1. Introduction

2. Experimental Procedures

3. Grafting Reactions: The Selectivity Challenge

3.2. Initial and Final Graft Selectivity

The ratio of OH^∙ radicals that reacts either with starch or with monomer determines the selectivity of the reaction in the early stage. This ‘initial selectivity’ is a matter of chance. The probablity to start grafting can be influenced by reaction conditions, for example by the relative concentrations of the AGU in the starch backbone versus AA present at the vicinity of the activated starch site. The initial selectivity may not be the only factor that determines how much grafted starch versus homopolymer is formed by the end of the process. It can be imagined that growing chains of grafted polymer terminate with a radical end group of a homopolymer chain. We can designate this as cross-termination. Wether this happens is also a matter of chance, in which a factor like mobility of the chains in a viscous solution may play a role as well. In literature, reported GE-data are usually those at the end of the reaction, which makes it impossible to see if there is a contribution of cross-terminations to the final selectivity. Calculations based on data reported by Witono et al. [7,12] give a rather unique view on how the parameter GE evolves during some reactions, shown in Figure 4. The calculations are in the supplementary material. These figures may have some margin of error since it they are derived from combining several experimental data, but they certainly allow for trend analysis. In Figure 4, it can be seen that GE rapidly increases in the first minutes of the reaction, then levels off or even slightly decreases, but in the end it is always higher then in the first minutes. With Fenton’s initiator, there is also a slow second step of radical creation from ferric ions (Fe³⁺) [13,14]. It is uncertain if HO₂^∙, the radicals created in that step, are sufficiently reactive to also start grafting reactions, but that is possible since it has been demonstrated that these do react with starch [13]. This can be a contribution to higher graft selectivity since if these radicals do start new grafted chains, that occurs later on in the reaction when most of the monomer has been consumed, hence at low actual M/S, see also section 4.2. However, if this effect can be considered minor, the course of the GE-lines in Figure 4 is mainly caused by cross-terminations, and those may indeed contribute to a higher ‘final selectivity’. The increment of GE during the reaction is higher when there is more monomer in the system (at M/S = 2). This is consistent with the matter of chance since then, there will also be more growing homopolymer chains that a grafted chain can meet to terminate with. The examples in Figure 4 also show that GE% at the end of the reaction is higher when initial selectivity is higher as well, at a lower M/S ratio. To conclude, cross-terminations may compensate to some extent for a low initial selectivity, but the overall grafting result will be better when the selectivity at the initiation stage (initial selectivity) is optimized.

4. Factors that Influence the Graft Selectivity

4.1. The Choice of the Initiator System

In the reviews already mentioned [1,2,3,4] a good overview can be found of the huge variety of initiation methods that has been reported throughout decades of starch grafting research. Most of the systems belong to one of the three different categories according to the principal mechanism of radical creation at the starch backbone. The three principles are listed in Figure 5 and designated ‘class I/II/III’. The respective classes are discussed below, as they are directly related to the potential to obtain good graft selectivity. Also aspects related to use in larger scale reactors are briefly assessed.

• Class I: Activation by radicals created at random in the solution

This is the class of initiation systems which produces radicals throughout the reaction solution. Redox reactions like Fenton’s as used in our work are an example of this class, as are agents that form radicals by thermal dissociation. Examples of these are potassium and ammonium persulfates and organic peroxides like butyl or benzoyl peroxide. All of these systems create radicals more or less at random in the usually aqueous solution, then the radical activity is tranferred to the starch backbone. As already mentioned in paragraphs 1 and 3.1, this can be designated as indirect initiation. Also by the very principle, homopolymer formation is inevitably assiociated with indirect initiation. This class of initiation systems is probably the most easy to scale up, provided that there is a good level of mixing in the reactor.

• Class II: Radicals via a direct reaction with starch

A second class defines initiation agents that also show a redox reaction but in this case, starch-OH groups are the reducing part of the redox couple. Cerium(IV) Ammonium Nitrate (CAN) is used in many graft research projects. Other examples include Manganese or other transition metal ion systems[15]. The reaction mechanisms of Cerium(IV) is clearly described in many literature sources, e.g. [1,3,4]. In principle, such systems are more selective towards grafting as they start from the starch-bound radical. In Liu et al. [16] and in other older literature it is stated that cerium is expensive, but that information appears to be outdated. Still, a major disadvantage of the CAN-system is that it need strond acidic conditions that can degrade the starch backbone [13]. Also, the complex formation with starch is a slow process [4,17], and there is some evidence that the promise of high selectivity is not always obtained [17,18]. The removal of residual cerium from the product after the reaction is never mentioned in literature, but perhaps it can be tolerated in most applications. One aspect of graft initiations that is also hardly ever addressed in literature is the low initiator efficiency, in other words only a small fraction of the radicals created shows up in a succesfull propagating graft chain. From results of Okieimen et al. [19] we could estimate, for Ce(IV)-initiated grafting of ethyl acrylate, an initiator efficiency in the order of 0.001, or 0.1%. Calculation examples are in the supplement. A better but still low initiator efficiency of <1% could be calculated from a paper of In-Hwan Park et al. [20] for ammonium persulfate/sodium metabisulfite initiation of acrylic acid grafting, so the example of cerium is not unique. These numbers are in sharp contrast to industrial polymerizations where initiator efficiency figures of 30-80% are the standard11. When the cost of initiator is a factor of importance in process economics, this aspect must be weighed as well. Still, for the reasons mentioned we consider CAN-initiation, in spite of its selectivity potential, more suitable for laboratory scale studies than for larger scale industrial operation.

• Class III: Activation by high energy or microwave irradiation

Irradiation as a method of starch activation has been investigated by several authors as listed by Meimoun et al. [1]. High energy irradiation was already studied extensively by Reyes et al. [21] in 1963, see also paragraph 6.3. Such systems need expensive equipment, which is an economic setback. More recently, microwave initiation methods, as such or in combination with thermal dissociation radical creating agents, has been reported [1,2] and e.g. by Singh et al. [22]. A common issue with irradiation is the depth of penetration. Microwave rays go no deeper than a few centimeters in water, high energy systems probably more. Still, insuffient or uneven distribution of radicals troughout the reactor is a potential disadvantage for application or irradiation initiation methods in commercial scale large tubular or tank reactors. Like with Cerium salts, we consider class-III initiation methods probably more suited for labscale research.

4.2. Reaction Variables Like Concentrations and Temperature

Reaction variables are a major factor when considering the GP and GE of the grafting reaction. The effect of reaction variables is dependent on the type of grafting reaction system. The reviews mentioned before [1,2,3,4] give a good overview already, so here we only dicuss the most important effects in relation to previous results from our own work. The concentration of initiator usually shows an optimum in both GP and GE. An example of that is shown in Figure 6, giving results published by Khalil et al. [23] with potassium persulfate, a class-I initiator, and acrylamide monomer. At low level of initiator dosage there are simply not enough radicals to start a polymerization so increasing the dosage improves that. However beyond a certain amount of initiator there are too many radicals that cause premature terminations as well as increased homopolymer formation. The reason for the latter is not fully clear but it is speculated that in a further stage of the reaction, there may be a limited number of grafting sites available from the backbone carbohydrate, perhaps due to steric hindrance [4]. The optimum value is different for each specific grafting system so it can only be determined by experiments. From the experimental design study published before [8] we could conclude that the dosage used in our work as mentioned in section 2, an Fe-AGU molar ratio of 1:100 and a 1:10 molar ratio of Fe²⁺ to H₂O₂, are values sufficiently near that optimum. In literature, the effect of temperature shows a pattern that is similar to the influence of the initiator concentration, with the same mechanistic explanation. For our system, it was found that increasing temperature beyond 40^oC gives lower GP and GE, so 40^oC was used in further experimental work [7].

The relative concentrations or the molar ratio of monomer to starch (M/S) appears to be a more important factor than absolute concentrations, in the grafting system with gelatinized starch, acrylic acid and Fenton’s. Unfortunately, the M/S ratio has an opposite influence on the most important result parameters GP and GE, as shown in Figure 7 (data from Witono7). Such a contrasting influence is seen in other literature as well [22,23,24]. The optimum will a trade-off between sufficient functional grafts for an intended application while not to produce a lot of homopolymer.

5. A chemical Engineering Approach to Different Grafting Systems

Table 1. Graft selectivity (GE) values from literature, for five monomers with different water solubility and with direct versus indirect initiation systems.

6. Methods to Improve the Graft Selectivity

6.2. A) Use of a Polymerization Crosslinker

In principle, the addition of a polymerization crosslinker can increase the overall or final selectivity. This remedy however only applies when an application is considered that asks for a network of polymer chains like in superabsorbent materials. There is ample literature on the topic of superabsorbents based on starch, including books and other good reviews [6,34,35]. In these reviews the chemistry of polymerization crosslinking has been described as well as other crosslinking options that are outside the scope of the present paper. N, N’-methylene bisacrylamide (MBAM) is the most common polymerization crosslinking agent and was also used in our laboratory [36]. The crosslinker molecule contains two C=C groups that get incorporated into the polymer structure and make it a three-dimensional network. It is quite imaginable that when one of the C=C has been taken up in a grafted chain, the other C=C can connect to a growing chain of homopolymer. If this would occur in sufficient numbers, the amount of polymer that is connected to starch is increased, in other words a higher GE is obtained at the end of the process. The factors chain mobility and meeting probability are of importance, as is the amount of crosslinker used. In literature on superabsorbent the effect of addition of soluble homopolymer to the grafted network is not discussed in great detail. Most usually, crosslinker dosage is optimized towards a compromise between a high absorption capacity and sufficient chain mobility [6]. Only two times, in different chapters of the book edited by Pradan34, it is briefly noticed that when there is insufficient crosslinking, there will be a soluble fraction in the synthesized superabsorbent material which affects absorption capacity. The presence of a soluble fraction can only mean that some of the polymer chains have not become attached to the 3D-network. Translated to the situation with grafted starch, that would correspond to homopolymer that has not got connected into the network of grafted polymer chains.

A few data that were obtained in our laboratory may further support the idea that polymerization crosslinker may connect homopolymer into the grafted network. These are shown in Figure 9. The data in Figure 9A are not really convincing yet, but a few more recent data in Figure 9B show a clear positive effect of increased crosslinker dosage on the grafting result parameters GP and GE. Since this potential remedy for better GE is limited to applications of grafted starch in superabsorbents, this method to reach a higher final graft selectivity is not explored deeper here.

6.3. B) Preceding Process Step(s) and Creation of Starch Macroradicals before Entering the Monomer

An extra process step prior to grafting makes the whole synthesis procedure more complicated and probably more expensive. An economic evaluation should be made wether a possible higher graft selectivity that may be obtained by additional process steps could be compensated for by lower cost for homopolymer separation. However, such data has not been reported yet in literature. So, in this paper we discuss the scientific aspects only. There are some interesting options to improve grafting results with preceding process steps and more selective starch activation also with water-soluble monomers.

Starch gelatinization is also a preceding process step which is standard for all experiments in our laboratory [7]. The potential advantages were already discussed before [7]. Shortly, by gelatinization the AGU-groups that are inside granules also get exposed and can take part in reactions, where internal diffusion limitations are likely to occur with starch in the granular state. By gelatinization, the starch chains are fully exposed so in fact the ‘effective’ concentration of AGU is increased, which would be beneficial in the competition with monomer molecules for initiating radicals. Also the overall rate of the reaction will be higher, and the issue of uneven distribution of grafts along the starch chains is avoided. Since we did not do experiments with granular starch, we can not make a complete comparison. Data from literature are not completely clear about it7 but several authors confirm an advantageous effect of using gelatinized starch in grafting processes, both on grafting percentage en grafting efficiency, e.g. Park et al. [20] and Hebeish et al. [37].

Vaquez et al. [14] adsorbed Fe²⁺ ions on starch prior to adding the monomer and H₂O₂ to have radicals created in the proximity of starch. This method had limited succes. A reasonable average graft selectivity of 45% was reported, but no comparison was made to situation without pre-adsorption. This system could not overcome the controverse effect of the concentration of the monomer methacrylic acid on GE and GP similar to what is shown in Figure 7. Why the Fe²⁺ ions would attach to starch is not mentioned in this paper, but it is most likely the effect of negatively charged phosphate groups, known to be present in potato starch only and in a ratio of about 1 per 250 AGU [38]. It can be imagined that inserting more phospathe groups in any starch prior to graft copolymerization may have a greater positive effect, especially with Fenton’s. There is no more literature about this idea, except one notition in a book from Hebeish and Guthrie [39], where prior to the grafting of several monomers from cellulose, phosphorous acid was reacted with the backbone molecules with the intention to improve graft selectivity. Also since phosphorylation is a proven modification process in starch industry [40], it would be an interesting topic to investigate further.

Many more premodification methods have been tested and reported in literature on carbohydrate grafting. A complete overview would be far beyond the scope of this article. However, since in most cases the first paper was never followed-up, perhaps they were either unsuccessful or produced unwanted by-products like methods involving halogens. Another reason may be that starch integrity was affected too much, as mentioned by Wilham et al. [41] in substituting starch with allyl groups.

Reyes et al. have published interesting papers on the creation of radicals in dry starch by irradiation with γ-rays from ⁶⁰Co, a radioactive isotope21, [42]. When the activated starch was brought in contact with an aqueous solution with acrylic acid monomer, starch became grafted with an almost 100% selectivity. This is by far the best result ever reported, and these results also suggest that water does not act as a chain-transfer agent for starch-bound radicals since apparently, no homopolymer was found in this system. However and despite a promising pilot-scale study [42], the method never resulted in a real process as far as we know. The application of high energy radiation may have been the bottleneck, both for economic and general safety reasons. Other methods based on chemical pre-initiation, with mixed success, have also been appeared in literature as listed in [13]. In that paper, we reported on an attempt to apply pre-initiation also with Fenton’s reagent. The results were really disappointing since the viscosity of the gel as reflected in the registrated torque values very rapidly deteriorated, as shown in Figure 10 (data from earlier work [7,13]). Oxidative degradation of the starch into smaller fragments is the most plausible reason for the observed loss of viscosity [13]. Since high viscosity or long intact chains of starch are the basis for most applications of grafted starches, we considered this method of pre-initiation not useful to investigate further.

To conclude, perhaps some pre-initiation or pre-modification methods with starch could be feasible and worth to investigate further, but with Fenton’s reagent it was unsuccesful. Therefore we tried the next option.

6.4. C) Keeping the Monomer to Starch Ratio Low, by Dedicated Dosage of Monomer

The disappointing results of pre-initiation with Fenton’s show that there must always be some monomer in the system at the moment of radical generation, else the OH^∙ will affect the starch backbone. Then, a logic idea is to divide onomer addition over several smaller portions that are subsequently fed, to the batch reactor of Figure 1. The addition intervals cannot be too long to ensure there is no monomer depletion but not too short that the effect of low M/S ratio at the grafting sites is overruled. In exploratory experiments we tested five equal portions, the first is mixed in before the initiator is there, then four more were added in five-minute intervals. The principle of this ‘dedicated dosage’ is shown in Figure 11. Experimental results are shown in Table 2. There, the method of adding monomer acrylic acid in five portions in intervals of 5 minutes is shown, and compared to results with the same other conditions but with all of the monomer added before the initiator, so before the start of polymerization reactions. For each condition there are three measurements done a different dates over a period of several months, and also they are done by different students. The nevertheless moderate standard deviation (SDEV) shows that the differences between both dosage methods are significant and reliable. More details, also on the calculation methods, can be found in the ‘supplementary materials’.

The method of adding monomer in portions over time shows a clear improvement of both grafting percentage (17 -> 28 wt%) and graft graft selectivity (19 -> 31 %). With the conventional method, a graft selectivity of over 30% could only be obtained at lower total dosage of monomer. Then, GP was also lower as already shown in Figure 7. This can only mean that the method of dedicated dosage has principally broken the reverse effect and the trade off between GP and GE that is characteristic for graft copolymerization with water solution monomers. This must be considered a success. However, 31% GE is still far from the values that are seen when grafting hydrophobic monomers as the data in Table 1 point out. So there is ample room to further improve the method of dedicated dosage, for example by varying the size of the respective portions and the time intervals. It would be most optimal if monomer could be suppleted at just about the exact moment when the earlier portions becomes depleted. It is also imaginable that feeding the monomer slowly in a continuous fashion, at just about the rate of its disappearance would be even more optimal. Such a refinement of a dedidated dosage method would however need much more experimental effort, perhaps even a complete new research project, so this is still in the future. For now, we have at least demonstrated that the principle of keeping the monomer to starch ratio low at the start of the reaction, while in the end still having added the full amount of monomer, is valid since it leads to better grafting results with the water-soluble monomer acrylic acid.

Table 2. Results of explorations of the dedicated monomer dosage method and comparison with graft polymerization at the same conditions with conventional dosage of acrylic acid monomer.

Table 2. Results of explorations of the dedicated monomer dosage method and comparison with graft polymerization at the same conditions with conventional dosage of acrylic acid monomer.

Method	Run nr	Total M/S molar ratio	Results: Grafting Percentage/Efficiency
Conventional dosage (A) = all monomer before adding the initiator	A-1 A-2 A-3 Average (StDev)	2.0 2.0 2.0	GP wt% 18 % 15 % 17 % 17 (± 1.3) %	GE wt% 20 % 17 % 20 % 19 (± 1.4) %
Monomer dosage in 20% portions, at 5 min intervals (B)	B-1 B-2 B-3 Average (StDev)	2.0 2.0 2.0	28 % 26 % 29 % 28 (± 1.3) %	32 % 29 % 33 % 31 (± 1.7) %

Other conditions: grafting of acrylic acid onto pre-gelatinized starch, T= 40 ^oC, starch load 7.5% in water, Fentons initiator, Fe²⁺: AGU molar ratio 1:100, Fe²⁺-H₂O₂ molar ratio 1:10. More details are provided in the supplementary material and in section 2 (standard run).

The idea to add monomer slowly to obtain a low M/S ratio during the course of the reactions is not completely new for starch grafting, but only seldom seen in literature. A paper of Trimnell and Stout [43] is a rare example of such an earlier application of a method of slow dosage. They show a good grafting efficiency, up to 59%, but with a much lower total monomer acrylic acid dosage, M/S is 4 times lower than in our case. A comparison with the situation of all monomer at the start is not made. Also the use of UV-irradiation initiation and oxidized starch as the backbone makes those results difficult to compare with ours. Brockway [44] has added monomer in portions but together with portions of initiator (Fenton’s). No appreciable effect on the graft selectivity is seen here. However, also this situation is not really comparable with our system since the addition of fresh initiator with every portion of monomer might as well increase the chance of new homopolymers to be formed. The main intention of that work was to increase the frequency of grafting, which was indeed achieved, as also reported in our previous review paper [5].

6.6. An Outlook towards Application of the Low M/S Principle in a Large Scale Continuous Process

It would be an interesting chemical engineering challenge to implement the princple of keeping the monomer concentration low during a starch grafting reaction at larger scale, in a continuous reactor. The first principal choice is between a Continuous Stirred Tank Reactor (CSTR) and a Plugflow Reactor (PR). In a CSTR, the feed stream is efimmediately mixed with the overall contents of the reactor, where the conversion of monomer has already progressed to or near depletion. So, in principle this concept assures a low monomer concentration, particularly when the recator is operated at high conversion. However, when starch is processed in the gelatinized state, there is high viscosity already at the start of the reaction. Cassava starch may be easier to process in the gelatinized state since, due to a relatively high amylopectine content, if forms softer gels than most other starches [7,45]. Upon graft polymerization however, as we reported in a previous article, there can be an increase of the initial viscosity by a factor 3-4 during the process [46]. It would require uneconomic low starch load to keep such a mixture stirrable. In industry, for the same reason, starch is more usually processed in the granular state, thus as a slurry [40]. In principe, it is possible to perform grafting reaction with granular starch, but this has other disadvantages like lower reaction rate and uneven distribution of grafts, as already mentioned in section 6.3. It was also stated there that with gelatinized starch, a better graft selectivity could be expected. For these reasons, a process with starch in the gelatinized state would be preferable.

There are several other flow reactors that have been proven able to handle high viscosity fluids. In our laboratory, a tube reactor with static mixing elements has been demonstrated suitable for hydroxypropylation of gelatinized starch [47], a concept that was also patented [48]. This reactor shows nearly perfect plug flow. There are several more literature sources that report the use of extruder reactor for starch grafting. The example of a bench scale extruder in the work of de Graaf in our laboratory was mentioned before [26]. Other examples are in the work of Willet and Finkenstedt who performed graft copolymerization with acrylamide in a bench scale extruder [49] and Kugler et.al who grafted several acrylic monomers in a labscale extruder [50]. There is another aspect related to polymerization reactions with vinylic monomers that is not yet all that relevant in labscale studies. Such polymerizations are quite exothermic, for example the reaction enthalpy of acrylic acid polymerization, ΔH = -77 kJ/mol [51]. Therefore, any conceptual reactor or process at larger scale needs to have the capacity to remove a substantial amount of heat while a high viscosity does not ease this. The extruders and the static mixer reactors as mentioned are promising also in this respect, but as yet not proven at any larger scale for starch grafting.

With such reactor concepts, a PR may be feasible for the graft modification of starch in the gelatinized state. The principle of keeping the M/S ratio low by monomer dosage in portions could be translated from a batch to a continuous reactor by substituting the factor time by the factor distance. Gradual dosage of monomer can then be brought about by a row of injection ports along the reactor tube, as illustrated by the sketch in Figure 12. This is of course only a projection, a bit premature perhaps since the concept op gradual/portionwise dose had better be refined at labscale first. Still, a continuous system may also be suited for such refinement and optimization, and should start with a small scale reactor. At present, this is just a challenge for future R&D.

7. Conclusions

References

1. Meimoun, J.; Wiatz, V.; Saint-Loup, R.; Julien Parcq, J.; Favrelle, A.; Bonnet, F.; Zinck, P. Modification of starch by graft copolymerization. Starch/Stärke 2018, 70, 1600351. [Google Scholar] [CrossRef]
2. Lele, V.V.; Kumari, S.; Niju, H. Syntheses, Characterization and Applications of Graft Copolymers of Sago Starch. A Review. Starch/Stärke 2018, 70, 1700133. [Google Scholar] [CrossRef]
3. Athawale, V.D.; Rathi, S.C. Graft Polymerization: Starch as a Model Substrate. J. Macromol. Sci. Rev. Macromol. Chem. Phys. 1999, 39, 445–480. [Google Scholar] [CrossRef]
4. Fanta, G.F.; Doane, W.M. Grafted Starches. In Modified Starches: Properties and Uses, O.B. Wurzburg, Editor. 1986, CRC Press.
5. Noordergraaf, I.W.; Fourie, T.K.; Raffa, P. Free-Radical Graft Polymerization onto Starch as a Tool to Tune Properties in Relation to Potential Applications. A Review. Processes 2018, 6, 31. [Google Scholar] [CrossRef]
6. Bucholz, F.L.; Graham, A.T. Modern Superabsorbent Polymer Technology; Wiley VCH: New York, NY, USA, 1998. [Google Scholar]
7. Witono, J.R. New Materials by Grafting of Acrylic Acid onto Cassava Starch. Ph.D. Thesis, Chemical Engineering Department, University of Groningen, Groningen, The Netherlands, 2012. [Google Scholar]
8. Witono, J.R.; Noordergraaf, I.W.; Heeres, H.J.; Janssen, L.P.B.M. Graft Copolymerization of Acrylic Acid to Cassava Starch—Evaluation of the Influences of Process Parameters by an Experimental Design Method. Carbohydr. Polym. 2012, 90, 1522–1529. [Google Scholar] [CrossRef]
9. Parkatzidis, K.; Wang, H.S.; Truong, N.P.; Anastasak, A. Recent Developments and Future Challenges in Controlled Radical Polymerization: A 2020 Update. Chem 2020, 6, 1575–1588. [Google Scholar] [CrossRef]
10. Witono, J.R.; Marsman, J.H.; Noordergraaf, I.W.; Heeres, H.J.; Janssen, L.P.B.M. Improved homopolymer separation to enable the application of 1H-NMR and HPLC for the determination of the reaction parameters in de graft copoly merization of acrylic acid onto starch. Carbohydr. Res. 2013, 370, 38–45. [Google Scholar] [CrossRef]
11. Odian, G. Principle of Polymerization, 4th ed.; Wiley: Hoboken, NJ, USA, 2004. [Google Scholar]
12. Witono, J.R.; Noordergraaf, I.W.; Janssen, L.P.B.M.; Heeres, J.J. A new kinetic model for the graft copolymerization of acrylic acid onto cassava starch. 9^th World Congres of Chemical Engineering (WCCE-9). Aug.2013, Seoul, Korea.
13. Witono, J.R.; Noordergraaf, I.W.; Heeres, H.J.; Janssen, L.P.B.M. Torque measurement as a tool to monitor the breakdown of cassava starch gels, by the effect of Fenton’s initiator for graft copolymerization. Results Chemistry 2022, 4, 100314. [Google Scholar] [CrossRef]
14. Vazquez, B.; Goni, I.; Gurruchaga, M.; Valero, M.; Guzman, G.M. A study of the graft copolymerization of methacrylic acid onto starch using the H₂O₂/Fe⁺⁺ redox system. J. Polym. Sci. 1989, A27, 595–603. [Google Scholar] [CrossRef]
15. Gao, J.; Yu, J.; Wang, W.; Chang, L.; Tian, R. Comparison of transition metals in the graft copolymerization of vinyl monomers to starch. J. Macromol. Sci. Pure Appl. Chem. 1998, A35, 483–494. [Google Scholar] [CrossRef]
16. Liu, Y.; Li, J.; Yang, L.; Shi, Z.; Deng, K. Graft copolymerization of methyl methacrylate onto starch using potas sium ditelluratocuprate(III). J. Macromol. Scic Pure and Appl. Chem. 2004, A41, 1025–1035. [Google Scholar] [CrossRef]
17. Gurruchaga, M.B.; Goni, I.; Vazquez, B.; Valero, M.; Guzman, G.M. An Approach to the Knowledge of the Graft Polymerization of Acrylic Monomers onto Polysaccharides Using Ce (IV) as Initiator. Journal of Polymer Science: Part C: Polymer Letters 1989, 27, 149–152. [Google Scholar] [CrossRef]
18. Okieimen, F.E.; Said, O.B. Studies on the graft copolymerization of methyl methacrylate onto starch. Acta Polym. 1989, 40, 708–710. [Google Scholar] [CrossRef]
19. Okieimen, F.E.; Egharevba, F.; Jideonwo, A. Graft copolymers of starch and (poly)ethyl acrylaat. Angew. Makromol. Chem. 1991, 184, 20–26. [Google Scholar] [CrossRef]
20. Park, I.H.; Song, S.Y.; Song, B.K. Graft polymerization of acrylic acid onto corn starch in aqueous isopropanol solution. Angew. Makromol. Chem. 1999, 267, 20–26. [Google Scholar] [CrossRef]
21. Reyes, Z.; Syz, M.G.; Huggins, M.L. Grafting acrylic acid to starch by preirradiation. J. Polym. Sci. 1963, C23, 401–408. [Google Scholar] [CrossRef]
22. Singh, V.; Tiwari, A.; Pandey, S.; Singh, S.K. Microwave-accelerated synthesis and characterization of potato starch- g-poly(acrylamide. Starch/Staerke. 2006, 58, 536–543. [Google Scholar] [CrossRef]
23. Khalil, M.I.; Mostafa, K.M.; Hebeish, A. Graft polymerization of acrylamide onto maize starch using potassium persulfate as initiator. Angew. Makromol. Chem. 1993, 213, 43–54. [Google Scholar] [CrossRef]
24. Athawale, V.D.; Rathi, S.C. Effect of chain length of the alkyl group of alkyl methacrylates on graft polymerization onto starch using ceric ammonium nitrate as initiator. Eur. Polym. J. 1997, 33, 1067–1071. [Google Scholar] [CrossRef]
25. Fanta, G.F.; Burr, R.C.; Doane, W.M.; Russell, C.R. Influence of starch granule swelling on graft copolymer Composition. A comparison of monomers. J. Appl. Polym. Sci. 1971, 15, 2651–2660. [Google Scholar] [CrossRef]
26. De Graaf, R.A. The use of twin screw extruders as starch modification reactors. PHD-Thesis, Groningen University, 1995.
27. Athawale, V.D.; Rathi, S.C. Role and relevance of polarity and solubility of vinyl monomers in graft polymerization onto starch. React. Funct. Polym. 1997, 34, 11–17. [Google Scholar] [CrossRef]
28. Trimnell, D.; Fanta, G.F.; Salch, J.H. Graft polymerization of methyl acrylate onto granular starch: comparison of the Fe²⁺/H₂O₂ and ceric initiation systems. J. Appl. Polym. Sci. 1996, 60, 285–292. [Google Scholar] [CrossRef]
29. Gao, J.; Tian, R.; Zhang, L. Graft copolymerization of vinylic monomers onto stach initiated by transition metal-thiourea redox systems. Chin. J. Polym. Sci. 1996, 14, 163–171. [Google Scholar]
30. Shogren, R.L.; Willet, J.L.; Biswas, A. HRP-mediated synthesis of starch-polyacrylamide grafts copolymers. Carbohydr. Polym. 2009, 75, 189–191. [Google Scholar] [CrossRef]
31. Okieimen, F.E.; Nkumah, J.E.; Egharevba, F. Studies on the grafting of acrylic acid to starch. Eur. Polym. J. 1989, 25, 423–426. [Google Scholar] [CrossRef]
32. Kislenlo, V.N. Emulsion graft polymerization: mechanism of formation of dispersions. Colloids Surf. A: Physi Cochemical Eng. Asp. 1999, 152, 199–203. [Google Scholar] [CrossRef]
33. Khaddazh, M.; Gritskova, I.A.; Litvinenko, G.I. An Advanced Approach on the Study of Emulsion Polymerization: Effect of the Initial Dispersion State of the System on the Reaction Mechanism, Polymerization Rate, and Size Distribution of Polymer-Monomer Particles, in: Polymerization Edited by Ailton De Souza Gomes, Intech Open, Rijeka, Croatia, 2019.
34. Pradan, S. Mohanty, S. (editors). Bio-based Superabsorbents. Recent Trends, Types, Applications and Recycling. Springer Nature Singapore Pte Ltd. 2023.
35. Athawale, V.D.; Lele, V. Recent trends in hydrogels based on starch-graft-acrylic acid: a review. Starch/Stärke 2001, 53, 7–13. [Google Scholar] [CrossRef]
36. Witono, J.R.; Noordergraaf, I.W.; Heeres, H.J.; Janssen, L.P.B.M. ; Water Absorption, Retention and the Swelling Characteristics of Cassava Starch Grafted with Polyacrylic Acid. Carbohydrate Polymers 2014, 103, 325–332. [Google Scholar] [CrossRef]
37. Hebeish, A.; Zahran, M.K.; El-Rafie, M.H.; El-Tahlawy, K.F. Preparation and characterisation of poly(acrylic acid) starch polyblends. Polym. Polym. Compos. 1996, 4, 129–141. [Google Scholar] [CrossRef]
38. Van Warners, A. Modification of starch by reaction with ethylene oxide. PhD-thesis Groningen University, 1992.
39. Heibish, A.; Guthrie, J.T. The Chemistry and Technology of Cellulosic Copolymers; Springer: New York, NY, USA, 1981. [Google Scholar]
40. Bemiller, J.; Whistler, R.L. Starch : chemistry and technology. eBook©2009 3rd ed., London : Academic, 2009. Printed book: Elsevier Academic Press Burlington, MA :, 2009.
41. Wilham, C.A.; McGuire, T.A.; Rudolphi, A.S.; Mehltretter, C.L. Polymerization studies with allyl starch. J. Appl. Polym. Sci. 1963, 7, 1403. [Google Scholar] [CrossRef]
42. Reyes, Z.; Clark, C.F.; Comas, M.; Russell, C.R.; Rist, C.E. Continuous Production of Graft Copolymers of Starch with Acrylamide and Acrylic Acid by Electron Preirradiation. Nucl. Appl. 1969, 6, 509–517. [Google Scholar] [CrossRef]
43. Trimnell, D.; Stout, E.I. Grafting acrylic acid onto starch and poly(vinyl alcohol) by photolysis. J. Appl. Polym. Sci. 1980, 25, 2431–2434. [Google Scholar] [CrossRef]
44. Brockway, C.E. Efficiency and frequency of grafting of methylmethacrylate to granular corn starch. J. Polym. Sci. 1964, 2, 3721–3731. [Google Scholar]
45. Swinkels, J.J.M. Composition and Properties of Commercial Native Starches. Starch/Staerke 1985, 37, 1–5. [Google Scholar] [CrossRef]
46. Witono, J.R.; Noordergraaf, I.W.; Heeres, H.J.; Janssen, L.P.B.M. Rheological behavior of reaction mixtures during the graft copolymerization of cassava starch with acrylic acid. Polym. Eng. Sci. 2017, 57, 1285–1292. [Google Scholar] [CrossRef]
47. Lammers, G.; Beenackers, A.A.C.M. Heat transfer and the continous production of hydroxypropyl starch in a static mixer reactor. Chem. Eng. Sci. 1994, 49, 5107. [Google Scholar] [CrossRef]
48. Lammers, G.; Balt, L.; Beenackers, A.A.C.M. Method of modifying starch, EP0778850A1, European Patent Offiice. 1994.
49. Willet, J.L.; Finkenstadt, V.L. Comparison of Cationic and Unmodified Starches in Reactive Extrusion of Starch–Polyacrylamide Graft Copolymers. J. Polym. Environ. 2009, 17, 248–253. [Google Scholar] [CrossRef]
50. Kugler, S.; Spychaj, T.; Wilpiszewska, K.; Gor, K. Starch-Graft Copolymers of N-Vinylformamide and Acrylamide Modified with Montmorillonite Manufactured by Reactive Extrusion. J. Appl. Polym. Sci. 2013, 127, 2847–2854. [Google Scholar] [CrossRef]
51. Evans, A.G.; Tyrrall, E. Heats of polymerization of acrylic acid and derivatives. J. Polym. Sci. 1947, 2, 387–96. [Google Scholar] [CrossRef]