Morphology and structure of acorn starches isolated by enzymatic and alkaline methods

2.1 Materials

Acorns from Quercus rotundifolia (QR) and Quercus suber (QS) were collected in “montados” located in Idanha-a-Nova. Three sets of 1 kg of each acorn fruits were randomly harvested at the maturity stage, dried, and milled as previously reported by Correia, Leitão, and Beirão-da-Costa (2009).

2.2 Starch extraction methods

Acorn starches were isolated according to the following methods (Correia and Beirão-da-Costa 2012).

2.3 Scanning electron microscopy

A scanning ISI-D 130 electron microscope (International Scientific Instrument) was used to observe the acorn starch samples. Samples were applied on an aluminum stub using double-sided adhesive tape, and the starch was coated with gold–palladium (80:20). An accelerating voltage of 10 kV was used.

2.4 Particle size and distribution analysis

To determine the size of acorn starch granules, a laser particle size analyzer Mastersizer X (Malvern Instruments Ltd, Malvern, United Kingdom) was used. A polydisperse mode of analysis and a 300 mm lens were used. Starch samples were dispersed in isopropyl alcohol to obtain an obscuration of 15–20%, using an equipment circulation unit, with mechanical shear and ultrasound for 5 min to disperse the starch clots. Measurements were taken at 2 min intervals. Size distributions were determined at three replications of duplicate samples (n = 6). Results were expressed in terms of volume (%) occupied by starch granules. The classes defined based on Lindeboom et al. (2004) scale was used to classify starches: large (>25 µm), medium (10–25 µm), small (5–10 µm), and very small (<5 µm).

2.5 Color evaluation

A Chroma Meter CR-300 Minolta (Osaka, Japan) colorimeter and the classification by CIELAB (1986) system were used to analyze color. From L* a* b*, chroma (c*) and hue angle (h°) were determined. Color lightness (value), L* (100: white to 0: black), measures how light/dark is the color of the object; chroma or saturation, c* (0–60), measures how dull/vivid is the object color; hue angle, h° (0–360°), expresses the characteristic/dominant color (0 – red/purple; 90 – yellow; 180 – bluish/green). Color difference (ΔE) was determined by comparison to a while standard tile (L* = 97.46; a* = –0.02; b* = 1.72) and using equation (1):

A commercial maize (Maizena^®, Copan, Portugal) starch was used as a reference.

Measurements were performed 25 individual times for each sample.

2.6 Fourier transform infrared (FTIR) spectroscopy

The FTIR spectra of acorn starches from the Q. rotundifolia and Q. suber were obtained on a Mattson 7000 spectrophotometer. Starch sample was deeply mixed with dried potassium bromide powder (KBr) (1:100 in weight) and pressed in a die at 10,000 psi to yield a disc. The spectra recorded with a resolution of 4.0 cm⁻¹ and 64 scans were registered in the medium-infrared area, which extends from 4,000 to 400 cm⁻¹. The calibration was carried out using KBr as a blank.

2.7 X-ray diffraction

The crystalline structure of the starches was carried out in a Philips diffractometer (X’Pert MPD, Almelo, Netherlands), using CuKα radiation (λ = 0.154 nm) generated at 40 kV and 20 mA, at a moisture content between 8 and 10%. The sample was scanned through the 2θ (diffraction angle) from 3 to 50° at a speed of 8°/min. The degree of relative crystallinity was determined following the method previously reported by Huang et al. (2007) using equation (2):

where I _a is the amorphous area on the X-ray diffractogram and I _c is the crystallized area on the diffractogram.

2.8 Solid-state NMR

Q. rotundifolia and Q. suber starches were characterized by ¹³C NMR using cross polarization and magic angle spinning (CP/MAS), using a Bruker 500 spectrometer (Bruker, Germany). Before NMR measurements, samples were kept at 100% relative humidity for 48 h. The spectra were obtained with static magnetic field of the 11.7 T. The sample was placed in a zirconium rotor sealed of 4 mm with Kel-FTM caps and rotated at 9 kHz. The acquisition parameters used were as follows: proton pulse of 4 µs, contact time of 2 ms, delay of 4 s, and 7,000 scans.

2.9 Statistical analysis

Experimental data were subjected to a one-way analysis of variance (ANOVA) using the Statistic^® vs 6 software. The significant difference or separation of means comparison of all parameters was tested by Tukey’s HSD test. The level of significance used for all the statistical tests was 95%.

1. Ethical approval: The conducted research is not related to either human or animal use.

3.1 Morphology of acorn starches

Isolated acorn starches appeared to the naked eye as a powder having a light greyish-brown color. Starches isolated by the ENZ method presented higher values of L*, meaning that they were whitest (Table 1). Starches presented a yellow predominant color (h° values near 90°). QR produces starches duller than QS, and the A3S methods presented the higher values. Furthermore, some of these aspects are probably related to color characteristics of the origin flours since this tendency was also observed for QR and QS flours (Correia et al. 2009). The ΔE calculation clearly puts in evidence major differences between samples. According to Drlange (1994), acorns starches presented values that are classified as great (between 6.0 and 12.0) to very great color difference (more than 12). Based on these results, acorn starches showed not to be recommended for use in products requiring light uniform color, but they could find applications in baked goods (e.g., cookies, donuts) that are not affected by the starch colour. The isolation method seems to affect the colour parameters. Starches produced by the A3S method presented higher values of total color difference parameter and lower L* values, probably due to the lower efficiency to extract and eliminate some compounds formed during the drying process of the respective fruits, such as those resulting from the Maillard reaction.

Granule size distribution and SEM images of acorn starches produced by alkaline and enzymatic isolation methods are shown in Figure 1. Granules are round and oval in shape, showing some fractures. Images analysis does not establish differences on the isolation method effects. Acorn starches consist of medium/small granules, with starch granules size lower than 18 µm, representing about 80% of the total volume of particles, with about 25% of volume granules lower than 5 µm. The mean granule size ranged between 9 and 13 µm. These were also found for other acorn starches (Cho and Kim 2000; Stevenson et al. 2006; Molavi et al. 2018). Wilson et al. (2006) used a size scale where starch granules are classified into three size ranges: A-type granules (>15 µm), B-type granules (5–15 µm), and C-type granules (<5 µm). Based on this classification, acorn starch granules should be classified predominantly as B-type granules, and thus, some equipment used for extraction starch from wheat cam be used for acorns. However, QS starch showed a higher range of sizes, with some granules larger than 60.0 µm in diameter. The ENZ method had removed/destroyed larger granules when compared to granule size distribution showed by the A3S method. This effect was also reported by other authors in their studies (Daiuto et al. 2005). From the results, it is evident that the granule diameter of acorn starch varies greatly depending on species and starch isolation methods.

Figure 1

SEM (×1,500) and granule size distribution of acorn starch: QR – Q. rotundifolia; QS – Q. suber; A3S – alkaline pH and using successively three sieves method; ENZ – enzymatic method.

3.2 Fourier transform infrared spectroscopy

Fourier transform infrared (FTIR) analysis is a powerful tool to analyze, in a qualitative way, the interactions between starch molecules. It analyzes the presence of functional groups in the starch granules to confirm the structure of starches, which are detected through peaks, presented in a spectra, originated by vibrational modes of starch components. Usually, the spectra of sample show peaks that confirm the polysaccharide nature of the starches at approximately 3,320, 2,920, 1,645,1,426, 1,331, 1,156, 1,079, 1,020, 923, and 915–757 cm⁻¹ (Deng et al. 2020).

Figure 2 shows a comparison between the FTIR spectra obtained for the acorns starch granules isolated by A3S and ENZ methods. No differences can be observed in those spectra. The characteristic peaks of starch molecules interactions are presented in Table 2, and as an example, the spectra of the sample Q. suber isolated by the ENZ are shown in Figure 2. The peaks at 2,935 and 3,436 cm⁻¹ could be due to the complex vibration stretches associated with –CH stretches attributed to the ring methine hydrogen atoms and free intramolecular and intermolecular hydroxyl groups, respectively (Reddy et al. 2017; Xu et al. 2019). The absorptions near 1,376 and 1,648 cm⁻¹ may be attributed to water bending vibration (Zhang et al. 2007; Luo et al. 2009; Reddy et al. 2017). Between 933 and 1,160 cm⁻¹ appeared three characteristic peaks, which are associated with C–O bond stretching. Furthermore, the anhydro-glucose ring C–O stretch peaks are observed at 1,085 and 1,033 cm⁻¹ (Fang et al. 2002; Wu, Geng et al. 2009), and the peak at 862 cm⁻¹ is associated with the C–H of residual carbons of b-galactose (Freile-Pelegrín et al. 2007). Thus, the 800–1,200 cm⁻¹ region in these spectra reflects C–C, C–OH, and C–H stretching vibrations, starch polymer conformation, and hydration process (Li et al. 2020). Furthermore, the absorbance ratio of 1,045/1,022 cm⁻¹ show the ordered degree of starch, being the absorbance at 1,045 cm⁻¹ relative to the ordered/crystalline region of starch, and the one at 1,022/995 cm⁻¹ reflects the proportion of amorphous to the ordered carbohydrate structure in starch, indicating the internal changes in the double-helical structure (Li et al. 2020; Sevenou et al. 2002). These encountered results are closed to those obtained for native potato starch (Luo et al. 2009; Wu et al. 2009).

Figure 2

The FTIR spectra of acorn starch: QR – Q. rotundifolia; QS – Q. suber; A3S – alkaline pH and using successively three sieves method; ENZ – enzymatic method.

3.3 X-ray diffraction pattern and crystallinity

The X-ray diffraction patterns of acorn starches are shown in Figure 3, which also presents the corresponding X-ray diffraction parameters and crystallinity degree. Starch samples exhibit a X-ray pattern, where the polymorphs were present in varying proportions, whit a strongest diffraction peaks at 2θ of about 17° and 23° and a few small peaks at around 15°, 20°, and 26°. These patterns may be classified as a C-type X-ray pattern (a mixture of “A” and “B” unit cells) (Chung et al. 2010). B-type starch has one strongest diffraction peak at around 17° 2θ, a few small peaks at around 15°, 22, and 24°, and also a characteristic peak at about 5.6° 2θ, like potato starches and some legumes starches (Jayakody et al. 2007; Singh et al. 2008; Guo et al. 2017). Conversely, pure A-type starches such as wheat and maize exhibit a shoulder/doublet at around 17° and 18° 2θ, which do not show the 2θ peak at 5.6°, but show a unique and strong peak at around 15 and 23° 2θ, instead of the doublet 22–24° 2θ (Jayakody et al. 2007; He and Wei 2017), and another peak at 27° (Yu and Wang 2008). A- and B-types starches presented the same double helical conformation but different packing arrangement, being the A-type denser and with less intracrystalline water content (Wang et al. 2009). Furthermore, the structure of the C-type starch granules is distinct, being an aggregation of both A- and B-type polymorphs, giving to C-type starches different processing suitability (Guo et al. 2017), more complex than the other two types due to the different contents and distribution patters of A- and B-type crystals in them (Wang et al. 2009).

Figure 3

X-ray diffraction patterns of acorn starches: QR – Q. rotundifolia; QS – Q. suber; A3S – alkaline pH and using successively three sieves method; ENZ – enzymatic method; DC – crystallinity degree (%). Data with the same superscript letter are not significantly different (p < 0.05).

The C-type spectrum are classified into C_A-type, C_B-type, and C_C-type (Cai et al. 2014; He and Wei 2017) based on their similarity to A and B-type polymorps. According to He and Wei (2017), the X-ray pattern of Cc-type starch shows a single peak at about 17° and 23° 2θ, and a few small peaks at around 5.6° and 15° 2θ, and that of C_A- and C_B-type starches are similar to that of the C_C-type, but they presented a shoulder peak at around 18° 2θ and a strong singlet at 23° 2θ for C_A-type starch, and two shoulder peaks at about 22° and 24° 2θ for the C_B-type starch. In this sense, because of the peaks revealed by X-ray, the Q. rotundifolia and Q. suber acorn starches could be classified as a C_A-type. It should also be noted that the presence of a peak near 20° indicates the occurrence of crystalline amylose–lipid complexes (Yang et al. 2010). The complex amylose–lipid complexes are more evident on starches isolated by the A3S method, which means that this method is less affective in destroying these complexes. Furthermore, the type of X-ray diffraction patterns of acorn starches seems to be dependent of acorn species because Steverson et al. (2006) observed a A-type pattern for pin oak acorn starches. Kim and Lee (1976) reported acorn starch has B-type, but Molavi et al. (2018) found that Q. brantii a C-type pattern.

Concerning crystallinity, no significant differences are presented between acorn starches, showing a high percentage of this parameter, when compared, for example, with pin oak acorn starch (Stevenson et al. 2006) but similar to Persian starch (Molavi et al. 2018). QR starches showed higher crystallinity values, and the A3S method produced also starches with higher degree of crystallinity, which could mean that its original structure is less disturbed, mainly the amylopectin crystallites.

3.4 Solid-state NMR

Molecular order in a starch granule is composed of two types of helices from amylopectin side chains: (1) packed helices in regular arrays, which form crystallinity, measured by ¹³C CP/MAS solid-state NMR and X-ray diffraction, and (2) unpacked helices that are in a regular form or packed in short-range distance, which can be detected by ¹³C CP/MAS solid-state NMR but cannot be found by X-ray diffraction (Cooke and Gidley 1992).

Figure 4 summarizes the ¹³C CP/MAS NMR spectra of starch Q. rotundifolia and Q. suber, with reference to ¹³C chemical shift values for all resolved signals. Assignments of resonances are consistent with the literature data (Gidley and Bociek, 1985; Veregin et al. 1986; Bogracheva et al. 2001; Atichokudomchai et al. 2004; Wang et al. 2009). RMN spectra present a different crystallite types. Signal of C1 resonance contains information both on the noncrystalline (but rigid) chains as well as the crystalline nature. A triplet characteristic resonance for A-type crystallites was observed for all the starch (Figure 4). The multiplicity of the C1 resonance is related to the type of packing of granules in the region 99–104 ppm intensity. Maltotriose is the repeat unit in an A-type starch, and the twofold axis generates the double helix causing three different environments for C1; thus, the C1 peak in A-type starch spectra is a triplet (∼102, 101, and 100 ppm). Moreover, maltose is the repeat unit for B-type starch, and the threefold screw axis generates a double helix, which provides two different environments for C1. Thus, the C1 peak in B-type starch spectra is a doublet (∼101 and 100 ppm) (Gidley and Bociek 1985; Veregin et al. 1986). Signal at 81–84 ppm is attributed to C4. The peaks in the C2, C3, and C5 region (70–79 ppm) were sharpened, and the signal near 76 ppm (B-type double helices from residues of free amylose) was pronounced for the nonwaxy starches (Atichokudomchai et al. 2004).

Figure 4

¹³C CP/MAS NMR spectra of acorn starch: QR – Q. rotundifolia; QS – Q. suber; A3S – alkaline pH and using successively three sieves method; ENZ – enzymatic method.

The results show that Q. rotundifolia and Q. suber have similar starch structures. Moreover, the isolation methods clearly affected both varieties. Since C-type starches have both A-and B-type crystallites, it can be suggested that the resonances in the spectrum mainly depend on the relative proportions of A- or B-allomorph in the sample (Bogracheva et al. 2001). Signals at 99–104 and 58–65 are attributed to C1 and C6 in hexapyranoses, respectively. As described earlier, the C1 resonances for the ENZ method, in both acorns, indicates that B-type allomorph is actually predominant in the acorn starch isolated by this method, whereas a typical A-type characteristic can be found in the A3S method. The two broad shoulders that appear at ∼103 and ∼82 ppm resonances could arise from the amorphous domains for C1 and C4 (Wang et al. 2009). The overlapping signal around 72 and 74 ppm is associated with C2, C3, and C5, and the resonance around 61.4 ppm is assigned to C6. Except the above peaks, the weak peak appearing at 94.5 ppm could be attributed to high-energy, twisted conformations remote from those characteristic of single helices (102–103 ppm) and double helices (99–101 ppm) (Wang et al. 2009).

Generally, one particular difference was observed in the ¹³C CP/MAS NMR patterns for acorn starch isolated by the A3S and ENZ methods. The intensity of C-1 and C-4 amorphous resonances seems to be smaller in A3S method than in the ENZ method. Amorphous compounds give broad resonances as the distribution of local molecular environment gives rise to a wide distribution of chemical shifts for each carbon. Some researchers mentioned that ordered materials show narrower resonances due to more regularity of the environment (Gidley and Bociek 1985; Veregin et al. 1986), which reflected the stricter polymer configurations in the ordered parts of the starch (Paris et al. 1999). Because of the decrease in the amount of amorphous phases in the starch granules, the resonances decreased, which leads to obtain more crystalline material in the compounds obtained by the A3S method. These results are also corroborated by the X-ray diffraction crystallinity results.