deprotection quantitatively occurred in order to provide the copolymer PG-stat-PEGE without substituents at the α-end. Additionally, we used SEC equipped with a multiangle laser light scattering (SEC-MALS) instrument to test their actual weighted average molecular weights (M w,MALS's ), and the results for PG-stat-PEGE and PG-b-PEGE ranged from 7.2 to 9.7 kg mol −1 (entries 15-28 in Table 2). It is worth noting that as the α-end of copolymers are OH groups without characteristic peaks, we adopted the SEC measurements equipped with multi-angle laser light-scattering (SEC-MALLS) instruments to provide their weight average molecular weights (M w, MALS 's). As shown in Table 2, the actual weight average molecular weight (M w, MALS ) of the copolymers PG x -stat-PEGE y and PG x -b-PEGE y is in the range of 7.4-9.7 kg mol −1 .

Thermoresponsive Properties of PG x -stat-PEGE y and PG x -b-PEGE y
The thermal phase-transition behavior of copolymers was determined by the transmittancevs-temperature curves; that is, the T cp was defined as the temperature when the transmittance reached 50%. The transmittance-vs-temperature curves are formed by monitoring the polymer solution at a concentration of 10 g L −1 with an ultraviolet-visible spectrophotometer at a wavelength of 500 nm. Figure 5 shows the dependence of the transmittance of PG-stat-PEGE and PG-b-PEGE on temperature. Among the obtained copolymers, both PG-stat-PEGE and PG-b-PEGE were soluble in pure water at room temperature. It was found that the aqueous solution of PG 10 -stat-PEGE 86 , PG 20 -stat-PEGE 79 , PG 30 -stat-PEGE 70 , PG 40 -stat-PEGE 60 , PG 50 -stat-PEGE 50 , PG 60 -stat-PEGE 40 , and PG 70 -stat-PEGE 30 showed a sharp drop in transmittance when heated, indicating the existence of an T-type phase transition. In order to confirm the reversibility of the phase transition, we found that the optical transmittance of the solution returned to 100% after cooling, and the hysteresis was negligible. On the other hand, although PG 78 -stat-PEGE 20 and PG 87 -stat-PEGE 10 are completely soluble in water, the transmittance slightly decreases with increasing temperature. Surprisingly, it does not show any significant phase change at 90 • C. The aqueous solution of PG x -stat-PEGE y (x = 10, 20, 30, 40, 50, 60, and 70; y = 30, 40, 50, 60, 70, 79, and 86) reacts violently to thermal stimuli, as shown in Figure 5a, and the T cp are listed in Table 2. For PG-stat-PEGE, we observed that T cp increased with an increasing hydrophilic segment PG and a decreasing thermoresponsive segment PEGE, accompanied by decreasing of M w, MALS , which is usually observed in most thermoresponsive polymers. In other words, an earlier reduction in optical transmittance can be observed for PG x -stat-PEGE y with a higher EGE ratio, which is in line with our expectations. When the value of PG unit x increased to 78 and the value of PEGE unit y dropped to 20, it was difficult to exhibit LCST phase-transition behavior. Different from PG-stat-PEGE's performance, PG 10 -b-PEGE 86 , PG 20 -b-PEGE 79 , and PG 30 -b-PEGE 70 did not show a sharp drop in transmittance from 0 • C to 90 • C, only slightly changing in transmittance. Interestingly, the transmittance of PG 40 -b-PEGE 60 and PG 50 -b-PEGE 50 in the range of 0-60 • C appears to first decrease with increased temperature and then rise to a high value after further heating, as shown in Figure 5b. Similar results have occurred when we can decompose this special phase-change behavior into first LCST-type and then temperature (UCST)-like (actually not UCST) phase transitions. In summary, the PG x -b-PEGE y phase transition significantly differs from that observed for PG x -stat-PEGE y when the values of x and y are the same. In addition, taking the thermoresponsive property characteristics of PG 50 -b-PEGE 50 as an example, as shown in Figure 6, it can be concluded that the concentration of the polymer aqueous solution significantly affects the phasetransition process. In addition, the phase change becomes more obvious with increasing concentration, which may be attributed to the higher concentration and easier-to-form large aggregates. Therefore, in order to more directly understand the aggregation behavior of the polymer solution, we performed DLS measurements.
The hydrodynamic-diameter (D h ) results of PG x -stat-PEGE y and PG x -b-PEGE y are summarized in Table 2. As an example, PG 50 -stat-PEGE 50 was observed with a monodispersed distribution, and the D h was 10.4 at 20 • C, as shown in Figure 7a. This value conforms to the order of a single polymer chain, which proves that there is only a single polymer chain in the solution. When at a high temperature of 70 • C, the DLS measurement curve of PG 50 -stat-PEGE 50 still shows the monodispersed scattering distribution, whose D h value calculated from the peak is 722 nm, and it is used as a typical distribution of PG x -stat-PEGE y in Figure 7b. Obviously, the D h results show that PG 50 -stat-PEGE 50 forms uniform aggregates with larger particle sizes at a higher temperature than T cp , resulting in a polymer aqueous solution with a permeability of 0. For PG 50 -b-PEGE 50 , there is a two-stage phase transition during the heating process; that is, the average measured value of D h is only 9.6 nm at 5 • C. It rose to 268 nm quickly when the temperature increased to 25 • C. The D h then undergoes a second transition to 26 nm due to heating to 40 • C. We consider that there may be a micelle-like structure at 10 • C, so in the first stage of raising the temperature, the micelles aggregate to form large aggregates with a uniform D h of 268 nm. After the second stage of heating to above 40 • C, D h decreased to 40 nm, and the aggregates rearranged to form more stable, tiny aggregates. In other words, the solution is in a metastable system state at 25 • C, and a phase transition from the first stage to the second stage can occur when the temperature reaches 45 • C. Comparing the phase-transition behavior of PG x -stat-PEGE y and PG x -b-PEGE y , we infer that the phase-transition behavior is completely different due to the disorder of the polymer chain. In order to demonstrate the point of disorder, we designed the synthesis of multiblock copolymers.

Thermoresponsive Properties of Multiblock Copolymer and Homopolymer
In order to evaluate the thermoresponsive properties of multiblock polymers, we designed and synthesized four triblock copolymers, PG 15 35 . As shown in Figure 8a,b, our analysis shows that the thermal-response behavior of these four triblock copolymers significantly depends on the ratio of PG/PEGE. As observed, PEGE 35 Table S4, it is not difficult to find that the higher percentage of EGE units tends to have lower T cp . It should be noted that the pentablock copolymers PEGE 20 -b-PG 20b-PEGE 20 -b-PG 20 -b-PEGE 20 and PG 20 -b-PEGE 20 -b-PG 20 -b-PEGE 20 -b-PG 20 both exhibited LCST-type phase transitions in aqueous solutions, which were in sharp contrast to the results of triblock and diblock copolymers. Finally, we confirmed our conjecture that the dependence of T cp on the disorder of copolymers is undeniable. However, the exact of cause this phenomenon is unclear. As one example, Figure S2 shows the 1 H NMR spectra of PEGE 15 -b-PG 70 -b-PEGE 15 with the temperature changed in D 2 O. Since the instrument cannot cool down at room temperature, the temperatures were initially set to 23 • C, increased to 33 • C and 43 • C, then finally cooled to 23 • C. For the initial 23 • C measurement, the signals of the main chains and side groups could be clearly observed at 3.81-3.22 ppm (a, b, c, d). The signals obviously shifted to the low magnetic-field regions when the temperature increased from 23 • C to 43 • C. Similar tendencies were observed for the other signals in the range of 1.03-1.27 ppm (e). This result can be explained by the theory that the electron densities of the protons decrease due to intramolecular/intermolecular bond interactions, resulting in the shift of the signals to the low magnetic-field regions. Notably, the signals were not significantly broadened but significantly enhanced by the increasing temperature. We cannot provide a reliable explanation for this unexpected result, but the enhancement of the signal corresponds to this increase in transmittance behavior. For the final 23 • C measurements, the spectral profile agreed with that of the initial one, proving that the phase-transition phenomenon is reversible. In order to further determine the interesting phase-transition behavior, we performed DLS measurements on these multiblock polymers to understand the polymer at each stage of the phase-transition hydrodynamic diameter (D h ) at 20 and 60 • C. At high temperatures above 60 • C, DLS measurements show a monodispersed scattering distribution, which is a typical distribution curve for PG 15 -b-PEGE 70 -b-PG 15 ( Figure S3). For PG 15 -b-PEGE 70 -b-PG 15 , the D h at 60 • C is calculated as 545 nm. The results showed that although the light transmittance did not reach 0%, PG 15 Table S4.
We considered that while exploring the thermoresponsive property of the copolymer of PG and PEGE, it should be necessary to explore the homopolymer of PEGE to compare. Our group reported the synthesis of PEGE with the n-butoxy group as the α-chain end in previous studies [34]. However, we first used tBBA as an initiator to carry out ring-opening polymerization to obtain tBBA-PEGE in order to obtain the PEGE with a hydroxyl group at the α-chain end. Here, we designed the ratio of the initial monomer to the initiator ([M] 0 /[tBBA]) to be 25, 50, 75, and 100 to produce tBBA-PEGE with different molecular weights, which is obtained with a corresponding molecular weight of 2.7-10.4 kg mol −1 (Table S5). In addition, by observing the signal displayed by the 1 H NMR spectrum, we can confirm that the resulting product affords a tBBA group at the end of the α-chain ( Figure S4a). For the 13 C NMR spectrum ( Figure S4b), not only did the three signals appear at 61.0, 79.5, and 69.4 ppm (a, b, and c, respectively), but the signals of tBBA can also be observed. Figure S5 shows the matrix-assisted laser-desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) spectrum, which provides favorable evidence for the structure assignment of tBBA-PEGE. It can be seen from Figure 9a that the aqueous solutions of tBBA-PEGE 25, 50, 75, 100 all become opaque during heating, indicating the existence of the LCST-type phase transition. As listed in Table S5, the T cp s of the polyethers were observed in the temperature range of 9.1-12.1 • C and increased in the following order: tBBA-PEGE 100 (9.1 • C) < tBBA-PEGE 75 (10.4 • C) < tBBA-PEGE 50 (11.5 • C) < tBBA-PEGE 25 (12.1 • C). We synthesized the polymer PEGE with the α-chain end as the hydroxyl group through deprotection experiments and through a 1 H NMR spectrum test to confirm that the tBBA group of the previous α-chain end had disappeared. In Figure S6, two signals appeared at 1.23-1.42 ppm and 3.48-3.85 ppm (e and a, b, c, d, respectively). The size exclusion chromatography (SEC) experiment ( Figure S7) with the obtained products, PEGE, showed that the narrow dispersions (M w /M n ) were less than 1.10. However, these results indicated that we successfully afforded PEGE without substituents in both ends of the chain after deprotection. The aqueous solutions of PEGE (10 g L −1 ) were characterized by UV-Vis spectroscopy to determine their T cp s, which are listed in Table S6. The T cp value of PEGE 25,50,75,100 was in the range of 25.2 to 32.5 • C, as shown in Figure 9b. It is easy to find that both tBBA-PEGE and PEGE follow the same rule that T cp will decrease as the degree of polymerization increases. The exciting phenomenon is that PEGE with the same degree of polymerization has a higher T cp value than tBBA-PEGE, so we consider the α-end group has a considerable influence on the thermoresponsive property of the polymer. We further performed the DLS measurement of PEGE to monitor the D h at different temperatures. Figure S8 depicts the curve of hydrodynamic diameter (D h ) before and after the phase transition of PEGE 50 . Thus, we concluded that the thermal-response behavior of tBBA-PEGE and PEGE made our research on the thermoresponsive properties of its copolymers more colorful.

Conclusions
The ROP, using the combination of t-Bu-P 4 and the tBBA initiator, was efficient for synthesizing PBnGE-stat-PEGE and PBnGE-b-PEGE in different DPs with the intended molecular weights and relatively narrow molecular weight distributions, the controlled/living nature of could also afford PG-stat-PEGE and PG-b-PEGE. PG x -stat-PEGE y exhibited LCSTtype thermal-response behavior, and adjusting the x/y value caused a cloud-point temperature (T cp ) in the range of 30.5 to 70.4 • C, which may expand its application possibilities. To further study the thermal-response behavior of its copolymers, we synthesized triblock and pentablock copolymers. Therefore, we found that PEGE 35 15 . In addition, we found that PEGE homopolymers without specific α-chain ends also exhibited regular thermoresponsive properties. In general, the properties of all copolymer structures were evaluated, and the importance of copolymer disorder to thermoresponsive properties is emphasized.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.