3.1 Printing properties evaluation
The printing products obtained from all formula groups of ink are shown in Figure. 3. The banana paste ink without adding potato whole powder had strong flowability and little formability, so it was not included in the picture. As shown in the Fig. 3, as the amount of potato whole powder increased, the printing products’ support gradually enhanced. When the amount of potato whole powder reached 12.5%, the shape of the printing product could be well maintained. Moreover, during the printing process, the printing inks were extruded smoothly, and the surface of the printing products also appeared smooth and delicate. However, as the amount of potato whole powder further increased, the extrudability of the ink decreased, and the extruded material would experience line breakage during the deposition process, resulting in poor printing performance. Therefore, the ink of Ink-12.5% was the group with the best printing performance, and this group formula was selected for further research.
3.2 LF-NMR properties analysis
LF-NMR was applied to explore the water distribution in different formula printing inks, as shown in Fig. 4. In Fig. 4a, all peaks appearing on the spectrum were divided into three groups on the basis of the relaxation time of the abscissa: T21 (≤ 10 ms), T22 (10ms-100 ms), and T23 (≥ 100 ms). It is reported that the relaxation time reflected the degree of water freedom, and the smaller the relaxation time, the lower the level of water freedom (Chitrakar et al., 2019). The peak relaxation time of T21 was less than 10 ms, indicating the tightly bound solid water. The peak relaxation time of T22 was between 10–100 ms, indicating the semi-bound water. The peak relaxation time of T23 was greater than 1000 ms, indicating stronger fluidity of water (Chen et al., 2023). As the amount of potato whole powder continued to increase, the relaxation time of the free water, semi-bound water, and bound water all continued to decrease. This indicated that adding potato whole powder would combine the moisture in the material to form a denser network structure, leading to a reduction in the water activity and fluidity (Liu et al., 2020). The peak area reflects the moisture content of each peak. For the purpose of showing the moisture content of each group more intuitively, the proportion of moisture content in different groups of peaks is shown in Fig. 4b. From the Fig. 4b, it could be observed that as the amount of potato whole powder increased, the percentages of the three groups T21, T22, and T23 did not show significant changes. T23 was the highest proportion of water in the inks, indicating that free water dominated among all materials. This is the same as the LF-NMR properties of 3D printing inks in multiple published papers (Liu et al., 2018a; Liu et al., 2020; Cui et al., 2022). Thus, it could be seen that adding potato whole powder could combine the moisture in the materials and limit the fluidity of the inks, but it did not change the dominance of free water T23 in any inks.
3.3 Rheological properties analysis
There is a related link between the rheological properties and the printing performance of the printing materials, including the extrusion force during printing, and the stability after printing (Zhu et al., 2019). The rheological properties of each formula group are shown in Fig. 5. From the Fig. 5a, it could be discovered that as the increase of shear rate, all formula inks’ apparent viscosity (η) value decreased, indicating that all inks had the characteristic of shear thinning. The appropriate viscosity value contributes to the smooth extrusion of the printing inks from the printer nozzle, which is of great significance for 3D printing (Wang et al., 2023). Moreover, as the amount of potato whole powder continuously increased, the η value of the printing inks also gradually increased. This indicates a gradual decrease of the fluidity of water molecules, which is conform to the results of LF-NMR analysis.
The storage modulus represents the behavior of elastic solids and can the mechanical strength of materials; The loss modulus represents the viscous behavior and is related to the flow performance (Feng et al., 2022; Wang et al., 2023). From Fig. 5b and Fig. 5c, it could be found that the storage modulus (G') and loss modulus (G'') both increased as the frequency increased, exhibiting a frequency-dependent property. The G' and the G'' increased with the increase of the potato whole powder. Moreover, the value of the G' was always higher than the value of the G'', indicating that the behavior of inks tended towards elastic materials. This is a necessary condition for 3D printing (Montoya et al., 2021; Habuš et al., 2021). This material property is beneficial for maintaining the stability of the product shape after printing (Niu et al., 2023). As shown in Fig. 5d, tanδ (G''/G') was less than 1, which also indicated that the inks were like elastic materials and had solid properties. As potato whole powder continued to increase, tanδ continued to decrease, indicating that the mechanical strength of the materials was constantly increased, the fluidity was weakened, and the support and stability post-printing were continuously improved, which was consistent with previous analysis.
3.4 Browning analysis
Visual quality is crucial for the food industry, and color as one of the visual attributes is an important sensory quality attribute of products. The most intuitive manifestation of enzymatic browning is the deterioration of color, which can affect consumer acceptance (Wang et al., 2022a). The changes of L* value, a* value, and b* value are shown in Table 1. The L* value indicates brightness, the a* value indicates red, and the b* value indicates yellow (Kim et al., 2022). The decrease in L* and b* values, as well as the increase in a* value, represents that the color turns to darken, blue, and red, which is a manifestation of the browning reaction (Kim et al., 2021a). From Table 1, it could be found that the color after printing compared to that before printing, the L* value and the b* value both decreased, the a* value increased, which indicated that the color of the printing products was darker and redder than that of the inks. Moreover, as the oxygen content in the printing environment decreased, the degree of the changes of L* value, a* value, and b* value significantly reduced. In the 21% oxygen (Control) group, after printing, the L * value decreased by 20.4 ± 0.45, the a * value increased by 4.39 ± 0.10, and the b * value decreased by 5.88 ± 0.90; When the oxygen content was less than 1%, after printing, the L * value decreased by 5.77 ± 0.36, the a * value increased by 0.74 ± 0.20, and the b * value decreases by 1.52 ± 0.67. This indicated that the low oxygen environment reduced the degree of browning of the printing products.
Table 1
Changes of sample color parameters L*, a*, b* during 3D printing under different oxygen content conditions
Proportion of oxygen content | △L* | △a* | △b* |
21%(Control) | − 20.40 ± 0.45a | + 4.39 ± 0.10a | − 5.88 ± 0.90a |
15%±0.2% | − 15.44 ± 0.59b | + 4.60 ± 0.15a | − 4.54 ± 0.27ab |
10 ± 0.2% | − 12.21 ± 1.02c | + 2.92 ± 0.14b | − 4.12 ± 0.74bc |
5%±0.2% | − 10.14 ± 1.58d | + 1.77 ± 0.10c | − 2.80 ± 0.30cd |
༜1% | − 5.77 ± 0.36e | + 0.74 ± 0.20d | − 1.52 ± 0.67d |
Note: The "-" sign in the table indicates the reduced value, while the "+" sign indicates the increased value; different letters indicate significant differences in the data in the same column. |
From Fig. 6a, it could be seen that the browning index (BI) of the printing products was much higher than that of the inks before printing, indicating that the printing material underwent a browning reaction during the printing process. And as the oxygen content in the printing environment continued to decrease, the ΔBI after printing also continued to decrease, just as shown in Fig. 6b, when the oxygen content was 21% (Control), the ΔBI was 5.26 ± 0.50, and when the oxygen content was less than 1%, the ΔBI was only 0.58 ± 0.13. In order to more intuitively represent the color change, a flower model was established, and 3D printed in the control environment with an oxygen content was 21% and the best gas environment with an oxygen content less than 1%. The product photos were placed in Fig. 6b. Visually, the product printed in the control environment had a darker and redder appearance, resulting in a browning reaction. In the gas environment with an oxygen content of less than 1%, the color characteristics of the printing product were basically the same as the color of the printing ink.
However, the group with the lowest oxygen content did not completely inhibit the occurrence of browning reaction, because the oxygen in the printing environment was not completely removed. It should be noted that the BI of the inks before printing also reduced with the reduction of oxygen content in the printing environment because the inks were exposed to the printing gas environment during the preparation and filling process, the browning reaction also occurred in these processes. It can be seen that the printing inks are very easily oxidized and very sensitive to the oxygen content in the environment. In summary, compared to the control environment, reducing the oxygen content in the printing environment could prevent the occurrence of browning reactions, and as the oxygen content continued to decrease, preventing browning was more effective. This showed that the products printed in low-oxygen environments could maintain the original colors better, which had a high application value.
3.5 Total polyphenol and flavonoid analysis
Oxygen, phenolic compounds, and polyphenol oxidase are key factors in the enzymatic browning reaction of fruit and vegetable materials. Oxygen and phenolic compounds are substrates for enzymatic browning reactions. The changes in TPC and TFC during 3D printing with different oxygen content are shown in Fig. 7. As shown in Fig. 7a and Fig. 7c, TPC and TFC in the printing product were lower than those in the ink before printing that was, there was a loss of the TPC and TFC during the printing process. For example, when the oxygen content was 21% (control), the TPC of inks before printing was 13.10 ± 0.32 mg/g, and TFC was 4.12 ± 0.24mg/g. In the printing products after printing, the TPC and TFC reduced to 9.64 ± 0.16mg/g and 2.89 ± 0.03mg/g, respectively. Moreover, for the control environment, the TPC and TFC in the printing products were the lowest among all the test samples. This is because phenolic substances are the main substrates for enzymatic browning and are consumed during the browning process (Derardja et al., 2022). When the oxygen content was less than 1%, the TPC and TFC in the inks before printing were the highest among all the test samples, with values of 14.83 ± 0.46mg/g and 5.78 ± 0.15mg/g, respectively, and the loss during the printing process was small.
Figure 7b and Fig. 7d represent the loss rates of the TPC and the TFC during the 3D printing process. From Fig. 7b and Fig. 7d, it could be discovered that with the decrease of the oxygen content in the printing environment, the loss rate of TPC and TFC during the printing process also gradually decreased. In the printing environment with 21% oxygen content (control), the loss rate of total polyphenols and flavonoids during printing was as high as over 25%, while when the oxygen content is less than 1%, the loss rate was only about 1%. This indicates that reducing the oxygen content could reduce the loss degree of TPC and TFC during the printing process. This is because the presence of oxygen can promote enzymatic oxidation, so reducing the availability of oxygen is crucial for preventing the oxidation of bioactive substances (Kim et al., 2021b). It is noteworthy that when the oxygen content was below 1%, the TPC after printing might even be slightly higher than before printing. This might be because the printing inks were extruded through the printer nozzle, the shear made the materials more delicate and uniform, so that the phenolic substances were easier to be extracted.
3.6 Antioxidant capacity analysis
The DPPH reagent and the ABTS reagent are commonly applied to explore the radical scavenging activity of phenolic compounds (Xie et al., 2017). In this study, the radical scavenging activities of DPPH and ABTS were selected as indicators to characterize the antioxidant capacity of the samples. From Fig. 8a and Fig. 8c, it could be seen that when the oxygen content was less than 1%, for the ink samples before printing and the products after printing, the DPPH radical scavenging rate could both reach over 80%, and the ABTS radical scavenging rate could both approach 100%. However, at the oxygen content of 21% (control), the DPPH radical scavenging rate of the printing product was only 65.34% ± 2.06%, and the ABTS radical scavenging rate was only 77.51% ± 1.44%. Moreover, the radical scavenging activity of ABTS was higher than that of DPPH. The difference in the radical scavenging rate values expressed in percentage between DPPH and ABTS might be because of the different antioxidant mechanisms. The antioxidant mechanism of DPPH is electron transfer, while ABTS is hydrogen atom transfer (Qiu et al., 2023).
Browning reaction occurred during the printing process, and also a loss of TPC, TFC, and antioxidant activity. Derardja et al. (2022) analyzed the enzymatic browning of apricots and found that polyphenols and antioxidant capacity in apricots were significantly affected by the browning reaction. Enzymatic browning significantly reduced total phenolic compounds and antioxidant capacity. This was consistent with our research conclusion. From Fig. 8b and Fig. 8d, it could be observed that as the oxygen content in the printing environment decreased, the loss rate of antioxidant capacity gradually decreased. When the oxygen content was 21% (control), the loss rates of DPPH and ABTS radical scavenging capacity were both over 15%, while when the oxygen content was less than 1%, the loss rates of DPPH and ABTS radical scavenging capacity were both less than 0.5%. The changes were the same as the changes in the total polyphenols in the above section. Just as the research results about persimmon showed by Chang et al. (2019), the positive correlation between phenol content and antioxidant activity was significant. Therefore, owing to the antioxidant capacity being closely related to the total phenol content, the decrease of the total phenol content was the cause of the decrease of antioxidant capacity probably (Alipoorfard et al., 2020).
3.7 Oxidase activity analysis
PPO and POD are the enzymes that participated in browning reactions (Cho et al., 2016). The enzyme activity changes of PPO and POD were shown in Fig. 9. From Fig. 9a and Fig. 9c, in any printing group, the oxidases' activity in the printing products was lower than that in the inks before printing. Moreover, among all the test samples, the PPO activity and POD activity were the highest in the printing inks at the oxygen content of 21% (control), with 4.49 ± 0.08 △OD420/min·g and 1.69 ± 0.05 △OD470/min·g, respectively. From Fig. 9b and Fig. 9d, it could be seen that when the oxygen content in the printing environment was in the range of 21% (Control) − 5% ± 0.2%, with the continuous reduction of oxygen content, the loss rate of PPO activity decreased during the printing process, reducing from 19.20% ± 3.24–9.41% ± 1.81%. The enzymatic reaction catalyzed by polyphenol oxidase requires oxygen to start, and low oxygen level inhibits the activity of polyphenol oxidase (Duan et al., 2009). Therefore, in the printing process, the activity of PPO was inhibited, and the consumption of PPO was reduced. This might be the reason why the loss rate of PPO activity decreased with the decrease of oxygen content in the printing environment. Kim et al. (2018) ground apples in air atmosphere and nitrogen atmosphere, and then stored them in the corresponding gas environment for six hours after grinding. It was found that the PPO activity of samples in the air atmosphere was 8.71 ± 3.32 U/g·min, and the PPO activity of samples in the nitrogen atmosphere was 16.66 ± 3.13 U/g·min, the decrease of PPO activity of the sample in nitrogen atmosphere was lower than that in air atmosphere, which was the same as our results. In terms of POD activity, shown in Fig. 9c and Fig. 9d, under all test printing gas environments, the loss rate of POD during the printing process gradually decreased with the oxygen content in the printing environment decreasing.
It is worth noting that when the oxygen content was less than 1%, the activity of PPO in the materials before and after printing decreased significantly, the loss rate in the printing was as high as 28.69% ± 4.17%, which indicated maybe that the inhibition mechanism of the gas environment on PPO has changed, not only the dilution effect of enriched nitrogen on the oxygen concentration but also the interaction of the active site or the induced change of protein conformation (O’Beirne et al., 2011). Soliva et al. (2000) found that a nitrogen atmosphere could inhibit the activity of PPO and thereby reduce the browning effect of PPO. Zhang et al. (2001) introduced nitrogen into the PPO system and the mixture system of PPO and substrate continuously to create a nitrogen atmosphere and discovered that nitrogen treatment could inhibit the activity of PPO in both systems. Besides reducing the content of available oxygen, it might also inactivate the active site of the enzyme. The loss of PPO activity might be the reason for the significant increase of total phenol content.