Introduction

One of the most severe hazards to civil aviation safety is non-containment accidents that involve aero-engines. When high-speed rotating rotor components (such as fan blades and turbine discs) suddenly fail due to fatigue fracture, bird strike, or manufacturing defects, the broken fragments are ejected at supersonic speeds under the action of enormous centrifugal force. This can result in accidents that can range from minor flight function loss to catastrophic crashes1. Containment is a universal requirement for turbine engines worldwide. This means that a single blade must be contained after fracturing at a section outside the root2,3,4,5,6,7.

The blade separation speed is the primary challenge in whole-engine containment experiments. Precise control is required. The blade must fracture at a specific position and speed, and it must not generate an inordinate amount of residual speeds or debris spray8,9. The blade can be effectively separated within the designated timeframe and energy control can be achieved through the use of blasting technology. The effectiveness and reliability of this technology in the application of explosive separation of aero-engine blades were verified by Guo Mingming et al.10. through static blasting cutting experiments. The combination of pre-set fractures and shaped charge cutting blasting separation technology was employed. Lü Dengzhou11 verified the practicality of linear-shaped charge cutters in practical applications through numerical simulation and successfully conducted a constant-speed explosive separation test on titanium alloy blades. The shaped charge discharge and the rotating centrifugal load, which were controlled with high precision, were combined to cause the blades to break, according to the study. Hu et al.12 drilled holes in the center of the side of a titanium alloy composite plate and filled them with explosives. The optimal ratio of the minimal wall thickness of the explosive separation component to the charge diameter was determined through the use of experimental and numerical simulation methods. Hu et al.13 subsequently conducted research on the explosive separation of thinner plates by pre-fabricating guide perforations and guide grooves on the plate surface. The structural parameters of the shaped charge cutting device were optimized by Xia Zhiyuan et al.14 through the use of numerical simulation methods. Chen Gang15 investigated a method for effectively separating titanium alloy plates at a specific slot depth by studying the linear shaped charge jet explosive separation of titanium alloys.

Nevertheless, the high-speed cameras are unable to capture the separation and movement of titanium alloy blades in a timely and effective manner due to the significant amount of titanium fire that is generated during explosive separation tests using linear shaped charges. In the current research on explosive detachment experiments, investigations have primarily focused on cutting methods and performance metrics. However, there has been a significant gap in the exploration of suppressing the generation of titanium sparks during the explosive moment. By conducting a thorough analysis of the formation mechanism of titanium sparks in explosive detachment, this study introduces a novel concept-the barrier layer protection technology, which is proposed for the first time. Utilizing the velocity disparity between the metal flow from the linear shaped charge and the diffusion speed of the detonation products, the titanium fire is eradicated. High-temperature-resistant materials were employed to construct a barrier layer, and verification experiments were conducted. A new engineering solution for titanium fire suppression in aero-engine containment tests was demonstrated by static cutting test data, which demonstrated that this technology can effectively reduce the degree of combustion reaction. Additionally, the high-speed photography system ensured the observability of the blade separation process.

Investigation into titanium fire mechanism during explosive blade release

The field of view of high-speed cameras can be obscured by dazzling titanium fires that are the result of explosive cutting of titanium alloy blades in aero-engine containment tests, as illustrated in Fig. 1. Research has demonstrated that metals such as steel and aluminum, as well as their alloys, do not spontaneously combust due to the self-suppression mechanism of metal oxidation. Self-quenching is the term used to describe the phenomenon in which the oxide film that forms during melting obstructs the reaction between oxygen and the metal substrate. This results in a significant amount of heat loss, which exceeds the heat generated by the oxidation process, thereby making it difficult for the temperature to continue to increase. Consequently, these metals are not combustible.

Fig. 1
figure 1

Bright titanium fire produced by shaped energy cutting titanium alloy blades.

Full size image

The ignition point of titanium in air is 1627℃. A dense oxide film of TiO₂ is typically formed on the surface of titanium at normal temperatures, which serves a protective function. Ti₂O₃ and Ti₃O₅ are produced when titanium is subjected to elevated temperatures. The internal metal is exposed to oxidizing agents as a result of the original TiO₂ oxide layer’s easy cracking and loss of its sealing function. This results in a significant increase in temperature as heat production exceeds heat loss in a brief period of time. Titanium is capable of undergoing exothermic reactions at elevated temperatures due to its high oxygen reactivity, which is triggered by high-speed, high-pressure gas fluxes. Ultimately, this results in the construction of a bright titanium fire and combustion. Typically, the emergence of titanium fire in aero-engines necessitates specific temperature and pressure conditions, with the primary causes of titanium fire being violent impact and friction.

Figure 2 illustrates the cross-sectional dimensions and actual appearance of the linear shaped charge employed in the titanium alloy explosive separation experiment. The model number is 1145. Hexanitrotoluene (HNS) is incorporated into the linear shaped charge, while the container is composed of a lead alloy. The shell and the charge mold are simultaneously drawn and formed, and the charge angle is 75°. The shell and charge mold of the linear shaped charge generated by this process are typically constructed from the same metal material. The ball is pressed and subsequently compressed into a V-shaped groove after the explosion’s density is enhanced through multiple drawing processes.

Fig. 2
figure 2

Cross-section dimension and physical drawing of Linear shaped charge. (a) Linear shaped charge size diagram, (b) Cross section of linear shaped charge.

Full size image

The linear shaped charge detonated in a vacuum environment during the experiment, resulting in the immediate generation of high-temperature, high-pressure explosive gases and metal projectiles. The hexanitrotoluene explosive gas attained a temperature of 3700℃. The titanium alloy reacts with the explosion gas products of the explosive, including H2O, CO2, CO, N2, and C, at high temperatures, resulting in the production of Ti2O3, Ti3O5, TiN, TiC, and other products. This process results in a brilliant titanium flame. The surface oxide protective layer peels off at high temperatures.

The intensity of the explosive’s local destructive effect is closely correlated with the temperature of the detonation. While simultaneously reducing the detonation temperature to suppress the titanium fire, it is contradictory to maintain the explosive’s local destructive effect. The primary cause of the dazzling titanium fire was determined to be the high-temperature explosive gas products through experiments and computational studies that were conducted to address this challenging issue. The material is initially cut by a shaped charge metal jet generated by the cutting cord, which is subsequently followed by the advent of high-temperature explosive gas products. The metal jet’s speed is approximately 2700 m/s, while the explosive gas’s expansion speed is approximately 1500 m/s, making the metal jet’s speed substantially greater than that of the explosive gas, this information is based on calculations. The solution for the suppression of titanium fire is provided by utilizing the time difference between the two.

A barrier layer can be formed by a specific material that adheres to the titanium alloy’s surface. The titanium fire can be eliminated by reducing the temperature rise of the titanium alloy, blocking and delaying the contact between the detonation products and the titanium alloy during the jet penetration cutting process. In order to prevent the excessive consumption of jet kinetic energy that occurs when the material’s thickness is high, it is imperative that the thickness should be as small as feasible. The micro jet cutter initially penetrates the material and subsequently penetrates the titanium alloy plate during shaped charge cutting, resulting in a micro cut that exclusively permits the jet to pass through. The coating prevents detonation products from contacting the titanium alloy before they expand to the barrier layer. The barrier layer effectively reduces the generation of titanium fire by causing the temperature of the detonation products to drop significantly within a short period and temporarily isolating the titanium from the detonation products, preventing them from immediately reacting with the high-temperature detonation products.

Numerical simulation

The finite difference analysis program AUTODYN software is employed to achieve a comprehensive understanding of the obstruction effect of the obstruction layer in the metal discharge. The Euler algorithm is more suitable for solving the large deformation problem that arises during the explosive cutting process, as the material will endure significant deformations16.

Model parameters

The TC4 titanium alloy plate in the computational model is 10 mm by 5.6 mm. The titanium alloy plate features a central groove measuring 4.2 mm in width and 3.6 mm in depth. The barrier layer material in the simulation is designated as ceramic (Al2O3) with a thickness of 0.1 mm, closely affixed to the inner surface of the titanium alloy groove to prevent contact between the linear shaped charge and the titanium alloy plate. The characteristics of the Johnson-Cook material model were applied to the TC4 titanium alloy plate, shaped charge shell, and titanium fire suppression material. The parameters are specified in Tables 1, 2 and 317. The JWL equation of state was employed to characterize the detonation products of the explosive18,19, with parameters specified in Table 4.

To examine the obstructive influence of the barrier layer on the detonation products, two monitoring sites were established in the constructed model, positioned on either side of the barrier layer, to observe the velocity variations of the shaped charge jet. Figure 3 illustrates the numerical simulation and monitoring point diagram, with light blue denoting the titanium alloy plate, beige indicating the ceramic, dark blue representing the lead outer shell of the cutting cord, and green signifying the cutting cord core (HNS).

Fig. 3
figure 3

Linear shaped charge cutting titanium alloy model.

Full size image
Table 1 The Johnson-Cook material model parameters for TC4 (Ti6Al4V).
Full size table
Table 2 Material parameters of lead-antimony alloy.
Full size table
Table 3 The material model parameters of the Al2O3.
Full size table
Table 4 Parameters of the JWL equation of state for HNS.
Full size table

Calculation results and analysis

Figures 4 illustrate that subsequent to the detonation of the linear shaped charge, the jet commences its penetration of the ceramic barrier layer at t = 0.8 µs. At this juncture, the jet arrives at monitoring point 1, with the jet head velocity around 2703 m/s. At t = 1 µs, monitoring point 2 indicates that the jet’s velocity post-penetration is 2,388 m/s, while the jet head pressure exceeds 69 GPa, thereby guaranteeing the subsequent cutting efficacy of the titanium alloy. At t = 3 µs, the precision jet cutting tool infiltrates the titanium alloy, resulting in a minimal incision of merely 0.2 mm due to the rupture of the copper barrier layer. At this juncture, only a fraction of the detonation products could directly engage with the titanium alloy plate via the narrow aperture, while the remainder of the detonation products were precluded from direct contact with the plate due to the intervening barrier layer. Upon the completion of jet penetration, residual ceramic material remained affixed to the groove of the titanium alloy plate.

Fig. 4
figure 4

Solving results at different times in shaped-energy cutting titanium alloy.

Full size image

This method clearly illustrates the function of the barrier layer in postponing the interaction between detonation products and titanium alloy, while maintaining the cutting efficiency of the jet cutting tool. The barrier layer mitigates the thermal impact of high-temperature detonation products on the titanium alloy by limiting their initial interaction, hence significantly decreasing oxidation reactions. The simulation confirmed the viability of the barrier layer technique in mitigating titanium combustion during shaped charge cutting.

Materials and methods

Materials and instruments

Experimental materials and instruments: TC4 titanium alloy plate (100 mm×60 mm×23 mm), energy gathering cutting cable, detonator, high-speed camera, silica sol, aluminum nitrate solution, etc.

Preparation of barrier layer materials

Research revealed that aluminosilicate is lightweight and possesses excellent heat resistance and flame retardant qualities20. Aluminosilicates are a class of inorganic, non-metallic materials composed of aluminum oxides, silicon oxides, and alkali or alkaline earth metal oxides. Their melting point exceeds 1850 °C, and they undergo no crystalline transformation or decomposition at high temperatures. They maintain a stable volume and structure, avoiding material cracking or pulverization caused by phase transitions. Aluminosilicates are lightweight with an extremely low thermal expansion coefficient. During rapid temperature changes, minimal thermal stress develops within the material, making it resistant to cracking. They also possess high mechanical strength and inherent toughness, enabling them to withstand thermal stress-induced damage. Even near their maximum operating temperatures, aluminosilicate materials maintain high mechanical strength without exhibiting creep under self-weight or external loads.

It does not interact with detonation byproducts or titanium during the cutting procedure. Consequently, aluminosilicate was employed as the barrier layer in this study. The preparation procedure entails the gradual incorporation of silica sol into an aluminum nitrate solution, while sustaining the pH at 3.5–4.0.5.0 with a buffer solution. The solution is agitated for one hour to achieve homogeneity. The mixed sol is placed in an open container, heated in a water bath to sustain a temperature of around 40℃, and let to stand for 24 h to gel. The resultant wet gel is submerged in a 1:1 ethanol/water combination, aged at roughly 25℃ for 48 h to fortify the network structure, thereafter dried, and sintered to produce a high-temperature-resistant fireproof material suited for experimental applications.

Experimental design

The processed aluminosilicate material is combined with water to create a paste, which is subsequently uniformly applied to the inner surface of the grooves on the TC4 titanium alloy plate. Subsequent titanium fire suppression experiments are conducted after the complete evaporation and drying of the moisture in the coating. This treatment approach seeks to maximize the flame-retardant qualities of aluminosilicate to efficiently mitigate titanium fires. The shaped charge cutting is positioned near the bottom of the groove and extended 5 cm beyond the groove’s length. Figure 5 displays the assembled titanium alloy plate. The control experiment is designated as #1, whereas the aluminosilicate substance is designated as #2. Table 5 presents a total of two sets of tests.

Fig. 5
figure 5

Assembly of titanium alloy plate and barrier layer material before blast. (a) Non-barrier layer design experimental setup diagram, (b) Barrier layer design experimental setup diagram.

Full size image
Table 5 Number and thickness parameters of different materials.
Full size table

Results and discussion

Titanium flame test results

Figures 6 and 7 illustrate the phenomena seen at various time intervals during the explosive cutting of titanium alloy utilizing a linear shaped charge, both with and without an obstruction layer. The high-speed camera employed for filming operated at a frame rate of 5000. Figure 6 illustrates that in the control group experiment, the detonator activated and exploded the cutting cord 0.6 ms following the commencement of recording. At 1 ms, the linear shaped charge of the cutting cord detonated, emitting a substantial amount of light, subsequently generating a metal jet that commenced severing the plate. At 4 ms, a substantial quantity of metal sparks was detected through the viewing window, resulting from the synergistic influence of the metal jet and explosion products. The sparks repeatedly reacted with atmospheric oxygen during the next 2 ms, producing a significant amount of light. Approximately 6 ms later, the reaction intensity commenced to diminish, and the sparks progressively lessened before vanishing.

Fig. 6
figure 6

Results of the titanium fire elimination experiment-#1.

Full size image
Fig. 7
figure 7

Results of the titanium fire elimination experiment-#2.

Full size image

Figure 7 illustrates the procedure subsequent to the incorporation of the barrier layer. At approximately 0.8 ms, the ignition time of the detonator occurs marginally later than that of the blank group experiment, and the extent of radiated light exceeds that depicted in Fig. 7 at 0.6 ms. In the following 4 ms, minimal sparks were produced, and subsequent to the coating of the refractory material, the illumination from the detonation of the explosive core was significantly diminished. At 5.6 ms, following the destruction of the barrier layer, a minor eruption of sparks occurred, and after 1.6 ms, metallic sparks erupted once more, reacting with atmospheric oxygen to emit a dazzling light that persisted for approximately 2.4 ms. The temporal course of the two experimental sets was analogous, with titanium flames becoming visible at approximately 4 ms, succeeded by secondary reactions that persisted in emitting light. A substantial decrease in light emission was distinctly evident following the addition of the barrier layer.

Quantitative determination of titanium flame brightness

The brightness value of an image denotes the intensity of brightness, quantified in candelas per square meter (cd·m−2) or nits. In digital photographs, brightness is generally associated with the grayscale values of pixels. In grayscale images, brightness corresponds to grayscale values, often ranging from 0 to 255, with 0 being black and 255 denoting white. Grayscale values can be directly interpreted or altered when assessing brightness21.

A quantitative investigation of the brightness of titanium fire produced during shaped charge cutting of titanium alloy is conducted by processing photos captured using high-speed photography to determine the average brightness across several photographs. Standard techniques for image brightness adjustment encompass the utilization of specialized image processing tools, programming environments like MATLAB, and Python. In MATLAB, the imread function is initially employed to read the image file, subsequently converting the color image to grayscale, and then the mean function is utilized to compute the average brightness value of the image pixels, thereby determining the average brightness value of the image21. The mean brightness-time curve is illustrated as depicted in Fig. 8.

Fig. 8
figure 8

Results of the titanium fire suppression experiment.

Full size image

At t = 0 ms, the two experimental groups demonstrated uniform beginning performance, with average brightness values of 10.70 and 10.75 cd·m−2, respectively. The mean luminance of both groups rose at 1.2 ms, with Experimental Group #1 attaining its initial apex of 26.21 cd·m−2 at 2.4 ms, correlating with the illumination produced by the explosive core of the linear shaped charge. Subsequent to the detonation of the charge core, a titanium fire emerged and engaged in a prolonged reaction, resulting in an initial fall in brightness, followed by a continuous increase, ultimately attaining a maximum of 49.00 cd·m−2 at 5.4 ms. Experimental group #2, with an obstacle layer coated with aluminum silicate refractory material, attained its initial peak of 13.61 cd·m−2 solely upon the ignition of the detonator. The luminosity of the core ignition was markedly diminished thereafter and did not exhibit the same rising trajectory as experimental group #1. At 6 milliseconds, a minor quantity of metallic sparks emerged, resulting in a resurgence of brightness to a maximum of 12.1 cd·m−2. In comparison to experiment group #1, the performance time trajectory of both groups was analogous, peaking at approximately 5–6 ms; nevertheless, the luminance of experiment group #2 was merely 12.1 cd·m−2, with a titanium fire suppression rate reaching 75%. Figure 9 illustrates a three-dimensional representation of the gray values from the two sets of experimental findings at approximately 6 ms.

Fig. 9
figure 9

Graphic model of gray value of titanium fire suppression experiment results.

Full size image

The process by which the barrier layer mitigates titanium fire is examined by numerical modeling and the results from titanium fire suppression experiments as follows: The barrier layer efficiently obstructs and postpones direct interaction between detonation products and titanium alloy, while maintaining the cutting efficacy of the jet knife, thus markedly diminishing the swift temperature increase and redox reactions induced by high-temperature detonation products in the titanium alloy. The pressure at the jet cutter’s head remains at 69 GPa after traversing the barrier layer, the energy dissipation rate is below 20%, and the jet head velocity is 2388 m/s, thereby ensuring the effective cutting capability of the titanium alloy plate. This study also confirms the viability of employing the barrier layer approach to mitigate titanium combustion during the linear shaped charge of titanium alloy plates.

Conclusion

This paper investigates the mechanism of titanium fire generation through simulation calculations and conducts titanium alloy cutting experiments using aluminosilicate materials as an obstacle layer, aiming to study the technology for eliminating titanium fire caused by explosive detachment in aircraft engines. The conclusions are as follows:

(1) The protective mechanism of the barrier layer involves regulating the titanium alloy substrate and laying the barrier layer along the linear shaped charge. When the jet cutter penetrates the barrier layer, it only forms a microchannel with a width of 0.2 mm, resulting in minimal impact on the cutting efficiency of the jet at the tip. An undamaged barrier layer can effectively delay and reduce the direct contact between detonation products and titanium, significantly suppressing the intensity of the titanium fire reaction.

(2) The use of a thin barrier layer can effectively suppress the titanium fire generated during shaped charge cutting of TC4 titanium alloy. The experimental results show that the titanium fire suppression rate of the aluminosilicate coating is as high as 75%.

This study presents an efficient technique for mitigating titanium fire during the shaped charge cutting of titanium alloy blades, offering a technical solution to the issue of titanium fire interference observation during the cutting and separation of blades in aero-engine containment tests. The experimental and computational findings corroborate each other, offering a realistic and efficient technical solution to the issue of brilliant titanium fire obstructing visibility during the shaped charge Data availability.However, this study only investigates static cutting experiments. After drying, the coating material may exhibit delamination or thermal cracking during dynamic high-speed rotational testing. Future research could explore the performance of aluminosilicate materials under high-speed vibration conditions.