Solid-state polymerization of EDTA and ethylenediamine as one-step approach to monodisperse hyperbranched polyamides

Abstract

Hyperbranched polyamides (HBPAs), a special class of polymers with extended use in colloidal systems, is synthesized by an innovative method taking advantage of the ordered aggregation of positive and negative molecular ions in solid-state. This novel and facile procedure introduces a one-step approach to convert ethylenediaminetetraacetic acid (EDTA) and ethylenediamine (EDA) in macromolecules with ordered framework. EDTA and EDA are converted in negative and positive molecular ions through acid–base reaction. Then, Coulomb interaction among the charged molecules are used to drive the aggregation of the ions during the precipitation induced by casting process. The solid material composed by the molecular ions are subjected to thermal treatment to promote amide formation reaction in solid-state. The synthesis condition is evaluated to understand the macromolecule growth. Solid-state reaction products are chemically characterized by FTIR and ¹H and ¹³C NMR. Molecular weight is determined by gel permeation chromatography (GPC) and the particles diameter in solution and charges are measured by dynamic light scattering (DLS) and zeta potential, respectively, at different pH values. The results attested the synthesis of hyperbranched polyamides with features similar to polyamide dendrimers. Surprisingly, the method enables the synthesis of macromolecules with very low dispersity index (DPI), that in some cases can be as low as 1.1.

---

Introduction

Hyperbranched macromolecules, such as star-polymers and dendrimers, have been playing important roles in the development of applied systems in colloidal chemistry, since they possess a vast list of potential applications due their unique characteristics in solution. For instance, as excellent host material to stabilize metal nanoparticles or as biomaterials to promote special interactions with biomembranes.^1–4 A remarkable feature of hyperbranched macromolecules is the high content of terminal groups.^5–7 These terminals groups can be easily converted to address the design of desired systems, affecting their solubility in different solvents, biocompatibility and reactivity.^8–10 In addition, stable conformations assumed by hyperbranched macromolecules in solutions provide outstanding capping properties for nanoparticles or drugs.^11–14 As a consequence of their remarkable properties in solution, hyperbranched macromolecules are widely explored in nanotechnology, biomedicine, coatings process and resins formulations.^15–17

A large diversity of hyperbranched macromolecules were previously reported.^18–22 Despite the large diversity of structures, two main classes of hyperbranched macromolecules emerge as more prominent: dendrimers and hyperbranched polymers. Dendrimers are perfectly branched and monodisperse macromolecules. Dendrimers are prepared by complex synthesis approaches of successive cyclic steps that end up in expensive materials and therefore, materials that are not attractive for large-scale industrial process.^23,24 On the other hand, hyperbranched polymers are usually prepared through one-pot synthesis. The issue of one-pot synthesis is the limited control on molecular weight and branching accuracy. Therefore, hyperbranched polymers are heterogeneous materials in terms of molecular weight and branching.^25,26 Despite the more heterogeneous structure, hyperbranched polymers have the facile synthesis as their great advantage, which ease their fabrication in large amount at reduced cost. Hence, they can be thought to be used in general applications.

The comparison between dendrimers and hyperbranched polymers rises a challenge in polymer science, regarding the development of systems to combine the advantages from these two polymers classes. A convergence point between dendrimers and hyperbranched polymer would result in outstanding macromolecules by a technology point of view. The convergence point should correlate very simple synthesis and purification procedures to result in dendritic polymers with well-defined molecular size and shape structure.

Organic solid-state reactions are normally uncontrolled processes with very few examples in the literature.^27–29 Inherent solid-state features, what could be thought to drive reactions to be hardly controlled, such as low reactants diffusion and degree of freedom, can be explored in a way to favor the formation of organized frameworks. Molecules with specific and strong intermolecular interaction, aggregate in the solid-state forming crystals, where the orientation of the molecules are assumed to benefit the attractive forces. Quasi-static location of molecules in a crystal in relation to their neighborhood could indeed be explored to drive reaction at specific molecular positions. It could be done by placing reactive groups of different molecules in close proximity to favor a determined reaction.

The methodology based in the use of aggregated reactants organized in the solid-state to favor the formation of macromolecules with organized framework is for the first time demonstrated in the present work. Herein, the preparation of hyperbranched polyamidoamine by a single reaction step procedure is presented and discussed. The materials prepared have chemical structures akin to the well-known polyamidoamine dendrimers, known as PAMAM,^30,31 which is a widely studied dendrimers in fields of drug delivery and nanoparticles preparation.^32,33

Experimental

Precursor salt

Ethylenediaminetetraacetic acid (EDTA) and ethylenediamine (EDA) were purchased from Sigma-Aldrich and used as received. EDTA and EDA solution were prepared using deionized water.

Ethylenediaminetetraacetic acid 1 mol L⁻¹ (EDTA) solution and ethylenediamine 1 mol L⁻¹ (EDA) solution were mixed in proper amounts to prepare solutions with EDTA to EDA ratio, of 1 : 1.5 or 1 : 2. For instance, 100 mL of 1.0 mol L⁻¹ EDTA solution was poured into 150 mL of 1.0 mol L⁻¹ solution of EDA. The molecular ratio between EDTA and EDA in the final solution is 1 molecule of EDTA to 1.5 molecules of EDA. Similarly, a solution with EDTA to EDA ratio of 1 : 2 was also prepared by adding 100 mL of 1.0 mol L⁻¹ EDTA solution into 200 mL of 1.0 mol L⁻¹ of EDA. The formed solutions were dried in a heating plate at constant temperature of 65 °C. In this step, the solvent in a beaker was evaporated without stirring. The solids obtained were transferred to Petri dishes and stored under vacuum in a desiccator using silica beads as desiccant prior their use in the next step.

Solid-state polymerization

Solid-state polymerization were carried out by the thermal treatment of the EDTA : EDA salts. Different thermal treatment conditions were tested. Three different temperature levels (120, 140 and 160 °C) and five thermal treatment time (2, 4, 8, 12 and 24 hours) were used.

The solid materials obtained after the thermal treatment were dissolved in deionized water to produced solution having 1 g of polymer in 10 mL of solution. The obtained solutions were purified using a size-exclusion chromatography (SEC) with Sephadex LH-20 as stationary phase. Sephadex are dextran beads with particles size in the range of 25 to 100 μm. Sephadex column were prepared to have 15 cm. Prior the application in the SEC column the solutions were filtered using a sintered glass filter with pore sizes in the range 40 to 100 μm. After the filtration the polymer solution was applied to the column and the column was eluted using distilled water. The first fraction containing around 15% of the applied mass is separated for characterization.

Characterization

Chemical characterization was carried out using FTIR and ¹H and ¹³C NMR. FTIR spectra were recorded in a Bomen MB-100 and analyzed using PeakFit software. ¹H and ¹³C NMR spectra were acquired in a Varian Mercury Plus BB spectrometer, working at 300 and 75.45 MHz, respectively.

Gel permeation chromatography (GPC) were carried out to determine the products molar weight (M_n and M_w) and dispersity index. GPC were determined using a gel permeation chromatography GPCmax VE2001 Viscotek model equipped with triple detection system consisting of refractive index (RI), viscosity and light scattering detectors. In the GPC experiment, water was used as the mobile phase under flow rate of 1.0 mL min⁻¹. The integration of the three detectors allowed to obtain the absolute molar weight of the sample.

Dynamic light scattering (DLS) and zeta-potential measurements were carried out at 25 °C using a Brookhaven Instruments Corporation instrument equipped with a He/Ne ion laser (λ = 633 nm). DLS measurements utilized a 90° detection angle: hydrodynamic diameter (D_h) was calculated by the Stokes–Einstein equation, and size distribution was obtained by CONTIN analysis (90Plus/BI-MAS software). Zeta-potential was calculated by the Smoluchowski equation (Zeta Plus Analyzer software). In the DLS and potential measurement, the pH of the polymeric solutions was adjusted with HCl (0.1 mol L⁻¹) or NaOH (0.1 mol L⁻¹).

Results and discussion

Polyamidoamines were obtained by the polymerization of ethylenediaminetetraacetic acid (EDTA) and ethylenediamine (EDA) in solid-state, wherein amide formation reaction causes the condensation of EDTA and EDA. EDTA and EDA are two very widely used substances and both are very well-known as excellent chelating agents. Nevertheless, direct polymerization between EDTA and EDA is not verified in aqueous solution due to acid–base equilibrium. In aqueous solution of EDTA and EDA, acid–base equilibrium favors the neutralization products (K ≫ 1). Therefore, primary amine groups of EDA and carboxylic acid groups of EDTA are mainly present in the form of protonated amine cations and carboxylate anions, as represented in the eqn (1).

As consequence of above equilibrium, amide formation by the condensation reaction of EDTA and EDA is hindered. The direct use of EDTA and EDA to form polymers cannot be executed in aqueous condition. On the other hand, acid–base equilibrium provides a very interesting opportunity, since acid–base reaction promotes the formation of molecular ions in solution. In the Fig. 1 is presented a schematic representation of the synthetic approach used to obtain hyperbranched polyamide. The acid–base neutralization reaction, that initially difficult amide formation in solution and avoid the polymerization of EDTA and EDA, can be used to the formation of organized macromolecule as represented in the Fig. 1. The molecular ions in solution can be converted into molecular crystals by the simultaneous precipitation of EDTA and EDA. In that case, negatively charged EDTA⁴⁻ acts as anion specie and positively charged EDA²⁺ as cation specie. In the experimental procedure, aqueous solutions containing a mixture of EDTA and EDA was heated at 65 °C to allow a slow precipitation process to optimize the formation of a solid crystal dominated by Coulomb interaction among counter ions, and therefore to drive the formation of organized array of EDTA and EDA molecules. The dried materials obtained from EDTA : EDA solutions consist of yellowish and very hygroscope solids. After dried the solid must be kept in a desiccator, otherwise, the solid material obtained quickly absorbs water and becomes a viscous liquid. The strong hygroscope behavior of the solids is attributed to the ionic features of the species associated to a low lattice energy.

Fig. 1 Schematic representation of the hyperbranched polyamide synthesis through a solid-state polymerization. The carboxylate (yellow) and aminium ions (blue) reacts in solid-state to form amide groups (green).

The amide formation reaction can be more effectively carried out in solid-state than in aqueous media, since water is a product of the reaction in the amide formation. In the complete absence of water, the condensation reaction is not a reversible process, therefore higher yields would be easily obtained. The amide reactions occur as demonstrated in the reaction scheme shown in the eqn (2).

The condensation reactions in solid-state is carried out by thermal activation. The crystals are heated in the temperature range of 120 to 160 °C. Temperatures higher than 160 can cause decomposition of the organic material. The amide formation during the thermal treatment is easily verified in the FTIR spectra of the samples prepared in different EDTA to EDA ratio (Fig. S1–S3†). The spectra of the samples after the thermal treatment evidenced the appearance of a signal at 1680–1630 cm⁻¹, characteristic of amide C=O (ν_C=O) groups.³⁴ In addition, FTIR spectra present band at 1588 cm⁻¹ characteristic of asymmetric angular deformation of the NH bond of secondary amines (δ_N–H),³⁵ bands between 2800 and 2947 cm⁻¹ assigned to the stretching vibration of the bond CH of CH₂ groups and aliphatic CH₃ (ν_(s)C–H and ν_(as)C–H) and a broadband after 3000 cm⁻¹ which may be associated with the presence of many –OH groups in HBPAs structure.³⁶

A study of variable was carried out to analyze the formation of the amide group and to determine the more suitable condition to the thermal reaction between EDTA and EDA in the solid in function of time and temperature of the thermal treatment. In this step the EDTA/EDA crystals were heated from different times (2, 4, 8, 12 and 24 hours) and temperatures (120, 140 and 160 °C). The reaction conversions were analyzed by FTIR (Fig. S1–S3†). The relative area of the amide ν_C=O signal were used to evaluating the conversion of aminium and carboxylate groups to amide groups in function of temperature and reaction time. Thus, reaction yield was analyzed between the relative area of the ν_C=O signal of amide and the relative area of the δ_N–H signal of secondary amine, which was set as a reference, in the spectra of normalized FTIR (Fig. S4 and S5†).

Comparing the FTIR results, it is possible to verify that the temperature has high influence on the relative area of the amide group at 1660 cm⁻¹. It is observed the amide peak increase with the increase of the thermal treatment. On the other hand, reaction time does not appear to provide practical changes to the relative area of the amide group. To optimize the synthesis and to reduce possible side reactions as consequence of thermal degradation process, the chosen condition to further experiments was the thermal treatment condition with the highest temperature (160 °C) and the shortest time (2 hours).

The materials obtained in the thermal treatment were submitted to a purification step using a cross-linked dextran gel (Sephadex-LH20).³⁷ The Sephadex-LH20 acts as a size exclusion chromatography.³⁸ The spectra of the purified material were different in relation of the thermally treated ones in term of the intensity of the carbonyl band of amide groups. It is verified that the signal of the amide around 1660 cm⁻¹ has a higher intensity in comparison with the amine band around 1600 cm⁻¹ for both EDTA : EDA ratio, as presented in the Fig. 2.

Fig. 2 FTIR spectra of HBPAs synthesized from thermal treatment at 160 °C for 2 hours and EDTA : EDA ratio 1 : 1.5 and 1 : 2 (i) before and (ii) after purification by SEC.

The change in the amide signal intensity clearly indicates that the purified materials have higher content of amide groups in relation to the as-synthesized ones. Since the purified materials have higher molecular weight, because small molecular mass oligomers were removed in the purification process, it is possible to affirm that the growth of the molecular weight during the thermal treatment is directly related to the amide formation reaction. The chemical characterization of the polymers was carried out by NMR. In the Fig. 3 and 4 are presented ¹H and ¹³C NMR spectra for the purified HBPAs prepared with different EDTA : EDA ratio at 160 °C for 2 hours, respectively. Using NMR spectra of Fig. 3 and 4 and literature data, attributions were made to the major signs.^35,39,40

Fig. 3 ¹H NMR spectra in D₂O for HBPAs obtained at 160 °C for 2 hours from different EDTA : EDA ratio: (a) 1 : 1.5 and (b) 1 : 2.

Fig. 4 ¹³C NMR spectra in D₂O for HBPAs obtained at 160 °C for 2 hours from different EDTA : EDA ratio: (a) 1 : 1.5 and (b) 1 : 2. Scheme of the chemical structure of a section of HBPA synthesized using as monomers EDTA and EDA. The letters a, b, c, d, e and f correspond to the type of carbon in the structure.

In the Table 1 it is presented the peak attribution to the main hydrogens signals in the NMR spectra. The amide formation can be observed by the presence of the peaks 2, 4, 5 and 9. It can be observed that the formation of the amide after the thermal treatment occurs in the both EDTA to EDA conditions. Peak 2 and 4 are attribute to the methylene protons of the carbon in the core of the EDTA molecule direct linked to the tertiary amine. Peak 2 is attribute to the EDTA molecules with only one of the two carboxyl in one side of the EDTA converted to amide, while is the peak 4 is attributed to the ETDA molecules with complete converted to amide. The comparison between the intensity of peak 2 and 4 can provide an important understanding of the degree of amide formation on the reaction. It can be observed a severe change in the extend of amide formation between the samples with different EDTA : EDA ratio. In the sample prepared with the EDTA : EDA ratio of 1 : 1.5, peak 2 and 4 have intensity that are not very distinct. However, the sample prepared with EDTA : EDA ratio of 1 : 2, peak 4 is much more intense than peak 2. These result indicates content of amide groups formed is much higher in the sample prepared with EDTA : EDA ratio of 1 : 2. The same conclusion can be drawn based in the intensities of peak 5. Peak 5 is the predominant peak in the sample EDTA : EDA 1 : 2 for the methylene group, which is neighbouring to the carbonyl group in the EDTA. Peak 5 referred to methylene groups neighbor to carbonyl in amide groups while peak 3 is referent to the methylene group neighbor to carbonyls in carboxylic acid. In the sample EDTA : EDA 1 : 1.5 the non-modified carbonyl groups are predominant. In the sample EDTA : EDA 1 : 2, peak 5 is a dominant signal while peak 3 is a weak signal.

Table 1 Peak assignment for ¹H NMR of the polymerized EDA and EDTA

Peak	Group	Peak	Group
1		2
3		4
5		6
7		8
9

---

In terms of polymerization these molecules are very interesting, EDTA is a tetrafunctional molecule (AB₄-type) and EDA is bifunctional molecule (AB₂-type), where A and B refer to two reactive functional groups. The NMR results indicate that the stoichiometric sample (EDTA : EDA 1 : 2) had a higher conversion to amide. It suggests that the better organization of the EDTA : EDA crystal provided by the stoichiometric sample induce to a higher polymerization level.

In the Fig. 4 the ¹³C NMR spectra of the samples EDTA : EDA 1 : 1.5 and 1 : 2 are presented. ¹³C NMR spectra shows signals of amide C=O in the region of δ_C 178.3–175.6, related to carbons refereed as a in the schematic chemical structure in Fig. 4, confirming the polymerization of the monomers and HBPAs formation. There are signals in δ_C 182.2–180.0, related to C=O of carboxylate terminal groups (b carbons of the chemical structure of Fig. 4). Signals at δ_C 62.0–60.5 and δ_C 55.7–54.2 are attributed to the carbons linked to the branching units (c and d carbons, Fig. 4). Finally, carbon directly linked to the aminium terminal groups appear in δ_C 42.8–42.0 and δ_C 41.6–40.8 (represented by e and f carbons in Fig. 4).

The molecular weight (M_n) and dispersity index (PDI) of the HBPAs were analysed by GPC in water, Fig. S6 and S7.† HBPAs obtained from EDTA : EDA ratio 1 : 1.5 and 1 : 2 showed very low PDI of 1.30 and 1.12, respectively. The value obtained for the EDTA : EDA is comparable to dendrimers synthesis and it can be considered almost monodisperse. Dispersity values in the range of 1.2 to 1.45 were found for dendritic polymers prepared by multiple steps approach (branch generations).⁴¹ The very low dispersity index found for the HBPAs surpass the expectation, since they are produced by a very simple polymerization reaction performed in just one synthesis step. This result indicates that the solid reaction of organized molecular ions can provide great control of molecular mass.

In addition, in the GPC study it was obtained M_n values of 7.3 kDa for HBPA prepared with EDTA : EDA ratio 1 : 1.5. HBPA obtained from EDTA : EDA 1 : 1.5 has molar weight very close to PAMAM G3-NH₂ (M_n = 6.9 kDa). A slight lower M_n values was obtained for HBPA from EDTA : EDA 1 : 2 5.5 kDa. HBPA prepared with EDTA : EDA ratio of 1 : 1.5 is bigger than HBPA prepared with EDTA : EDA 1 : 2. Even though the molecular weight is slightly difference, it has to be considered the NMR results indicate a higher content of the amide groups in the EDTA : EDA 1 : 2. NMR and GPC results can be correlated to understand a different branching level between the two HBPAs. While the molar weight of the sample from EDTA : EDA 1 : 2 is lower its amide content is much higher due a greater branching level.

HBPAs pH-dependent properties were evaluated by DLS and zeta potential. In the Fig. 5 is presented hydrodynamic diameter and zeta potential as a function of pH. A general trend is observed, values of hydrodynamic diameter and zeta potential increase by decreasing pH values.

Fig. 5 Graphics hydrodynamic diameter and zeta potential due to pH for HBPAs synthesized at 160 °C for 2 hours and with EDTA and EDA ratios: (a) 1 : 1.5, and (b) 1 : 2.

The pH-dependence profile of HBPAs shows three distinct regions of protonation in the zeta potential for the HBPA prepared with EDTA : EDA ratio of 1 : 2 and two regions for HBPA prepared with EDTA : EDA 1 : 1.5. It is in agreement with the distinguish pK_a values of interior tertiary amine and surface primary amines in dendrimers. In the case of PAMAM, interior tertiary amine groups have pK_a around 6.5, while primary amine groups have higher pK_a values around 9.2.^42,43 In addition, terminal carboxylic acid has pK_a around 2.5 and they are prominently deprotonated in the range analysed. Considering that only high branching level is obtained at EDTA : EDA of 1 : 2, the formation of a core with lower pK_a value of internal tertiary amine, creates a new transition in the protonation condition. It has to be thought that the increase content of EDA in the sample also generates a higher content of amine terminal groups that form the transition around pH value of 8, that is not observed for HBPA prepared with EDTA : EDA 1 : 1.5.

In the DLS measurements is observed that the hydrodynamic diameter of both samples are around 25 nm at pH values close to 10. Lowering the pH the polymer particles starts to growth due aggregation of particles caused by the change in the zeta potential. Zeta potential is more negative in higher pH values and it becomes closer to zero as the pH change to higher acidic condition. Lowering the surface charges, the repulsion among the particles became weaker and the van der Waals interactions drive the particle growth. Nevertheless, the ration among EDTA and EDA has an intense effect in the susceptibility to the particle aggregation in function of pH. HBPA from EDTA : EDA 1 : 2 rapidly increases its hydrodynamic diameter in pH close to 9, reaching values higher than 200 nm in the pH range of 9 to 7.5. At lower pH values, HBPA from EDTA : EDA 1 : 2 dispersion becomes unstable due to zeta potential close to zero and the DLS was not able to provide values of hydrodynamic diameter. It occurs because the particle size exceed the maximum particles size measured by the equipment (5 μm). Nonetheless, in the sample prepared by EDTA : EDA 1 : 1.5, hydrodynamic diameter is increasing occurs at pH values below 8. In this case the increase in the hydrodynamic diameter follow a linear trend in function of the pH decrease.

Conclusions

Solid-state synthesis of hyperbranched polyamides (HBPAs) is demonstrated and the achievement of the polymerized products verified and characterized by FTIR, NMR, GPC, DLS and zeta-potential analyses. Solid-state synthesis demonstrated to be a viable approach to produce dendritic polymer in a one-step process. In addition, the synthesis condition based in the simple thermal treatment of the EDTA/EDA crystals in the temperature range of 120 to 160 °C was shown to be effective to induce the condensation reaction. NMR and GPC results indicates that the EDTA and EDA ratio can be used to change the polymers structure due observed changes in the branching levels. It unfolds many possibilities to the HBPAs synthesized by solid-state procedure since its general properties can be tuned by a simple change in the precursors ratio.

Despite the success in the carried out the polymerization synthesis of a HBPAs for the first time by a solid-state reaction, the most important result obtained was the dispersity index of the macromolecules obtained. Dispersity index of 1.3 and 1.12 for the two synthesized HBPAs is similar to the obtained by many steps synthesis of dendrimers. As conclusion, solid-state methodology is an appealing method for dendritic polymers synthesis, since it provides an easy way for large scale synthesis of practically monodisperse polymers.

Acknowledgements

R. S. and A. F. R. acknowledges the financial supports given by CNPq, Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES-Brazil), and Fundação Araucária-Brasil. V. A. T. also thanks the CAPES for the graduate fellowships.

Notes and references

1. R. Silva, M. H. Kunita, E. M. Girotto, E. Radovanovic, E. C. Muniz, G. M. Carvalho and A. F. Rubira, J. Braz. Chem. Soc., 2008, 19, 1224–1229.
2. E. Boisselier, A. K. Diallo, L. Salmon, C. Ornelas, J. Ruiz and D. Astruc, J. Am. Chem. Soc., 2010, 132, 2729–2742.
3. R. Guo, Y. Yao, G. Cheng, S. H. Wang, Y. Li, M. Shen, Y. Zhang, J. R. Baker Jr, J. Wang and X. Shi, RSC Adv., 2012, 2, 99–102.
4. D. Astruc, L. Y. Liang, A. Rapakousiou and J. Ruiz, Acc. Chem. Res., 2012, 45, 630–640.
5. S. E. Stiriba, H. Kautz and H. Frey, J. Am. Chem. Soc., 2002, 124, 9698–9699.
6. R. Silva, E. C. Muniz and A. F. Rubira, Appl. Surf. Sci., 2009, 255, 6345–6354.
7. W. Liu and C.-M. Dong, Macromolecules, 2010, 43, 8447–8455.
8. I. J. Majoros, A. Myc, T. Thomas, C. B. Mehta and J. R. Baker, Biomacromolecules, 2006, 7, 572–579.
9. A. Akesson, M. Cardenas, G. Elia, M. P. Monopoli and K. A. Dawson, RSC Adv., 2012, 2, 11245–11248.
10. L. Zhou, J. L. Geng, G. Wang, J. Liu and B. Liu, ACS Macro Lett., 2012, 1, 927–932.
11. R. M. Crooks, M. Q. Zhao, L. Sun, V. Chechik and L. K. Yeung, Acc. Chem. Res., 2001, 34, 181–190.
12. R. W. J. Scott, O. M. Wilson and R. M. Crooks, J. Phys. Chem. B, 2005, 109, 692–704.
13. S. K. Choi, P. Leroueil, M.-H. Li, A. Desai, H. Zong, A. F. L. Van der Spek and J. R. Baker Jr, Macromolecules, 2011, 44, 4026–4029.
14. M. Salem, S. Rohani and E. R. Gillies, RSC Adv., 2014, 4, 10815–10829.
15. R. Silva, G. M. Pereira, E. C. Muniz and A. F. Rubira, Cryst. Growth Des., 2009, 9, 3307–3312.
16. F. Daebritz, B. Voit, M. Naguib and M. Sangermano, Polymer, 2011, 52, 5723–5731.
17. J. Khandare, M. Calderon, N. M. Dagia and R. Haag, Chem. Soc. Rev., 2012, 41, 2824–2848.
18. K. L. Wooley, C. J. Hawker and J. M. J. Frechet, Angew. Chem., Int. Ed. Engl., 1994, 33, 82–85.
19. K. Inoue, Prog. Polym. Sci., 2000, 25, 453–571.
20. D. A. Tomalia, Prog. Polym. Sci., 2005, 30, 294–324.
21. R. Silva, E. C. Muniz and A. F. Rubira, Langmuir, 2009, 25, 873–880.
22. D. A. Tomalia, Soft Matter, 2010, 6, 456–474.
23. S. M. Grayson and J. M. J. Frechet, Chem. Rev., 2001, 101, 3819–3867.
24. G. J. Chen, L. Tao, G. Mantovani, V. Ladmiral, D. P. Burt, J. V. Macpherson and D. M. Haddleton, Soft Matter, 2007, 3, 732–739.
25. B. I. Voit and A. Lederer, Chem. Rev., 2009, 109, 5924–5973.
26. S. Li, J. Han and C. Gao, Polym. Chem., 2013, 4, 1774–1787.
27. A. Natarajan, C. K. Tsai, S. I. Khan, P. McCarren, K. N. Houk and M. A. Garcia-Garibay, J. Am. Chem. Soc., 2007, 129, 9846–9847.
28. R. Silva, E. C. Muniz and A. F. Rubira, Polymer, 2008, 49, 4066–4075.
29. K. Varga, J. Volaric and H. Vancik, CrystEngComm, 2015, 17, 1434–1438.
30. I. J. Majoros, C. R. Williams, D. A. Tomalia and J. R. Baker Jr, Macromolecules, 2008, 41, 8372–8379.
31. P. Bhattacharya, N. Conroy, A. M. Rao, B. A. Powell, D. A. Ladner and P. C. Ke, RSC Adv., 2012, 2, 7997–8001.
32. K. Liu, Z. Xu and M. Yin, Prog. Polym. Sci., 2015, 46, 25–54.
33. M. Labieniec-Watala and C. Watala, J. Pharm. Sci., 2015, 104, 2–14.
34. L. F. Tian, X. Shu and J. Zhu, Adv. Mater., 2007, 19, 4548–4551.
35. F. D. R. Amado, C. C. Silveira, L. F. Rodrigues Jr, C. A. Ferreira and A. Meneguzzi, Polim.: Cienc. Tecnol., 2008, 18, 244–248.
36. H. Hui, X. D. Fan and Z. L. Cao, Polymer, 2005, 46, 9514–9522.
37. K. Y. Cheah, G. S. Howarth, K. A. Bindon, J. A. Kennedy and S. E. P. Bastian, PLoS One, 2014, 9, e98921.
38. M. Liu, C. Xie, H. Pan, J. Pan and W. Lu, J. Chromatogr. A, 2006, 1129, 61–66.
39. Y. Jin, L. Song, D. Wang, F. Qiu, D. Yan, B. Zhu and X. Zhu, Soft Matter, 2012, 8, 10017–10025.
40. R. Silva, G. M. Carvalho, E. C. Muniz, G. J. Vidotti and A. F. Rubira, e-Polym., 2007, 7, 1574–1585.
41. H. E. Rogers, P. Chambon, S. E. R. Auty, F. Y. Hern, A. Owen and S. P. Rannard, Soft Matter, 2015, 11, 7005–7015.
42. M. H. Kleinman, J. H. Flory, D. A. Tomalia and N. J. Turro, J. Phys. Chem. B, 2000, 104, 11472–11479.
43. A. Kulczynska, T. Frost and L. D. Margerum, Macromolecules, 2006, 39, 7372–7377.

---

Footnote

† Electronic supplementary information (ESI) available: Chemical characterization of products prepared at different syntheses conditions by FTIR. See DOI: 10.1039/c6ra01023g

---