Abstract: Q-P martensitic steel is the latest developed third-generation structural steel. Specifically, the carbon-rich retained austenite can be formed by austenitizing the steel of medium and low carbon, high silicon and low alloy, quenching it to the temperature between the beginning of martensite transformation (MS) and the end of martensite transformation (MD), and then distributing the carbon element from martensite to austenite for a short isothermal time above the quenching temperature or MS temperature. At present, because of its excellent strength, plasticity and easy realization, it is often used in the production of automobile industry. However, its application in impact wear condition needs to be studied. With the rapid development of metallurgical and chemical industries, such as manganese steel and bainitic steel, it is increasingly difficult to meet the requirements of high impact wear resistance. Q-P steel is expected to become a new generation of wear-resistant materials due to its low cost and high performance. However, the effect of carbon content on the impact abrasive wear behavior of Q-P martensitic steel remains unclear. In order to reveal the mechanism of carbon content on the impact abrasive wear resistance of Q-P martensitic steel, this paper compared the wear loss, wear mark surface and sub surface morphology of two kinds of Q-P martensitic steel with carbon mass fraction of 0.3% and 0.4% respectively. To be specific, the impact of ore on the ball mill in reality was simulated, and gravel was used as the wear medium. After Q-P heat treatment of two kinds of independently developed cast low-alloy steels with carbon content, they were processed into a rectangle with a size of 10~30 mm³ and an arc shape at the front end. In addition, the lower test block (45 steel with hardness value of 455 HV) was impacted at the rate of 100 times/min using a drop hammer with the total impact energy of 3j and the test sample. During this period, quartz sand of 0.18~0.25 mm was used to flow through the impact surface at the rate of 50 kg/h. The sample should be weighed after ultrasonic cleaning and drying every half an hour. After the test, the wear morphology of the sample surface and the cross section of the sub surface layer were observed and tested by scanning electron microscope and projection electron microscope. Combining with the structural damage, this paper analyzed the formation mechanism of the sub standard layer of the sample. The V-notch impact test was carried out on the sample, the micro Vickers hardness value of the matrix structure and sub surface was measured, the carbide content on the matrix surface was analyzed by electron backscatter diffraction, and the residual austenite content was analyzed by X-ray. As shown by the results, the impact wear resistance decreased by about 7% with the increase of carbon content. In the process of impact wear, the material loss of Q-P martensitic steel was mainly ploughing and fatigue spalling. Due to the small difference in hardness between the two sub surfaces, the decrease of impact wear resistance was mainly attributed to the increase of fatigue cracks. In the process of impact wear, the nanostructure of lath structure and the transformation of residual austenite to martensite occurred on the sub surface layer. At the same time, the initiation of fatigue crack mainly came from the difference of microstructure and hardness between the interface structure of deformation layer and matrix layer. In conclusion, with the increase of carbon content, carbides precipitate in the matrix of Q-P martensitic steel, which intensifies the initiation and propagation of fatigue cracks in the process of impact wear, and finally triggers the reduction of impact abrasive wear resistance of the material.