Piercing injury and pulling-out mechanical properties of garlic scapes based on X-ray CT

1 Introduction

China is the largest garlic producer in the world, with an annual cultivation area of approximately 800,000 ha and an annual output accounting for about 70% of the global total (Zhang et al., 2026; Jiang et al., 2025; Shao et al., 2025). Garlic production in China is developing towards mechanization and suitable for mechanization, and the level of mechanization is still low (Zhang et al., 2023; Li et al., 2024; Sun et al., 2018; Tian et al., 2018; Jat et al., 2023; Zhang et al., 2025). Garlic scapes as the byproduct of garlic cultivation are one of the most widely grown vegetable varieties. Currently, garlic scape harvesting is still primarily manual. Low harvesting efficiency, high labor intensity, and high labor costs have become major factors limiting the large-scale production of garlic scapes.

In recent years, some scholars have conducted research on garlic scape harvesters and proposed an operating method in which a front-mounted cutting tool first processes the pseudostem of the garlic plant to locally damage the fibrous layer before pulling-out the scapes. However, the aforementioned research is still in its early stages. During harvesting, there is a high rate of injury to garlic scapes and a low harvest yield, making it difficult to meet the requirements for efficient, low-waste harvesting in actual production, which hinders its widespread adoption (Feng et al., 2021; Geng et al., 2021). Therefore, investigating the mechanical propertiess and internal tissue injury mechanisms induced by the piercing of garlic plants with front-end tools and the subsequent pulling-out of garlic scapes during harvesting is of great significance for the design and optimization of garlic scape harvesting equipment (Yang J. et al., 2026; Yang X. et al., 2026).

Currently, the primary methods for manually harvesting garlic scapes include the piercing method, the pressing method, and the method of dissecting the stalk to harvest the scapes. Among these, the piercing method is widely used due to its high harvesting efficiency and minimal injury to the plants. Consequently, most existing garlic scape harvesters are designed with corresponding piercing and harvesting mechanisms based on this method. When operating the machine, its parameters must be properly configured to prevent excessive injury to the garlic scapes and garlic plants. Therefore, investigating the mechanical properties of garlic scapes and garlic plants during harvesting is fundamental to the design of garlic scape harvesters. Mechanical parameters such as the compressive failure load, bending failure load, maximum shear force during the cutting of garlic scapes, and the tensile force and fracture displacement of injury garlic scapes during the pulling-out processes (Du and Wang, 2016). These parameters provide a basis for determining the gripping force, pulling-out force, gripping position, and piercing position of the harvesting device, ensuring that stem breakage occurs as close as possible to the predetermined piercing position, thereby reducing harvest injury and yield loss. Previous studies did not establish a relationship between the mechanical properties of garlic scapes and the internal injury caused by the tools during pulling-out, leaving the injury mechanisms of the piercing tools unclear. Therefore, in this study, CT scanning was used to characterize the injury features of garlic scapes inside garlic plants after pulling-out. These observations were combined with the results of mechanical tests to analyze the relationship between garlic scape fracture and the action mechanisms of the piercing tools, thereby providing a design basis for the development and optimization of garlic scape harvesting devices.

In this study, garlic scapes at the harvesting stage were selected as the research object. Tensile, bending, compression, and shear tests were conducted on garlic scapes, followed by pulling-out tests after piercing treatment using different tools. Furthermore, X-ray CT scanning was employed to non-destructively observe the internal injury of garlic plants. The mechanical properties of the whole garlic scape and its different sections were analyzed. In addition, the mechanical responses and internal tissue injury characteristics of garlic plants during the pulling-out process after treatment with different piercing tools were investigated, and the effectiveness of different piercing tools was evaluated. This study provides a theoretical basis for the design parameters of garlic scape harvester and the research foundations for improving the harvest rate of garlic sprouts.

2 Experiments and methods

2.1 Experiment materials

The garlic plants used in the experiment were obtained from a garlic scape cultivation base in Yichuan County, Luoyang City, Henan Province. The samples used in the pulling-out tests were obtained from seven garlic scape planting areas in the same farmland, which based on consultations with local garlic farmers, combined with local garlic cultivation practices and garlic scape harvesting experience. These samples covered, to some extent, different planting regions and plant growth conditions, providing a certain degree of regional representativeness. The moisture content of garlic plants may change during storage, and the samples used for pulling-out tests must be obtained from the same batch of garlic plants on the same day. The variety is Da Qingke, which matures in late spring and is harvested in April. Prior to the experiment, garlic plants were dug up from the same plot of land using tools. During the digging process, soil was left attached to the roots of the plants. The garlic plants, with soil still on their roots, were then placed in buckets, covered with water to submerge the roots, and stored in a cool, well-ventilated area.

2.2 Determination of basic physical parameters

The basic physical parameters of garlic plants and scapes serve as a crucial basis for determining the design parameters of garlic scape harvesters. Parameters such as plant height, scape length, diameter, pseudostem height, pseudostem diameter, and leaf distribution directly influence the dimensional design and installation positioning of the gripping device, support device, scape-piercing mechanism, and pulling-out mechanism. Specifically, plant height is used to determine the installation height of the support and guidance mechanisms as well as the clearance required for obstacle avoidance; scape diameter is used to determine the gripping gap and force of the gripping mechanism; scape length is used to determine the stroke of the pulling-out mechanism; pseudostem height and leaf distribution are used to determine the installation height of the cutting mechanism and the piercing tool’s operating position; and pseudostem diameter provides the basis for designing the piercing tool’s penetration depth. Therefore, measuring the basic physical parameters of garlic plants provides foundational data for the design and parameter determination of garlic scape harvesters.

2.2.1 Measurement of dimensions

Prior to the experiment, basic morphological measurements were taken of the garlic plants in this batch, as shown in Figure 1. A tape measure was used to measure the distance from the ground to the top of the scape cap when extended (total plant height), the distance from the ground to the bottom of the scape cap (scape height), the height at the point where the scape emerges from the plant (pseudostem height), as well as the heights of the third and fourth branches and the heights at two evenly spaced positions between them (labeled a, b, c, and d from the top to the base); Use a vernier caliper to measure the diameters at the upper, middle, and lower evenly spaced positions on the garlic scape on the pseudostem and the pseudostem, respectively. The measurement results are shown in Tables 1, 2.

TABLE 1. Heights of garlic plants.

TABLE 2. Diameters of garlic plants.

2.2.2 Measurement of moisture content and density

To measure the moisture content of garlic plants, each plant was divided into three equal segments. From each segment, a 25-mm-long sample was cut 25 mm above and below the midpoint, as shown in Figure 2a. The samples were then placed together in a halogen moisture analyzer (accuracy: 0.005 g, SN-SH-10A, Shanghai Shangyi Instrument Equipment Co., Ltd., Shanghai, China) to measure the moisture content, as shown in Figure 2b. The experiment was repeated five times, and the results were averaged. The final measured moisture content of the garlic plants was 89.18%.

To measure the density of garlic scapes emerging from the plants, the garlic scapes at the base of the pseudostem, and the leaf sheaths of the pseudostem, respectively. The garlic plant was divided into upper and lower parts at the point where the scape emerged. A 25 mm-long sample was cut from each part, centered on the midpoint of each segment. The segments were numbered a and b from top to bottom, as shown in Figure 3a. Use a knife to cut open the leaf sheath of the pseudostem segment b and take out the scape, as shown in Figure 3b, where the circular mark indicates the scape and the fan-shaped mark indicates the multi-layered leaf sheaths enveloping the scape.

The mass of the cut segments was measured using an electronic balance (accuracy: 0.001 g, FA-LQC2003, Suzhou Zhengfeng Electronic Technology Co., Ltd., Suzhou, Jiangsu Province, China), as shown in Figure 4a, and the volume of the samples was measured using the water displacement method with a graduated cylinder (accuracy: 0.2 mL), as shown in Figure 4b.

The formula for calculating density is as follows:

ρ = m v (1)

In the Equation 1: ρ is the density of the segment, in g/cm³; m is the mass of the segment, in g; v is the drained volume of the segment, in mL. Each experiment was repeated five times, and the results were averaged. The densities of the various segments are shown in Table 3.

TABLE 3. Density measurement results.

2.3 Mechanical properties of garlic scapes experiment

The mechanical properties of garlic scapes were determined during an experiment using a universal testing machine (JHY-5000M, Xiamen Jinhuayuan Technology Co., Ltd., Xiamen, Fujian). In preliminary experiments, the tensile strength of garlic scapes was measured to be 30–80 N, the compressive strength 80–150 N, and the bending and shear strengths 25–30 N. Therefore, a 50 N range load cell was used for the bending and shear experiments on garlic scapes, while a 200 N range load cell was used for the tensile and compressive experiments.

2.3.1 Garlic scape tensile experiment

To determine the optimal clamping position for garlic scapes on the harvesting device and breaking at the intended breaking point during harvesting, segmental tensile experiments were conducted to investigate the tensile properties of different segments of the garlic scapes. Prior to the experiment, a single plant was randomly selected from the batch of garlic plants to serve as a test group. The portion of the plant below the bulb was removed, and the remaining parts were separated from the pseudostem and leaf sheaths and divided into three equal segments (numbered a, b, and c from top to bottom), as shown in Figures 5a,b.

Because garlic scapes have a smooth surface, to prevent specimen slippage, compression damage, and stress concentration within the double-sided clamping fixture during the experiment, foam tape was applied to the inner surfaces of the fixture, and both ends of the specimen were wrapped twice with medical paper tape. This treatment increased the friction between the specimen and the fixture while cushioning the clamping load, as shown in Figures 5c, 6. Since the adhesive layer on the surface of the foam tape wears down during experiments, reducing mechanical, a new piece of foam tape is applied after 5 to 6 experiments to prevent the garlic scapes from shifting. Paper medical tape, foam tape, and fixtures of identical specifications were used in all tests, and all experimental procedures were performed by the same operator.

During an experiment, clamp both ends of the prepared specimen using double-sided locking fixtures and perform tensile tests on specimens a, b, and c at a speed of 10 mm/min, as shown in Figures 6a,b.

To improve the efficiency of garlic scape harvesting, experiments were conducted on whole garlic scapes at different speeds to investigate their tensile mechanical properties under varying speeds. Before the experiment, remove the entire garlic scape from the garlic plant and remove the scape cap. As shown in Figures 7a,b, the same treatment method used for the tensile experiment on segmented garlic scapes was applied to both ends of the specimen, which was then clamped at both ends using double-sided locking clamps. To avoid significant deviations in the data during the experiment, the machine’s tensile speed should not be too high; the loading speeds should be 5 mm/min, 10 mm/min, 15 mm/min, and 20 mm/min (Shen et al., 2024; Wang et al., 2025; Nazari Galedar et al., 2009).

All of the above experiment termination methods are set to force sudden change mode.

During the experiment, a test was considered invalid if the specimen cracked or slipped in the clamping fixture, fractured near the fixture opening. Each test was repeated seven times.

2.3.2 Garlic scape compression experiment

During scape harvesting, the gripping device applies a certain amount of force to the garlic scapes. If the gripping force is too weak, the scapes will slip, reducing the harvest yield; if the gripping force is too strong, it can injury the scapes, compromising their harvest quality. Therefore, when pulling-out the scapes, the device needs to grip the garlic scapes at the appropriate point with right force to improve the effectiveness of the process. For this reason, it is important to conduct compression experiments on different parts of garlic scapes to investigate the radial load-bearing capacity of these segments (Kovács and Kerényi, 2019).

Before the experiment, a garlic plant was randomly selected to serve as an experiment group. The scape cap and the portion of the plant below the bulb were removed, and the remaining parts were taken out of the pseudostem sheaths. The garlic scapes used in the experiment were 450–600 mm in length. A 20 mm sample was cut from the top of each scape and set aside. Another 20 mm sample was then taken 80 mm below the first sample on the same garlic scape. This procedure was repeated to obtain a total of five segments, which were numbered a, b, c, d, and e in sequence, as shown in Figures 8a,b.

During the experiment, place the sample horizontally in the center of the lower indenter of the universal testing machine, as shown in Figure 7c. The indenter used in the experiment was a steel circular compression fixture with a diameter of 100 mm. Start the universal testing machine and perform radial compression at a loading speed of 30 mm/min based on the preliminary experiment results. Set the compression stroke to 4 mm to ensure that the specimen fractures. Each set of experiments is repeated five times.

2.3.3 Garlic scape bending experiment

During the mechanized harvesting of garlic scapes, the gripping, guiding, supporting, and piercing processes all generate lateral loads, causing the scapes to bend and deform. If the bending load exceeds the garlic scapes’ own load-bearing capacity, it can easily cause them to break at unintended points, resulting in broken or shortened scapes (Weng et al., 2024). Therefore, a three-point bending experiment was conducted on garlic scapes to determine the bending loads that different segments of the entire scape can withstand.

Before the experiment, a garlic plant was randomly selected. First, the scape cap and the portion below the bulb were removed, and the remaining part was peeled out of the pseudostem sheath. An 80 mm segment was measured from the top and cut off for later use.

Another 80-mm segment was then taken from the same scape, 20 mm apart from the first. This procedure was repeated to obtain a total of five segments, which were numbered a, b, c, d, and e in sequence. Place the cut sample horizontally on the support. Since the sample diameter is 6.5–8 mm, the span of the support brackets is set to four times the diameter, or 30 mm, as shown in Figure 9b. Next, start the universal testing machine. Based on the preliminary experiment results, lower the indenter at a loading speed of 60 mm/min, set the downward stroke to 4 mm to ensure the specimen fails, and repeat the experiment five times for each group.

2.3.4 Garlic scape shear experiment

During the mechanized harvesting of garlic scapes, the piercing device applies a shearing force to specific segments of the scapes, causing localized injury at predetermined points to facilitate subsequent pull-out. To determine the optimal piercing position for the garlic scape piercing device and the radial shear loads that different segments of the garlic scape can withstand, shear experiments were conducted on various parts of the garlic scape.

Before the experiment, a garlic plant was randomly selected. First, the scape cap and the portion below the bulb were removed, and the remaining part was peeled out of the pseudostem sheath. An 80 mm segment was measured from the top and cut off for later use. Another 80 mm segment was then taken from the same scape, 20 mm apart from the first. This procedure was repeated to obtain a total of five segments, which were numbered a, b, c, d, and e in sequence, as shown in Figure 10c. A double-shear experiment was conducted using a custom-made shearing tool, as shown in Figures 10a,b. The sample was placed horizontally on the support brackets, with a span of 50 mm between the brackets, as shown in Figure 10d. In the preliminary experiments, sample a and b were difficult to fracture under low-speed shearing; therefore, the shearing tool was lowered at a loading speed of 60 mm/min with a downward stroke of 4 mm to ensure fracture of the samples. Each set of experiments was repeated five times.

2.4 Mechanical experiments of pulling-out and CT characterization

The mechanical properties of different sections of garlic scapes under axial loading, radial compression, lateral bending, and local shear conditions were first characterized through tensile, compression, bending, and shear tests. These mechanical properties provided a basis for the clamping and piercing methods used in the pulling-out tests. Based on these mechanical properties and the agronomic requirements of garlic scape harvesting, different piercing tools were applied to intact garlic plants, followed by pulling-out tests to evaluate the effects of piercing position and piercing tool type on the pulling-out force, fracture displacement, and pulling-out success rate. Therefore, X-ray CT scanning is used to characterize the internal structure of the piercing area and explain the reasons for the differences in pulling-out performance caused by different piercing tools from the perspective of injury internal tissues.

2.4.1 Mechanical experiment of pulling-out

Since garlic scapes are naturally enclosed by the garlic pseudostem and leaf sheaths, mechanical harvesting is not simply a matter of pulling-out and breaking the scapes, but rather a complex mechanical process involving both the trauma caused by piercing and the subsequent breaking of the scapes. An experiment on a single garlic scape can only reflect the mechanical properties of the scape itself; it is difficult to accurately characterize the actual stress conditions during the pulling-out process. Therefore, it is necessary to conduct an experiment on garlic plants to study the force required to break the garlic scape and the displacement of the garlic scape during the pulling-out process.

The tools for piercing the garlic scapes, as shown in Figure 11, are typically used during the harvest. Garlic farmer usually use them piercing the garlic scapes at the pseudostem of the garlic plants.

The mountain-shaped tool is shown in Figure 11a. The front cutting elements typically consist of three steel pins embedded in a plastic block; they are evenly spaced horizontally at the front end of the plastic block, with the middle pin being shorter than the two on either side; the cone-shaped tool, as shown in Figure 11b, features a steel cone at the front end; the fork-shaped tool, as shown in Figure 11c, features a front end consisting of an auxiliary piercing plate and piercing plates, with the piercing plates evenly distributed on both sides of the auxiliary piercing plate. The geometric parameters of the three garlic scape piercing tools were as follows: the mountain-shaped tool had a needle length of 8 mm, a needle-base diameter of 1.5 mm, and a spacing of 3.75 mm between adjacent needles; the cone-shaped tool had a cone length of 28 mm and a needle-base diameter of 4.5 mm; and the fork-shaped tool had a thickness of 1.5 mm, a bottom width of 3 mm and a right-angle side length of 23 mm for the piercing blade, while the auxiliary piercing blade had a length and width of 65 and 4.5 mm, respectively.

The determination of the optimal position for piercing garlic scapes relies largely on the experience of farmers, and there is no clear theoretical framework or scientific basis for this practice. Therefore, this study conducted experiments of pulling-out garlic scapes, which using different piercing tools, drawing on agronomic experience. Before the experiment, a garlic plant was randomly selected as a test sample, and the soil and roots beneath the bulb were removed. Place the plant horizontally and rest the scape on the lower support to simulate the condition where, during scape piercing, the bulb of the garlic plant remains in the soil while the farmer holds the scape. Using the three piercing tools described above (as shown in Figure 10), perform the piercing procedure on different parts of the garlic plant (the third and fourth sets of leaves counting down from the top, as well as two evenly spaced points between them, labeled a, b, c, and d, respectively), as shown in Figure 12a. Mechanical experiments on the garlic plants were then conducted.

To simulate the growth state of garlic plants during pull-out, the garlic were secured using a double-sided clamping fixture located at the bottom; wrap medical paper tape around the base of the curved garlic scape (as shown in Figure 7d), and apply foam tape to the inner walls of the double-sided clamping fixture located at the top. Tighten the clamp to secure the garlic scapes where they are wrapped with medical tape, preventing them from shifting within the clamp during the experiment, as shown in Figure 12b. The experiment termination condition is a sudden change in force (the experiment ends when the garlic scape breaks). To prevent the garlic scapes from breaking above the piercing point due to excessive speed during the tensile experiment, the loading speed was set to 10 mm/min, with each experiment group repeated seven times.

2.4.2 Internal characterization of CT scanning in the pulling-out experiment

Since the garlic scapes are enclosed by the garlic pseudostem and leaf sheaths, it is difficult to accurately determine, based solely on the plant’s appearance after piercing, whether the scapes have broken internally or whether the break occurred at the intended piercing point, and whether the pseudostem tissue has been excessively injured. To determine the actual effectiveness of the scape-piercing tool in damaging the internal garlic scapes, the internal structure of garlic plants after piercing was examined using a CT scanner (NanoVoxel-3000, Sanying Precision Instruments Co., Ltd., Tianjin, China) for CT scanning and subsequent image reconstruction to characterize the cross-segment of garlic scapes (Zwanenburg et al., 2022; Koetzier et al., 2023; Ghosh et al., 2023; Piovesan et al., 2021; Karahara et al., 2023; Guo et al., 2026).

Prior to the experiment, garlic plants that had successfully pull out (with the scape breaking off at the piercing point) were stored in sealed bags to prevent moisture loss from the pseudostem segment resulting from the bolting treatment during transport, which could affect the scanning results, and were then promptly transferred to the CT scanning equipment. Using the wound on the stalk as the center point, measure 40 mm above and below it to select the sample segments, cut them out, and perform CT scans on them. The scan parameters are shown in Table 4.

TABLE 4. X-ray CT scan parameters.

3 Experiment results

3.1 Mechanical properties of garlic scapes experiment

3.1.1 Garlic scape tensile experiment

The force of segmented tensile experiment on garlic scapes is shown in Figure 13. As displacement increases, the force follows a pattern of rising rapidly at first, then increasing slowly, and finally dropping sharply after reaching its peak. During the initial loading phase, the tensile force and displacement exhibit an approximately linear relationship, indicating that the garlic scapes are primarily in the elastic deformation stage; as the displacement continued to increase, the rate of increase in tensile strength gradually decreased, and the garlic scapes exhibited a distinct necking phenomenon, indicating that they were gradually entering the elastic-plastic deformation stage; toward the end of the loading phase, the tensile force rapidly decreases after reaching its maximum value, indicating that the sample has fractured. It can be seen that garlic scapes exhibit certain viscoelastic material properties in the tensile experiment. The force of different segments of the garlic scape, from top to bottom, show essentially the same trend, and the patterns of tensile failure are quite similar.

As shown in Figure 14, there are differences in the maximum tensile strength among the samples from different segments. As the sampling location moves from the upper to the lower part of the garlic scape, the maximum tensile force of the samples generally decreases. Among them, the maximum tensile force of the sample from segment a was the highest, significantly exceeding that of segments b and c, indicating that the upper tissue of the garlic scape has a greater tensile load-bearing capacity; the difference in maximum tensile force between samples b and c is relatively small, indicating that the tensile force of the middle and lower segments of the garlic scape is quite similar. It can thus be concluded that the tensile strength of garlic scape is unevenly distributed along their length, with a general pattern of higher tensile capacity in the upper segment and lower tensile capacity in the lower segment.

The condition of garlic scapes throughout the tensile testing process is shown in Figure 15. After clamping, the garlic scape remained in a relaxed state, as shown in Figure 15a. During the tensile test, the garlic scape was subjected to tensile loading, resulting in an increase in axial length and a reduction in radial diameter, accompanied by a necking phenomenon, as shown in Figure 15b. After the tensile test, the garlic scape fractured at approximately one-third of the ungripped section, as shown in Figure 15c.

Figure 16 shows the force of whole garlic scape under different loading velocities. The trend is identical to that of the segmented tensile experiment curve: in both cases, the tensile force rises rapidly with increasing displacement during the initial stage of the experiment, then the rate of increase gradually slows down, and after reaching a peak, it drops sharply, indicating that the sample has fractured at this time. As the loading velocity increased from 5 mm/min to 20 mm/min, both the maximum tensile force and the displacement at break of the garlic scape showed an increasing trend, indicating that the loading velocity has a significant effect on the tensile mechanical properties of the entire garlic scape. Within the experiment speed range, garlic scapes can withstand greater tensile loads at higher loading velocities and fracture at greater fracture displacements, demonstrating a certain degree of velocity sensitivity, as shown in Figure 17.

The fracture time of whole garlic scapes under different loading speeds is shown in Figure 18. As the loading speed increased, the fracture time of the garlic scapes gradually decreased, indicating that increasing the loading speed could shorten the scape breaking time and improve the scape pulling-out efficiency. Meanwhile, the difference in fracture time between adjacent loading speeds gradually decreased, suggesting that when the loading speed increased to a certain level, the effect of further increasing the speed on reducing the fracture time became weaker. Combined with the result in Figure 17 that a higher loading speed required a greater tensile force, the selection of the scape pulling-out speed should comprehensively consider the pulling-out efficiency and low injury harvesting requirements. Therefore, under the premise that the tensile force output of the harvesting device and low injury operation requirements are satisfied, the loading speed can be appropriately increased to obtain higher garlic scape pulling-out efficiency.

3.1.2 Garlic scape compression experiment

The compressive force of the garlic scape compression tests is shown in Figure 19. The force dropped sharply to a certain level after reaching the first peak. Subsequently, as the indenter continued to move downward, the force gradually increased and then dropped sharply again after reaching the second peak, indicating that the garlic scape segment underwent staged failure during compression. When the load reached the first peak, the garlic scape segment experienced initial compressive fracture, and an obvious radial crack appeared on its axial segment, dividing the specimen into two parts, as shown in Figure 20a. As the indenter continued to move downward, the two fractured parts were further compressed, and the local tissues were further injured, causing the curve to rise again and resulting in secondary fracture after the second peak was reached, as shown in Figure 20b.

As shown in Figure 19, sample a did not exhibit the same curve characteristics as the fractured samples b, c, d, and e, namely the process of reaching two force peaks followed by sharp decreases. Therefore, sample a did not undergo obvious fracture during the test, as shown in Figure 20c, indicating that sample a had strong compressive deformation resistance. The maximum failure loads that samples b, c, d, and e could withstand generally decreased with the decrease in sampling height. Among them, the maximum failure load of sample b was significantly higher than that of the samples below it, indicating that this position had strong compressive resistance. The maximum failure load of sample c was at a medium level, while the maximum failure loads of samples d and e were relatively close and both at a low level. These results indicate that the compressive resistance of different parts of the garlic scape varied significantly. Except for the uppermost segment, namely specimen a, the overall compressive resistance gradually weakened as the sampling height decreased, as shown in Figure 21.

The positions of sample a and sample b showed stronger compressive resistance and were therefore more suitable as clamping positions. Among the other positions, sample c was easily obstructed by leaves, while samples d and e were located in the region wrapped by the pseudostem, which was not conducive to mechanical clamping. Therefore, considering both compressive resistance and plant morphological characteristics, the upper-middle region of the garlic scape, corresponding to samples a and b, was more suitable as the clamping position for the harvesting device.

3.1.3 Garlic scape bending experiment

The bending force–time curves obtained from the three-point bending tests of different parts of garlic scapes are shown in Figure 22. The bending force–time curves first dropped sharply to a certain level after reaching the initial peak and then increased slightly. For sample a, the bending force did not drop sharply after reaching the peak but decreased slowly, indicating that sample a did not reach the bending failure load under the action of the indenter. As the indenter continued to move downward, sample a gradually slipped from the lower support. When the sample slid into the support, it tended to fold and came into contact with the indenter, thereby hindering the further downward movement of the indenter and causing the bending force to increase again, as shown in Figure 23a.

The bending force of samples b, c, d, and e dropped sharply after reaching the peak and then increased slightly, indicating that after bending fracture occurred, part of the epidermal or fibrous tissues at the fracture position remained connected, and the residual tissues continued to bear the load, as shown in Figure 23b. The maximum bending failure loads of samples from different positions were different, with sample b showing the highest value. Except for sample a, which did not undergo typical fracture, the bending forces of samples b, c, d, and e decreased successively, indicating that the bending resistance of garlic scapes weakened from the top to the base, as shown in Figure 24. Comprehensive analysis shows that the middle-lower parts of the garlic scape, namely samples c and d, were more likely to undergo bending fracture under transverse loading, which can provide an important basis for selecting the fixing position for scape piercing and the predetermined fracture position.

3.1.4 Garlic scape shear experiment

The shear force of different parts of the garlic scape is shown in Figure 25. For samples a–d, the shear force–time curves reached the peak after multiple fluctuations and then dropped sharply, indicating that shear fracture occurred in the samples.

For sample e, the test stopped before shear fracture occurred, and only bending deformation appeared at the tool contact position without being cut through, as shown in Figure 26. This was because the shear failure load of sample e exceeded the measuring range of the tension-compression sensor (50 N). To avoid sensors overload, the universal testing machine automatically stopped loading. This result indicates that sample e was more difficult to shear-fracture than the other parts and had stronger shear resistance. Except for sample e, the shear resistance of the other segments gradually decreased from the top to the bottom. Samples a, b, c, and d all showed the characteristic variation of reaching the peak after multiple fluctuations during shearing, followed by a sharp drop. This indicates that under the action of the tool, the samples were not completely fractured at one time. Instead, as the tool continued to move downward, the upper epidermis, pith, and lower epidermis, pith, and lower epidermis were successively injured, resulting in a layered shear fracture phenomenon.

Based on the experimental data, the shear stress of the garlic scape was calculated using the following Equation 2:where τ is the shear stress, MPa; P Ma x is the maximum shear force, N; and A is the cross-segmental area at one end of the sample, mm².

τ = P Max 2 A (2)

The shear stress of each sample segment was calculated based on the experimental data, and the average values are listed in Table 5.

TABLE 5. Result of mean shear stress.

Since sample e did not undergo shear fracture before the end of the test, only the shear stresses of samples a, b, c, and d were calculated. According to Figure 27, the results showed that the shear stress of the garlic scape decreased markedly from the top to the base, indicating that the middle-lower parts, namely samples c and d, were more easily injured by the cutting tool (Gao et al., 2024; Aydın and Arslan, 2018; Wang et al., 2024; Qin et al., 2026). Sample e was located at the base of the garlic scape, where more vascular bundles and a higher degree of lignification were present. As a result, the cutting edge of the tool could not effectively cut into the tissue, and bending deformation at the contact point, rather than shear fracture, mainly occurred during the test, as shown in Figure 26. Therefore, considering both the shear failure effect and the feasibility of scape piercing, the middle-lower parts of the garlic scape, namely samples c and d, were more suitable as the piercing positions.

3.2 Mechanical experiment of the scape pulling-out

To characterize the effects of different tools and different piercing positions on the garlic scape piercing performance, the garlic scape pulling-out success rate was introduced. Combined with the force–displacement curves, the optimal garlic scape harvest method was investigated. The garlic scape pulling-out success rate was calculated by Equation 3 as follows:where η is the scape pulling-out success rate, representing the probability that the garlic scape undergoes tensile fracture at the piercing position after the tool is used to pierce different positions of the garlic plant pseudostem; A is the number of samples in which tensile fracture occurs at the piercing position; and Ω is the total number of garlic plant samples treated by tool piercing.

η = A Ω × 100 % (3)

3.2.1 Mechanical experiment on scape piercing with the mountain-shaped tool

After the mountain-shaped tool was used to pierce garlic plants at different height positions, namely a, b, c, and d from the top to the bottom, the force–displacement curves during the scape pulling-out process are shown in Figure 28. After piercing treatment at different heights, the variation trends of force during scape pulling-out were basically consistent with those in the garlic scape tensile experiment. Specifically, the tensile force gradually increased with increasing displacement and then dropped rapidly after reaching the peak. This indicates that the mountain-shaped tool may have caused local injury to the garlic scape after piercing treatment, but did not completely cut off the internal garlic scape wrapped by the pseudostem and leaf sheath. Since part of the fibrous tissue around the piercing wound remained connected, the treated garlic scape still retained a certain tensile resistance.

According to the scape pulling-out experiment results at different piercing positions, position b showed the highest maximum tensile force, approximately 43 N, followed by position d at approximately 38 N, position a at approximately 35 N, and position c with the lowest value of approximately 29 N.

In terms of fracture displacement, position d had the largest value, close to 90 mm, followed by position a at approximately 82 mm, position b at approximately 61 mm, and position c with the smallest value of approximately 30 mm. These results indicate that the tensile load-bearing capacity and fracture deformation capacity of garlic scapes differed significantly among different piercing positions. Under the same tensile speed, the tensile force before fracture of the pierced garlic plants, ranging from 28 to 43 N, was lower than that of the whole garlic scapes before tensile fracture, ranging from 43 to 50 N. This was because after the mountain-shaped tool was used to pierce the garlic plants, part of the fibrous tissue of the garlic scape was fractured, while the unfractured fibrous tissue still remained connected. Therefore, the pierced garlic scape still retained a certain tensile resistance.

The success rates at positions a and b were the lowest, both at 42.8%; the success rate at position c was 57.1%; and the success rate at position d was the highest, reaching 71.4%. Overall, as the piercing position moved downward, the scape pulling-out success rate showed an increasing trend, indicating that after piercing treatment with the mountain-shaped tool at lower positions, the garlic scape was more likely to fracture in the subsequent scape pulling-out experiment.

In the garlic scape pulling-out experiment with piercing at position d, the sample showed a relatively large fracture displacement, indicating that it still had a certain deformation capacity before fracture. However, the scape pulling-out success rate at this position was the highest, suggesting that effective injury and stress concentration were more easily formed after the action of the mountain-shaped tool at position d, thereby increasing the probability of garlic scape fracture at the piercing position. Position b showed the highest maximum tensile force but a relatively low scape pulling-out success rate, indicating that when the garlic scape itself had strong tensile resistance, it was still difficult to fracture at the piercing position even after piercing treatment. Position c showed the lowest maximum tensile force and fracture displacement, but its success rate was only 57.1%, indicating that a lower tensile force alone did not completely determine the garlic scape pulling-out success rate, and the piercing position and injury effect were also important. Considering the fracture position and scape pulling-out success rate, the middle-lower region where position d was located, between the third and fourth leaves of the plant, was more suitable as the piercing operation position for the mountain-shaped tool.

3.2.2 Mechanical experiment on scape piercing with the cone-shaped tool

After the cone-shaped tool was used to pierce garlic plants at different height positions, namely a, b, c, and d from the top to the bottom, the force–displacement curves during the scape pulling-out process are shown in Figure 29. After piercing treatment at different positions, the variation trends of the mechanical curves during the garlic scape pulling-out process of garlic plants were similar to those of the tensile process of whole garlic scapes.

According to the garlic scape pulling-out experiment results at different piercing positions, position c showed the highest maximum tensile force, approximately 28 N, followed by position a at approximately 23 N, position d at approximately 22 N, and position b with the lowest value of approximately 18 N. In terms of fracture displacement, position a had the largest value, approximately 33 mm, followed by position c at approximately 25 mm, position b at approximately 16 mm, and position c with the smallest value of approximately 15 mm. These results indicate that the tensile load-bearing capacity and fracture deformation capacity of garlic scapes differed among different piercing positions.

Compared with the mountain-shaped tool, the overall fracture displacement after piercing treatment with the cone-shaped tool was smaller, all within 33 mm. This indicates that after cone-shaped piercing, the overall tensile resistance of the garlic scape at the wound position decreased significantly, and fracture occurred more rapidly.

As the piercing position moved downward, the garlic scape pulling-out success rate increased. Position d showed the highest success rate, reaching 85.7%; positions a and c followed; and position b showed the lowest success rate, at 42.8%. Compared with the mountain-shaped piercing tool, the cone-shaped tool showed garlic scape pulling-out success rates no lower than those of the mountain-shaped tool at all four positions. In particular, at position d, the scape pulling-out success rate reached 85.7%, indicating that the cone-shaped tool caused more significant injury to the garlic scape and was more conducive to garlic scape pulling-out.

Position highest d showed the garlic scape pulling-out success rate, with the shortest fracture displacement and a relatively small tensile force. This indicates that after the cone-shaped tool was used to pierce garlic plants at position d, the vascular bundles and fibrous tissues were severely injured, and the stress concentration effect was enhanced, making the garlic scape more likely to fail during tensile loading. The scape pulling-out success rates at positions a and c were 57.1% and 71.4%, respectively. Among them, position a showed the longest fracture displacement, which may be because the tissue continuity of the garlic scape above the pierced position was better, giving it stronger deformation capacity. Position b showed the lowest tensile force and a relatively short fracture displacement, but its scape pulling-out success rate was the lowest, at 42.8%. This may be because the garlic scape tissue above the pierced position still maintained good continuity. In addition, as the height decreased, the diameter of the garlic scape gradually increased, and the ratio of the injured area caused by the tool to the cross-segmental area at that position became lower.

3.2.3 Mechanical experiment on scape piercing with the fork-shaped tool

After the fork-shaped tool was used to pierce garlic plants at different height positions, namely a, b, c, and d from the top to the bottom, the force–displacement curves during the scape pulling-out process are shown in Figure 30. After piercing treatment at different positions, the variation trends of the mechanical curves during the scape pulling-out process of garlic plants were similar to those of the tensile process of whole garlic scapes. Among them, position d showed the highest tensile force, while the tensile forces at the two upper positions, a and b, were at relatively low levels. Position d had the largest fracture displacement, approximately 58 mm; the fracture displacements at positions a, b, and c were all between 20 and 30 mm.

As the piercing position moved downward, the scape pulling-out success rate gradually decreased. Positions a and b showed the highest success rates, both at 71.4%; position c followed; and position d showed the lowest success rate, at 42.8%.

Position d had the lowest scape pulling-out success rate, whereas both the tensile force and fracture displacement were the highest. This may be because, after treatment with the fork-shaped tool, the damage to the garlic scape tissue at the fourth leaf position of the garlic plant was relatively limited, and a large amount of fibrous tissue remained connected. Alternatively, the wrapping effect of the basal pseudo-stem on the garlic scape may have been enhanced, resulting in stronger tensile resistance. At position c, located above the fourth leaf of the garlic plant, the scape extraction success rate was also relatively low, while the maximum tensile force and fracture displacement were lower than those at position d. During the piercing process of the fork-shaped tool, the sheet-like blade may have caused damage to the cross-section of the internal garlic scape at the piercing position, effectively disrupting the tissue continuity of the internal garlic scape.

The trend in the scape pulling-out success rate of the fork-shaped tool with the downward shift of the piercing position was opposite to those of the mountain-shaped tool and the cone-shaped tool, as shown in Figure 31. Overall, the cone-shaped tool showed the highest scape extraction success rate, whereas the mountain-shaped tool showed the lowest scape pulling-out success rate.

Therefore, the different variation trends in the success rates of the three scape piercing methods may have resulted from the combined effects of piercing height, the tissue characteristics of the garlic scape at the piercing position, the wrapping degree of the pseudostem, and the different front-end morphologies of the tools.

4 CT scanning characterization

To further clarify the effects of different piercing tools on the scape pulling-out mechanics of garlic plants after piercing treatment, CT scanning and subsequent image reconstruction were performed on the internal structures of the pierced garlic plants, and the cross-segments of the garlic scapes were characterized.

4.1 Scape piercing with the mountain-shaped tool

The front end of the mountain-shaped piercing tool consisted of three steel needles. In the internal fracture image (Figure 32), only a triangular tearing wound caused by tensile loading after piercing was observed on the upper side of the fracture surface, as shown in Figure 32a. This indicates that the spacing between the steel needles on both sides of the tool was larger than the diameter of the garlic scape at the piercing position, and therefore these two needles did not cause injury to the garlic scape. In Figure 32a, the cross-segment corresponding to the red marked line is shown in Figure 32b, the cross-segment corresponding to the yellow marked line is shown in Figure 32c, and the cross-segment corresponding to the blue marked line is shown in Figure 32d.

As shown in Figure 32b, the fracture layer was elliptical, and the pseudostem and leaf sheath presented a dense and multilayer surrounding structure. It can also be observed from Figures 32c,d that the residual garlic scape tissues on the two fracture segments were unevenly distributed. In Figure 32c, the garlic scape tissue on the upper right side was missing, indicating that this part of the tissue was subjected to greater stress and fractured first, followed by the fracture of the remaining residual tissue.

4.2 Scape piercing with the cone-shaped tool

The internal fracture characteristics of the garlic pseudostem after treatment with the cone-shaped piercing tool were significantly different from those of the mountain-shaped tool, as shown in Figure 33a. In Figure 33a, the cross-segment corresponding to the red marked line is shown in Figure 33b, the cross-segment corresponding to the yellow marked line is shown in Figure 33c, and the cross-segment corresponding to the blue marked line is shown in Figure 33d.

At the middle positions on both the upper and lower sides of the fracture, the wounds were triangular in shape, and cracks extending upward and downward from the wound tip were observed. In addition, the garlic scape tissues on both the left and right sides of the wound were inclined outward in a branch-like shape, as shown in Figure 33a. This was because, in the later stage of tensile loading, the garlic scape on both the upper and lower sides of the piercing position was subjected to tensile force, followed by necking, at which point the deformation of the garlic scape became irreversible. While the garlic scape was undergoing tensile deformation, the internal wound at the piercing position also continuously extended toward both the upper and lower sides. When the failure load was reached, the garlic scape tissues on both sides of the wound fractured at the piercing position and could no longer recover their original morphology. After the garlic scape tissues on both sides of the wound lost the binding force provided by the fibrous connections, they still retained the necked shape and therefore inclined outward.

In the fracture layer image, a distinct longitudinal crack passed through the entire cross-segment of the garlic scape, and the crack exhibited an irregular serrated shape, as shown in Figure 33b. The garlic scape tissues on both sides of the piercing wound did not fracture on the same plane, resulting in a height difference; therefore, the crack indicates that the fracture surface was uneven. During tensile loading, stress was concentrated in the connected garlic scape tissues on both sides of the wound. The side with the smaller cross-segmental proportion was subjected to greater stress and was therefore more likely to fracture first, after which the stress on the other side increased to the failure load and caused fracture, as shown in Figures 33c,d.

The cone-shaped tool injured the garlic scape tissue and formed a cone-shaped wound. Different from the fracture tendency after treatment with the mountain-shaped tool, in which the piercing wound expanded from the tip toward the surrounding tissues, fracture under the cone-shaped tool treatment occurred directly on both sides of the penetrating wound. This also corresponded to the experimental results that the garlic scape showed a higher overall fracture rate and a shorter fracture displacement after piercing treatment with the cone-shaped tool.

4.3 Scape piercing with the fork-shaped tool

After piercing with the fork-shaped tool, the crack appeared at the middle position of the injured cross-segment, as shown in Figure 34b. This was attributed to the auxiliary positioning piece arranged at the middle of the front end of the tool, which enabled the blades on both sides to pierce accurately. In Figure 34a, the cross-segment corresponding to the red marked line is shown in Figure 34b, the cross-segment corresponding to the yellow marked line is shown in Figure 34c, and the cross-segment corresponding to the blue marked line is shown in Figure 34d.

The internal fracture image after treatment with this tool is shown in Figure 34a, where only slender cracks were observed on the upper and lower sides of the fracture. The tool pierced into the pseudostem from the left side and formed a triangular wound on the pseudostem, as shown in Figure 34b, which corresponded to the shape characteristics of the piercing blade at the front end of the tool. This fracture process was similar to that after piercing treatment with the cone-shaped tool. The difference was that the areas on both sides of the injured garlic scape were relatively uniform, as shown in Figures 34c,d. Before fracture, the stresses borne by the garlic scape tissues on both sides of the wound were not greatly different, and the connected parts on both sides fractured almost at the same time. However, since the blades on both sides of the fork-shaped tool were sheet-shaped, the wound area caused by the tool was smaller than that caused by the cone-shaped blade of the cone-shaped tool, resulting in a lower fracture rate than that under the cone-shaped tool treatment.

5 Discussion

5.1 Implications of garlic scape mechanical properties for harvesting machinery design

The mechanical properties of different segments along the same garlic scape vary from the top to the bottom part, which is mainly related to the differences in tissue composition among different parts. The upper part of the garlic scape exhibits higher tissue continuity and therefore possesses better resistance to tensile, compressive, bending, and shear loads. With decreasing height, the tissue structure gradually changes, and the fibrous tissues undergo increased lignification, resulting in reduced mechanical properties in the middle and lower regions.

In the tensile tests, the maximum tensile force and fracture displacement of garlic scapes varied with sampling position and also showed a certain sensitivity to loading speed. Within the selected loading velocity, increasing the loading velocity reduced the fracture time but allowed the garlic scapes to withstand higher tensile loads, which is associated with the viscoelastic behavior of agricultural biological materials. These findings indicate that the pulling-out speed of garlic scape harvesting machinery is an important factor affecting harvesting efficiency.

Compression, bending, and shear tests were conducted to investigate the mechanical characteristics of different garlic scape segments under various radial loading conditions. The compression test results demonstrated that the upper part of the garlic scape exhibited stronger radial load-bearing capacity, providing a basis for determining the clamping position and clamping force of the harvesting mechanism. The bending and shear test results indicated that the middle and lower parts of the garlic scape were more susceptible to mechanical injury, which was consistent with the empirical knowledge obtained from manual piercing-based harvesting practices. Therefore, the mechanical properties of garlic scapes under different loading conditions provides a theoretical foundation for the design of clamping, piercing, and pulling-out mechanisms in garlic scape harvesting machinery.

5.2 Relationship between pulling-out mechanics, CT characterization, and piercing tool

Among the three piercing tools, the cone-shaped tool achieved the highest pulling-out success rate, whereas the fork-shaped tool exhibited the lowest pulling-out success rate.

The steel conical blade at the front end of the cone-shaped tool and the blade composed of multiple steel needles at the front end of the mountain-shaped tool had similar sharp morphological characteristics. During piercing, these tool structures were more likely to penetrate the pseudostem and leaf sheath at the piercing point and act directly on the internal tissues of the garlic scape. Meanwhile, when the sharp part of the tool entered the garlic scape tissue, it compressed the surrounding tissue and formed a point-like injury, resulting in stress concentration and disruption of tissue continuity. During the subsequent pulling-out tests of intact garlic plants, such localized injury reduced the failure load required for fracture at the piercing position. Therefore, the mountain-shaped tool and the cone-shaped tool exhibited similar distributions of pulling-out success rates at the four piercing positions. However, the volume of each individual steel needle at the front end of the mountain-shaped tool was smaller than that of the steel cone at the front end of the cone-shaped tool.

The front blade of the fork-shaped tool was a sharp, plate-like triangular blade, which differed substantially from the other two tools in both structure and mode of action. During piercing, the inclined edges of the sharp triangular blade continuously compressed and displaced the garlic scape tissues on both sides. Owing to the plate-like structure of the piercing blade, it was difficult to generate pronounced local stress concentration in the pierced cross-section of the garlic scape. The pulling-out performance deteriorated as the piercing height decreased. According to the mechanical tests of garlic scapes, resistance to various forms of radial loading gradually decreased from the upper to the lower regions. Under the same experimental conditions, the garlic scape tissues at lower positions would therefore be expected to sustain more severe injury after piercing. However, after piercing with the fork-shaped tool, fracture became more difficult at lower positions. This indicates that the pulling-out performance after piercing with the fork-shaped tool was affected by the larger garlic scape diameter at lower positions, the stronger confinement provided by the pseudostem, the lower proportion of the injured area within the internal cross-section of the garlic scape, and the distinct morphology of the fork-shaped tool compared with the other two tools. These factors ultimately resulted in the unique distribution of its pulling-out success rate with piercing height.

To further explain the mechanical characteristics observed in the pulling-out tests, the cross-sectional images obtained by CT scanning were analyzed and discussed. The X-ray CT cross-sectional images revealed that the three piercing tools had different action mechanisms, thereby resulting in different pulling-out performances. After piercing and pulling-out with the mountain-shaped tool and the cone-shaped tool, the CT cross-sectional images showed that the cone-shaped tool produced a larger and more irregular injured region than the mountain-shaped tool. This suggests that the cone-shaped tool may have caused a larger injury area in the garlic scape tissue before pulling-out and disrupted tissue continuity at the piercing point to a greater extent, thereby resulting in lower pulling-out forces and fracture displacements at the lower piercing positions. Therefore, the cone-shaped tool exhibited higher pulling-out success rates than the mountain-shaped tool at all four piercing positions. After piercing and pulling-out with the fork-shaped tool, the CT cross-sectional images showed a relatively flat fracture surface, which was similar to that obtained in the tensile tests of garlic scapes. Moreover, specimens treated with the fork-shaped tool exhibited greater pulling-out forces and pulling-out displacements than those treated with the other tools. This may indicate that piercing with the fork-shaped tool produced a smaller injury area and had a weaker ability to disrupt the continuity of the garlic scape tissues. Therefore, the different performance trends among the piercing tools may have resulted from the combined effects of piercing height, the tissue characteristics of the injured region, the degree of pseudostem confinement, and the geometric characteristics of the tool tips.

6 Conclusion

• The mechanical properties of garlic scape vary along their height. The upper parts exhibited stronger resistance to tensile, compressive, bending, and shear loads, whereas the middle and lower parts can withstand relatively smaller loads. These differences in mechanical properties provide a theoretical basis for the design of clamping, piercing, and pulling-out mechanisms in garlic scape harvesting machinery.
• Piercing position and tool geometry jointly influence the pulling-out mechanical behavior of garlic scape. Among the tools, the cone-shaped piercing tool achieved the highest pulling-out success rate on the lower part of piercing region, indicating that an appropriate tool structure combined with a suitable piercing position can improve the harvesting efficiency of garlic scape harvesting machinery.
• X-ray CT analysis shows that different piercing tools produce different internal injury characteristics in garlic scapes, among which the level of injury is the main factor affecting the pulling-out mechanical property during the pulling-out process.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Wei Song, University of Illinois at Urbana-Champaign, United States

Reviewed by: Changsu Xu, Southwest University, China

Xin Pei, Sichuan University, China