Effect of iron oxide nanoparticles on the thermal characteristics of supramolecular, dendritic and macromolecular capping agents

The present study aimed at the enhancement of the thermal characteristics of different capping agents belonging to different macromolecular families, with the incorporation of iron oxide nanoparticles. Iron oxide nanocomposites were prepared following a simple reduction protocol using sodium borohydride (NaBH4) succeeded by air oxidation. Three iron oxide composites were prepared using different biocompatible capping agents. The capping agents used in the present study were supramolecular β-cyclodextrin, dendritic hyperbranched polyglycerol and macromolecular starch. The chemical nature of the composites was successfully confirmed by spectroscopy. The well-characterized composite systems were studied for their thermal characteristics using TG/DTG and DSC measurements. The primary focus of the study was to analyse any change in the thermal behaviour of the pure encapsulating systems on incorporation with the iron oxide species. The analysis was conducted by comparing the TG/DTG and DSC measurements of the pure capping agents- cyclodextrin, hyperbranched polyglycerol and starch with their respective nanoparticle incorporated composites. All the three capping agents were found to decompose in the temperature range 250–450 °C But these systems on incorporation of iron oxide nanoparticles have gained more thermal stability. Their decomposition shifted to a higher temperature with broadened curves suggestive of a slow and steady degradation pattern. DSC studies showed that the incorporation of IONPs could effectively prevent melting of the capping agents by which they could be propagated to a wide range of applications.


Introduction
Macromolecular materials belonging to different classes have been in various applications that attract the scientific community. For the past few decades, these giant molecules apart from their conventional characteristic applications, add more colour to the exciting world of materials science. They can be used as capping agents for the tiny nanoparticles to stabilize as well as enhance their characteristic properties (Philip and Kuriakose 2018;Jose and Kuriakose 2017). Both of them the nanoparticles and the macromolecular agents can be mutually benefitted from this encapsulation or incorporation. They are sometimes rather defined as nanocomposites because of this symbiotic effect on their properties. Research on one hand explores the property enhancement of nanoparticles by encapsulation with these stabilizing agents, while on the other hand it does emphasize on the improved characteristics of the macromolecular entities with the incorporation of the nanoparticles (Philip and Kuriakose 2019a).
The property enhancement of a macromolecular matrix by incorporation of various organic/inorganic or micro/nanoinclusion agents is actively pursued in polymer research (Sawada and Ando 1998;Kumar et al. 2000;Bian et al. 2002). Nanocomposites in their reduced size dimensions offer a wide range of superior properties when compared to their conventional microcomposites. These high-performance hybrid materials come from the need for significant improvements in the interfacial adhesion between the polymer matrix and the reinforcing material because the organic matrix is relatively incompatible with the inorganic phase. In recent years, polymers containing inorganic nanoparticles 1 3 have become one of the prime concerns of frontline research in materials science (Agag et al. 2001;Huang et al. 2000).
The stabilization of nanoparticles using macromolecules can facilitate the surface properties and many size-dependent properties of nanoparticles. On the other hand, these macromolecules which are either hydrophilic or lipophilic can be transformed into more solvent resistant with the introduction of inorganic nanoparticles especially of metallic origin. The nanoparticles can be loaded into the polymer matrix by different methods which can be broadly categorized as: (1) precursor metal salts along with the polymer in a solvent is reduced to their respective metal nanoparticles during stirring, heating or solvent evaporation Kuriakose 2018b, 2021;Lee et al. 2006); (2) a metallic precursor dissolved in a monomer can be reduced during polymerization (Lu et al. 2005;Haes et al. 2004); (3) a colloidal precursor solution is mixed with either the monomer or polymer solution and reduced via route (1) or (2) (Mibhele et al. 2003); and (4) a polymer matrix is impregnated with a precursor solution which then undergoes reduction or thermolysis (Boyes et al. 2003;Yoda et al. 2004). Metal nanoparticles uniformly dispersed in a polymer matrix furnish these composite materials with strong functional possibilities in catalysis (Zhong et al. 2005), sensing (Thurn-Albrecht et al. 2001), optical (Xu et al. 2001), magnetic (Chou et al. 1994;Baranauskas et al. 2005) or electrical (Delamarche et al. 2003;Luigi and Carotenuto 2005) properties. There are also evidences for the thermal property enhancement of these macro-stabilizing or capping agents on introduction of nanomaterials (Bian et al. 2002;Zhang et al. 2014). Therefore, many researchers have devoted their work to the development of these novel hybrid materials with improved performance (Aymonier et al. 2003).
The selection of polymers to build the matrix for these composites depends on many factors like their hydrophilicity, solubility, extent of polymerization in the presence of precursor, miscibility with the precursor, glass transition, melting temperature, method of preparation and so on. Rather simply, all polymers are not good candidates for the fabrication of nanocomposites. Our study used three macromolecular matrices for the preparation of iron oxide-induced nanocomposites, viz. supramolecular β-cyclodextrin (β-CD), dendritic hyperbranched polyglycerol (HPG) and macromolecular starch. All of them were randomly selected for their extensive hydrophilicity and intense biocompatibility. All of these three macromolecular species have already proven their proficiency in making themselves effective capping agents for metallic nanoparticles (Philip and Kuriakose 2018, 2019a, 2021. The availability of a large number of hydroxyl groups in these capping agents could be made use in the surface modification of the encapsulated nanoparticles rendering them more hydrophilic. Also, their biocompatibility can be a factor for using them in biomedical as well as any other applications. In the present study, these macromolecules were used to make biocompatible iron oxide nanocomposites which are significantly important in the field of biomedicine. These giant molecules could make the respective composites more water soluble. Thermal stability of these composites was the main concern of the present study since iron oxide nanoparticles are recently studied for their use in modern treatment protocols like magnetic hyperthermia which is a thermally sensitive procedure. All these factors justify the selection of these three macromolecules for the synthesis of iron oxide nanocomposites in the current study. Cyclodextrins, composed of glucose units linked by α-(1,4)-glycosidic linkage, contain a somewhat lipophilic central cavity and a hydrophilic outer surface (Valle 2004;Magnúsdóttir et al. 2002). β-cyclodextrin is the most accessible and cheap among the different classes of cyclodextrins (Shahgaldian and Pieles 2006). The propensity of cyclodextrins to form inclusion complexes is enabling them to be used in various fields of medicine like drug delivery. They possess supramolecular cages that can entrap small sized particles like nanomaterials through host-guest type interactions, which make them practically available for a wide range of applications. Since they can easily link covalently or non-covalently with each other, cyclodextrins form supramolecular aggregates which are capable for host-guest interactions. It is in this cavity formed by these supramolecular aggregates that the nanoparticle-like small molecules are trapped.
Hyperbranched polyglycerols explore the salient features of encapsulation by means of their bizarre potentiality for host-guest interactions. These giant molecules are peculiar for their phenomenal biocompatibility, low viscosity, high solubility and their three dimensional dendritic architecture (Frey and Haag 2002;Voit and Lederer 2009;Satoh 2012). The third capping matrix starch belonging to the family of macromolecules is a polysaccharide with many glucose units linked through glycosidic bonds and is one of the most common biocompatible polymers.
The study aimed at the thermal stability enhancement of these capping agents on incorporation of iron oxide nanoparticles (IONPs). Stabilized iron oxide nanoparticles often exhibit superparamagnetism Kuriakose 2021, 2019b) which can be propagated to be applied in most modern, suave theranostic techniques including hyperthermia, MRI contrasting and so on. Hence, the thermal properties of iron oxide nanocomposites are studied in detail.

Instrumentation
FTIR spectra of the samples were recorded as KBr discs in the frequency range 4000-400 cm −1 on Shimadzu-400 FTIR spectrophotometer. Microscopic characteristics of all the samples were studied using Jeol/JEM 2100 high-resolution transmission electron microscope (HRTEM). TG/DTG and DSC measurements were taken on a PerkinElmer thermal analyser. TG/DTG and DSC to measure mass change and heat flow rate were recorded simultaneously from 35 to 700 °C under nitrogen atmosphere at a heating rate of 10 °C/minute.

Synthesis of cyclodextrin/hyperbranched polyglycerol/starch-iron oxide nanoparticle composites
500 mg of the capping agent (β-CD/HPG/starch) was dissolved in 100 ml Milli-Q water. 100 ml (0.03 M, 0.8340 g) FeSO 4 .7H 2 O was also prepared in Milli-Q water. Each solution was sonicated separately by a probe sonicator at 25 W for 5 min to ensure complete and effective dissolution. After sonication, the Fe 2+ precursor solution was added dropwise to the aqueous solution of the encapsulating agent. Now the capping agent and precursor mixture was stirred for 4 h using a magnetic stirrer at 1000 rpm. 30 ml freshly prepared solution of sodium borohydride (0.5 M, 0.5675 g) was added with vigorous stirring till the completion of Fe(0) formation, and the solution was stirred further for 1 h. The filtered black Fe(0) was air oxidized to reddish brown α-Fe 2 O 3 , at room temperature. The sample was heated at 200 °C to oxidize all hydroxides as described in Philip and Kuriakose (2021). The composite corresponding to each capping agent and their representative notations used throughout this paper is denoted in Table 1.

Sample preparation for FTIR and HRTEM characterization
FTIR spectra of the four developed systems were recorded as KBr discs in the frequency range 4000-400 cm −1 on Shimadzu-400 FTIR spectrophotometer. About 10 mg of the sample is well mixed with 200 mg fine KBr powder.
Degassing is performed to eliminate air and moisture from the powder. The mixture is then finely pulverized and put into a pellet-forming die. The powder is dried at approximately 110 °C for three hours. A force of approximately 8 tons is applied under a vacuum of several mm Hg for several minutes to form transparent pellets. When performing measurements, the background can be measured with an empty pellet holder, with a pellet of KBr only, can inserted into the sample chamber to correct infrared light scattering losses in the pellet and for moisture adsorbed on the KBr.
HRTEM images of all the developed samples were recorded using Jeol/JEM 2100 high resolution transmission electron microscope (HRTEM).TEM grids were prepared by dispersing dried powder samples in isopropanol and dropping the solution onto a carbon-covered grid.

TG/DTG and DSC measurements
Thermogravimetric (TG) and differential (DTG) thermal analyses and differential scanning calorimetric studies were performed on a PerkinElmer thermal analyser. The thermal studies were used to establish the mass composition and thermal stability of each of the stabilizing macromolecules and to monitor any change on incorporation of iron oxide nanoparticles. Each of the developed nanocomposites was compared with the respective pure stabilizing agents used for their thermal properties. The samples for analysis were powdered and heated 35 °C-700 °C at a heating rate of 10 °C/ minute, under nitrogen atmosphere. The weight loss was measured in each case.

FTIR and HRTEM characterization
Fourier transform infrared spectroscopy and high-resolution transmission electron microscopy were used to confirm the successful synthesis of the newly developed composites. FTIR spectrum of the pure β-CD and IONP-CD nanocomposite was recorded as KBr discs in the wavenumber range of 4000-400 cm −1 (Fig. 1).
Pure The TEM image (Fig. 1) shows that the nanoparticles are well dispersed in the stabilizing agent medium thus confirming the successful formation of the nanocomposite. The image recorded at 100 nm resolution shows that the particles fall well below this size range, with the majority of them below 50 nm.
Pure The peaks below 600 cm −1 in IONP-HPG corresponding to Fe-O stretching vibrations confirm the successful encapsulation of iron oxide nanoparticles into the dendritic scaffolds.
The TEM image of IONP-HPG (Fig. 2) establishes a uniform distribution of the nanoparticles, all within a size range around 20 nm.
Pure Here also, the two dips below 600 cm −1 confirm the successful development of the nanocomposite.
The HRTEM image of IONP-STARCH (Fig. 3) shows that the nanoparticles are well dispersed uniformly. The size range of the nanoparticles, as in the case of IONP-HPG, falls around 20 nm.

Enhancement of thermal stability capping agents on incorporating iron oxide nanoparticles
The different thermal decomposition steps and thermal events of all the samples under study were monitored using TG/DTG and DSC measurements, respectively. TG/DTG measurements were recorded to monitor the thermal degradation of each sample with increase in temperature from 35 to 700 °C at 10 °C/min. TGA is a technique in which the weight of the sample was measured as a function of temperature, under controlled heating. The analysis gives an overall idea about the thermal stability of the samples. Different thermal events for each sample were monitored using DSC measurements in the same temperature range. Thermal characteristics of IONP incorporated nanocomposites were compared with those of the respective capping agents in their pure state.
TG/DTG curve of pure β-CD (Fig. 4a) showed two mass loss steps. The minor one centred at 97.82 °C contributing about 10.95% weight loss of the initial weight can be accounted for the loss of water molecules associated externally or internally with the supramolecule. The major weight loss of 67.98% at 324.61 °C is of course due to decomposition of cyclodextrin molecules (Trotta et al. 2000). At 700 °C, only 5.97% weight remained.
While TG/DTG curve of IONP-CD aggregates (Fig. 4  b) exhibited three major mass losses. The initial one at 63.35°Cwith a weight loss of 2.21% is obviously due to loss of external water molecules at the sample surface and the one at 258.78 °C contributing a total weight loss of 2.55% is due to the loss of chemically bonded or interstitially trapped water molecules in the sample.
The major loss of the aggregate appeared at 357.50 °C with 11.246% weight loss, which can be attributed to cyclodextrin decomposition on comparison with (Fig. 4a). It is quite worth to mention the increased stability of the supramolecule on incorporation with IONPs. The chemical decomposition of pure β-CD (Fig. 4a) started at 277.77 °C and reached a maximum at 324.61 °C, whereas on incorporation of IONPs, the decomposition of cyclodextrin started only at 296.86 °C with a maximum at 357.50 °C (Fig. 4b). This clearly indicates the increased thermal stability of β-CD on encapsulation of IONPs.
Differential scanning calorimetry was performed on all the samples synthesized to monitor the various thermal events. The samples were heated from 35 to 700 °C at the rate of 10 °C/minute. Figure 5 shows the DSC curves of pure cyclodextrin and IONP-CD nanocomposite. Pure CD showed (Fig. 5a) two endothermic peaks at 100.82 °C and 315.84 °C, which, respectively, corresponds to evaporation of water molecules and melting of CD, which can be correlated to the TG/DTG analysis (Fig. 4a). On incorporation of IONPs, melting of CD occurs at a lower temperature of 270.18 °C with an increase in ∆H (Fig. 5b).
The TG/DTG curve of pure HPG noted two mass losses: a minor one at 117.02 °C comprising of 8.96% mass loss and a major one at 402.47 °C which witnessed 67.71% mass loss (Fig. 6a). The initial weight loss can be attributable to loss of free water and water linked through hydrogen bonds and the second one to thermal degradation of the dendritic macromolecule (Li et al. 2015).
IONP-HPG aggregates displayed three major decompositions (Fig. 6b). The first one below 100 °C with a maximum at 52.19 °C is of no doubt due to loss of superficial water molecules present at the sample surface. The second decomposition starting at 176.89 °C and ending at 482.60 °C included two major dips that may involve a combination of loss of chemically bonded water molecules in the interstitial sites of IONP solid network and HPG degradation. From the nature of the second dip, we can understand that the decomposition of HPG shifted to a higher temperature range on encapsulation of IONPs in comparison with pure HPG.
In contrast to IONP-CD aggregates, IONP-HPG aggregates marked a third mass decomposition in the range 613.53 °C-695.74 °C with a maximum at 662.64 °C. Decomposition of HPG in the second weight loss step may result in evolution of oxygen and other gases which may result in crystal defects (Bora et al. 2012) or formation of any other oxides or compounds of iron (Carp et al. 2010). The decomposition of these newly formed compounds result in the third mass loss centred at 662.64 °C. Figure 7 describes the DSC curves of (a) pure HPG and (b) IONP-HPG nanocomposite. Pure HPG showed 3 endothermic processes. The first one at 121.95 °C is undoubtedly due to the loss of water molecules and the third one at 422.65 °C due to melting of HPG. But the endothermic peak at 379.07 °C, where no weight loss was observed, is probably the glass transition temperature (Tg) of this dendrimer.
IONP-HPG composite showed only one endothermic peak at 257.30 °C, which may be due to evaporation of trapped water molecules as described by the second weight loss of 21.46% in TG/DTG (Fig. 6b). On incorporation of IONPs, melting of HPG is found to disappear which justifies the perfect encapsulation. Figure 8a represents the TG/DTG curve of pure starch in which two mass loss steps were observed. The first sudden loss of 10.98% at 76 °C undoubtedly confirms the escape of surface water molecules. The second decomposition at 303.42 °C constitutes the major mass loss which is accountable for the carbohydrate degradation.
In case of IONP-STARCH composite (Fig. 8b), the first weight loss of 11.63% below 200 °C can be attributed to evaporation of surface water molecules or escape of chemically bonded or interstitial water molecules trapped in the crystal lattice of IONP. The second decomposition of 14.12% weight loss indicates the macromolecular disintegration. But the broadening of the decomposition peak from 220.71 to 214.83 °C with a maximum decomposition at 304.23 °C, in comparison with that of pure starch where the decomposition peak was more sharp and steep, confirms a slow decomposition of starch on encapsulation of IONP into the macromolecular scaffolds. IONP-STARCH aggregates as in the  (Carp et al. 2010). Or the increased oxygen evolution from decomposition of starch may result in some crystal defects due to hexagonal distortion that can cause increased compressive stress on the hematite surface (Bora et al. 2012).
Pure starch showed three endothermic thermal events (Fig. 9a). The one at 83.49 °C can be accounted for the water loss. The second one 279.59 °C is due to melting of starch and a weak peak with a very small ∆H (4.28 J/g) at 325.81 °C may be related to the oxidation reaction of the solid carbonaceous residue and is termed as a region of 'glowing combustion' (Rudnik et al. 2006). IONP-STARCH aggregates showed two endothermic curves at 102.59 °C and 138.66 °C, which can be associated with the evaporation of external and internal water molecules as explained by the first thermal weight loss in TG/DTG analysis.
The different thermal decomposition steps observed in TG/DTG analysis for all the IONP incorporated composite samples are enlisted in Table 2 and compared with those of the respective capping systems (marked in red).
The major thermal events (from DSC measurements) of all the synthesized IONP incorporated composite samples compared with the corresponding stabilizing systems (marked in red) are summarized in Table 3.
Iron oxide nanoparticles are of great advantage in many advanced biomedical procedures including magnetic hyperthermia, which is useful even in cancer treatment. The encapsulation of these medically relevant metallic nanoparticles into macromolecules of biocompatible origin can catalyse their potentialities. In applications like magnetic hyperthermia, the superparamagnetic nanoparticles are allowed to heat themselves under the influence of an external magnetic field. And also, the developed systems in our studies have clearly shown that the macromolecules within which these nanoparticles are encapsulated have greatly enhanced their thermal stability by incorporation of nanoparticles. The thermal degradation of the stabilizing macromolecules has been slowed down by encapsulation of metallic iron oxide nanoparticles. This synergic enhancing effect of macromolecules and the IONPs reveal that they could be developed into well-designed multi-tasking potential candidates that finds applications in the most advanced realms of therapeutic science with special reference to magnetic hyperthermia and targeted drug delivery.

Conclusions
The thermal studies performed on the newly developed iron oxide nanocomposites with β-cyclodextrin, hyperbranched polyglycerol and starch revealed that IONPs are suitable candidates for the thermal characteristic enhancement of macromolecular capping agents of different families and origin. All the pure stabilizing systems in the present study have poor thermal stabilities and decompose in the temperature range 250-450 °C. On encapsulation of IONPs, these systems have gained more thermal stability, especially CD where the decomposition shifted to a higher temperature. All the dips corresponding to decomposition of the encapsulating medium broadened on encapsulation of IONPs, which suggests a slower degradation. DSC studies revealed that encapsulation of IONPs have prevented melting of CD, HPG and STARCH which evidently confirms the effective encapsulation of nanoparticles into these systems. These systems can be explored for different functional utilities where the thermal properties of either the iron oxide or the macromolecular entities are scientifically exploited.