Hole quality and thermal defects in drilled CFRP by nanosecond pulsed laser

To explore the effect of nanosecond pulse laser drilling on the quality of carbon �ber reinforced polymer (CFRP). The effects of laser parameters such as laser power, scanning speed, pulse width and laser frequency on micro-hole quality and the thermal damage generated during the drilling process, were studied by using the univariate methods and the orthogonal experimental method for drilling micro-holes with a diameter of 0.4 mm using a 1064 nm �ber laser. The results indicate that with the use of pumping at the entrance, the hole diameter and the heat-affected zone (HAZ) width at the entrance and exit increased signi�cantly with the increase of laser power, pulse width and frequency, and the decrease of scanning speed. The variation in HAZ width at the exit was correlated with whether the energy reached the carbon �ber ablation threshold. All the holes were tapered and the hole taper was closely related to the magnitude of the change in the laser parameters on the hole diameter at the entrance and exit. Holes with a taper of about 0.3 were obtained when the parameter combination was 99 W, 50 mm/s, 13 ns and 1500 kHz. Experiments indicate that laser power and pulse width are important factors affecting the quality of micro-hole processing. However, thermal damage defects such as striations, micro-cracks, delamination, voids and surface �ber ‘�sh scale’ peeling can occur during nanosecond laser drilling.


Introduction
With the growing demand for lightweight structural products, composite materials have attracted increasing attention.To achieve both lightweight and structure-function integration, carbon ber reinforced polymer (CFRP) [1] has emerged as a typical representative composite material.CFRP has excellent properties such as high speci c strength, low density, strong corrosion resistance, high design capability and good fatigue resistance [2], and is widely used in aerospace, transportation, medical equipment, national defense and military products and other elds [3,4].Currently, CFRP products cannot be molded once before it meets people's needs and still needs to be processed by cutting, drilling, and other processing.However, the anisotropy of carbon ber composites poses great challenges to high-quality processing [5].Traditional mechanical drilling can easily lead to the formation of defects such as high material delamination, burrs, and cracks at the edge of the hole, causing severe tool wear and increasing processing costs [6,7].
In contrast, non-traditional machining, electrical discharge machining (EDM) [8] and ultrasonic machining [7] are ine cient and produce more hole defects [9].The resonance frequency matching problem of ultrasonic machining also needs to be solved [7].Abrasive water jet machining has low thermal damage, but the machining accuracy is di cult to control, resulting in defects such as high surface roughness and material delamination [10].In addition, due to the hygroscopic properties of CFRP, the mechanical properties are signi cantly reduced by abrasive water jet processing [11,12].The laser processing technology is a new process with high machining accuracy, high e ciency, no tool wear, non-contact, low cost and easy control [13], which can effectively reduce hole defects in CFRP hole drilling.However, laser processing of CFRP achieves thermal ablation removal of materials by absorbing laser energy.Due to the large difference in thermal properties of carbon ber and matrix material, it will inevitably cause thermal damage, especially the HAZ, which will greatly reduce the mechanical properties of the material [14,15].
To address the above issues in the laser processing of CFRP, Herzog et al. [16] compared the effects of pulsed Nd: YAG laser, disk and CO 2 lasers on the cutting quality and static strength of CFRP at different laser powers.The results showed that pulsed lasers could reduce the contact time between the laser and the material.Compared to the other two lasers, a smaller HAZ, higher static strength (1000 MPa) and bending strength (2600 MPa) were obtained.Li et al. [17] investigated the hole quality and tensile strength of CFRP using ber lasers in continuous and pulsed modes.It was found that the pulsed laser could achieve better processing quality.Ramanujam et al. [18] studied the hole quality of CFRP materials after continuous CO 2 laser processing and found that irregular micro-hole pro les appeared at the entrance and exit under all parameter conditions.Zhou et al. [19] analyzed the physical process of QCW ber millisecond laser processing of CFRP.They pointed out that hole drilling involves physical processes such as thermal evaporation, melting, plasma impact, mechanical ablation, pyrolysis and carbonization.Sato et al. [20] compared the effects of N 2 and air assistance on the quality of CFRP processed by a 1064 nm pulsed ber laser.The results showed that N 2 assistance could effectively inhibit HAZ formation and increase the cutting speed.Romoli et al. [21] showed that the epoxy resin hardly absorbs the 1064 nm laser but is absorbed by the carbon bers.Therefore, under the laser irradiation of this wavelength, it is mainly the heat conduction of the carbon ber that forms the HAZ.Takahashi et al. [22] analyzed the quality of Nd: YAG laser cutting of CFRP at 1064 nm and 266 nm from theoretical simulation and experiment.The results showed that the 1064 nm laser passes through the resin and heats the carbon ber strongly.Tao et al. [23] demonstrated that the 1064 nm ultra-short pulse laser drilling of CFRP laminates produces the bowl-bottom effect of the hole taper.This is related to the Gaussian distribution characteristics of the laser beam (taper initial formation stage I), the decrease of the laser average uence due to the expansion of the spot area and its projection area (taper accumulation stage II), and the secondary processing (bowl-bottom formation stage III ).
The laser drilling process mainly removes the molten material laterally and discharges it axially along the hole depth.In addition to the in uence of laser parameter factors, the choice of drilling method also helps to reduce the recast layer and improve the geometric accuracy of laser drilling [24,25].Li et al. [26] explored the in uence of different scanning strategies on hole drilling, and the results showed that spiral scanning could signi cantly reduce the width of the HAZ.Wang et al. [27] investigated the effects of nanosecond pulsed laser beam jump directions (outside-in and inside-out) and scanning modes (sequential and cyclical scanning) on the hole sidewall morphology in a spiral scanning mode.The results indicate that the outside-in laser scan path can achieve a atter hole exit morphology at the bottom.Li et al. [28] reported that the interlaced scanning mode can effectively reduce the heat accumulation effect of adjacent trajectories compared with the default sequential scanning mode.Appropriate scanning spacing can balance the two removal mechanisms of laser ablation and mechanical denudation in laser processing [29], and improve the processing quality and e ciency.
In summary, the pulsed laser can achieve better processing quality; the spiral scanning method of hole drilling is bene cial to the reduction of thermal defects; the carbon ber absorbs the laser energy at a wavelength of 1064 nm best, and the processing does not consider the resin matrix to absorb the laser energy at this wavelength.Meanwhile, ber laser has the advantages of high photoelectric conversion e ciency, low cost, and high stability, etc., is widely used in cutting, hole drilling and other industrial elds, is currently a hot spot for the development of industrial applications.However, the in uence of laser parameters on the quality of hole-making for nanosecond pulsed ber laser drilling of multi-orientation bers (MD CFRP) has not been su ciently studied, and the analysis of the thermal damage pattern obtained by pulsed laser drilling of CFRP is still lacking.Therefore, in this work, a 1064 nm nanosecond pulsed ber laser was used to produce 0.4 mm diameter microholes in 1 mm thick CFRP to investigate the corresponding hole-drilling phenomena and patterns.A controlled variable method was used to study the effect of laser processing parameters (laser power, scan speed, pulse width, laser frequency) on the hole quality during the hole drilling process.The response indicators of micro-hole quality were the hole diameter and the heat affected zone at the entrance and exit, and the taper.The L16 orthogonal experiments were designed, and the primary factors of the in uence of laser parameters on micro-hole quality were determined from the calculation results.The thermal defects were also analyzed based on the microscopic morphology of the micro-hole.This study contributes to the optimization of laser hole drilling parameters and provides a theoretical reference for the application of 1064 nm nanosecond pulse laser processing CFRP.
2 Experimental Details

Experimental materials and setup
The CFRP specimen is a laminate composed of carbon ber and epoxy resin matrix.The laminated structure is a multi-layer 0°/90° braided arrangement.The size of the length × width × thickness is 30 mm × 30 mm × 1 mm, respectively, as shown in Fig. 1a.
The experimental platform is shown in Fig. 1b and consists of a laser, beam splitter, galvanometer, 3D translation stage (movement accuracy of 1 µm), power meter, inspiratory device and computer.The laser is a pulsed nanosecond ber laser with a wavelength of 1064 nm and an average power of > 100 W. The pulse width is adjustable from 2 ns to 500 ns, the laser frequency is adjustable from 1 kHz to 4000 kHz, and the exit beam diameter is 7 mm.Part of the laser beam is re ected through the beam splitter into the power meter for real-time monitoring of the laser power, while the other part of the laser beam enters the galvanometer and is then irradiated onto the front surface of the CFRP sample through a focusing eld mirror lens with a focal length of 200 mm.The CFRP sample is placed on the sample holder of the 3D translation stage and clamped vertically to reduce the amount of carbon ber debris adhering to the front surface of the workpiece.An inspiratory device is used to pump air onto the front surface of the specimen to lter the smoke and exhaust gases generated during the drilling process to avoid harm to the experimental operators and environmental pollution.The laser parameters used in the experiments are shown in Table 1.

Experimental strategy
This work focuses on the effects of laser parameters such as power, scanning speed, pulse width, and frequency on the entrance and exit hole diameters and the taper, as well as the HAZ.The spiral scanning mode and the scanning sequence from outside to inside shown in Fig. 1c were used in the experiments.The focal spot measured by the beam quality analyzer is shown in Fig. 1d with a diameter of 50 µm (1/e 2 ).Combined with the energy distribution of the beam, the xed pitch during scanning is 0.02 mm (i.e., a path overlap of 60%).
The in uence of the laser parameters on the hole quality was analyzed using a single-factor experimental method.Based on the results of the single-factor experiment, a four-factor, four-level L16 orthogonal design experiment shown in Table 2 was designed to determine the primary and secondary relationships between the four factors of power, scanning speed, pulse width, and frequency on hole quality.The experiment was repeated at least three times for each group of parameters to avoid contingency in the results.

Measurement/Characterization
In uenced by factors such as the distribution of the laser beam and the thickness of the sample, the entrance and exit diameters of the micro-holes produce a certain diameter difference, which is expressed as a tapered hole.The calculation formula is: where t is the hole taper, D Entry and D Exit are the entry and exit hole diameters, respectively.T is the thickness of the CFRP.In the experiment, the difference in hole diameter of the micro-holes obtained for adjacent laser parameters for the same single factor variation is de ned as the magnitude of the variation ( ), and the calculation equation is as follows: where and are the magnitudes of the variation of the entry and exit hole diameters, respectively.X is a single dependent variable parameter, i.e. laser power (P), pulse width (τ), frequency (F) and scanning speed (V).i (i = 1,2,3,4,5) and j (j = 2,3,4,5) are consecutive numbers of variable parameters.Details are given in the accompanying notes to Table 1.
The thermal conductivity in the direction parallel to the carbon ber (50 W/(m•K)) is higher than that in the vertical direction (5 W/(m•K)) [19], so the HAZ was taken as the object of inspection with the laser scanning direction perpendicular to the carbon ber, as shown in Fig. 2a.A stereomicroscope was used to observe and measure the hole diameters at the entrance and exit, as well as the corresponding HAZ width.The measurement scheme is shown in Fig. 2. Multiple measurements were taken for each group of data, and the average value was selected for statistical analysis.Scanning electron microscopy (SEM) was used to measure the morphology of the microholes and to analyze the thermal damage defects.
3 Results and Discussion

Laser parameters versus hole diameter
Figure 3 and Fig. 4 show the entrance and exit micro-hole morphologies obtained for different laser parameters, respectively.It can be seen that D Entry is consistently larger than D Exit for different laser parameters.It is associated with the Gaussian distribution of the laser beam and the fact that the focal plane is maintained at the front surface during hole drilling, leading to a decrease in laser energy density with increasing depth.In addition, the high energy causes evaporation and burning of the material as the laser beam moves along the scan path.
The ablative debris produced in the process carries away some of the heat.Meanwhile, the ablation material that has not been removed in time and has accumulated around the hole wall blocks some of the laser energy, creating a shielding effect that reduces the energy absorbed by the material along the depth of the hole.
Figure 3 shows the expansion of the fractured ber ends and the convexity of the carbon ber layers around the micro-hole, which is directly related to the signi cant thermal expansion of the carbon ber and the superposition of thermal deformation of the individual ber layers.Meanwhile, the HAZ formed during the laser irradiation and the lack of resin reinforcement of the carbon bers became loose, which in turn led to phenomena such as carbon ber pull-out and pick-up in Fig. 3d and h.This phenomenon becomes more pronounced as the laser power, pulse width and frequency increase, and the scanning speed decreases.In addition, the vaporized material is ltered by pumping in the experiment, and the suction of this device may also cause the carbon bers to be picked up and then reattached to the surface.In contrast, for the rear surface shown in Fig. 4, only the broken carbon bers were picked up, as shown in Fig. 4c, and no pulling out of the carbon ber occurred.This may be related to the reduced energy absorbed by the material and the area of in uence of the HAZ.
For any combination of laser parameters, all of the micro-holes in Fig. 3 contain the notched pro le edges indicated by the arrows in Fig. 3a.This is because the start of the spiral scan path does not overlap with the adjacent spirals, and there is a distance gap between the scan spacing as shown in Fig. 1b, which leads to this defect in the drilling process.The gap is just not as obvious in some images, which is related to the amount of heat accumulation caused by different laser parameters.
In uenced by the reduced energy absorption in the thickness direction of the material, the laser dispersion outside the focal plane, the change in thermal conductivity in the ber direction and the high processing temperature [18], the micro-holes on the rear surface shown in Fig. 4 exhibit irregular contours and elliptical holes, the circularity is much smaller than that of the front surface.Meanwhile, the energy absorbed at the rear surface of the material is not su cient to vaporize the resin matrix directly.The temperature of heat transfer through carbon bers is also lower than the vaporization temperature of the resin.However, it has reached its liquefaction phase transition temperature, so a residue of pyrolysis and melting and then re-curing appears in the area near the micro-holes on the rear surface as shown in Fig. 4o and q.
Figure 5 shows the quantitative relationship between the hole diameter at the entrance and exit of the micro-hole and the laser parameters shown in Fig. 3 and Fig. 4.Both D Entry and D Exit become larger as the laser power, pulse width and frequency increase, and as the scanning speed decreases.This is directly related to the material removal rate.As the laser power and pulse width increase, the corresponding increase in laser energy directly leads to an increase in melting and vaporization of the material, which effectively enhances the material removal capability, and in turn, leads to progressively larger hole diameters of the D Entry and D Exit .The laser frequency and scanning speed inevitably affect the interaction time between the laser and the CFRP material.As the laser frequency increases and the scanning speed decreases, the time taken to heat the material increases, resulting in more material being removed and an increase in the diameter of the entry and exit holes.

Laser parameters versus taper
Figure 6a-f show the relationship between the taper and the laser parameters in Fig. 3 and Fig. 4, and the corresponding variation curves of D Entry and D Exit .It can be seen that when the laser power, pulse width and laser frequency are changed, the variation range of the exit diameter (ΔD Exit ) is rst larger than the variation range of the entry diameter (ΔD Entry ) and then smaller than ΔD Entry .However, when the scanning speed is changed, the corresponding pattern is exactly the opposite.
According to Eq. ( 1), the magnitude of the taper is mainly related to the difference in diameter at the entry and exit when the material thickness T is constant.From Fig. 6a-f, it can be seen that the smallest taper is obtained when the power, pulse width and frequency are 99 W, 13 ns and 1500 kHz, respectively, which are 0.296, 0.295 and 0.304.As the power, frequency and pulse width increase, so does the taper.As can be seen from Fig. 6g and h, the taper gradually decreases as the scanning speed increases and levels off at 50 mm/s, which is related to the fact that ΔD Exit is always greater than ΔD Entry .After increasing the speed to 50 mm/s, the variation of the entry and exit diameters is close to each other, and nally, a micro-hole with a taper of 0.295 can be obtained.In summary, changing the laser power, pulse width and laser frequency results in the same variation trend for the taper.The opposite trend is obtained by changing the scanning speed.
It can be seen that as the laser power, pulse width and frequency increase to a certain value, the laser energy absorbed by the material on the rear surface uctuates signi cantly, the energy utilization of the material removed on the rear surface increases, and ΔD Entry is smaller than ΔD Exit .However, as the value of the laser variable parameter continues to increase, the plasma shielding effect of the front surface increases, while the front surface also absorbs more laser energy, resulting in more intense ablation, increased material removal, and a signi cant increase in D Entry .The rear surface absorbs relatively little change in energy, material removal is limited, and the trend of increasing D Exit is relatively stable.This indicates that too much or too little laser power, pulse width and frequency will increase the hole taper.In this experiment, when the laser parameters were combined at 99 W, 50 mm/s, 13 ns, and 1500 kHz, a micro-hole with a taper of about 0.3 was obtained.

Laser parameters versus HAZ
CFRP consists of carbon ber and epoxy resin with different components.The thermal properties of the two materials are very different, resulting in CFRP being an anisotropic material.Carbon bers have a much higher vaporization temperature (3900 K) than epoxy resin (698 K) and have a better thermal conductivity [19].When the CFRP is irradiated by laser, the material absorbs heat, the epoxy resin reaches the vaporization temperature earlier, the resin matrix is removed rst, exposing the unremoved carbon bers and forming a HAZ.
Figure 7a shows that as the laser power increases, the heat absorbed per unit time on the surface material increases, except for the carbon ber material in the center of the micro-hole, which is removed, the heat accumulation around the micro-hole also increases, which does not reach the vaporization temperature, but already far exceeds the vaporization temperature of the resin matrix, making its heated vaporization area further expand, which in turn leads to a gradual expansion of the HAZ width at the entrance (HAZ Entry ).When the pulse width is increased, the heating time of the carbon ber material becomes longer and the cooling time is shorter, which is more likely to cause heat accumulation in the irradiated area, resulting in the HAZ variation shown in Fig. 7b.Where φ, V, d and F are the spot overlap rate, the scanning speed, the spot diameter and the laser frequency, respectively.According to Eq. ( 4), φ is proportional to the laser frequency and inversely proportional to the scanning speed when the spot diameter is constant.Increasing the frequency or decreasing the scanning speed can effectively reduce the amount of heat accumulation in the processed area of the material, and the area where the resin matrix fades is reduced, which helps to reduce the HAZ.When the pulse width is constant, increasing the laser frequency will decrease the time between adjacent pulses, decrease the cooling time in the processed area of the material, and increase the HAZ Entry .
The HAZ Entry versus laser parameters shown in Fig. 7 indicates that the HAZ Entry increases as the laser power, pulse width and frequency increase, and as the scanning speed decreases.The SEM image in Fig. 8 shows that as these parameters change, it leads to more carbon ber fracture around the micro-holes, increased ber tip swelling, signi cant carbon ber pull-out, as well as the formation of obvious cracks and striations.Meanwhile, it causes some ablation debris and matrix degradation residues to adhere to the raised carbon bers, which inevitably affects the mechanical strength and mechanical properties of the CFRP.
Figure 9 shows the relationship between the HAZ width at the exit (HAZ Exit ) and the laser parameters.The HAZ Exit is larger than the HAZ Entry at 93 W power, pulse width ≤ 9 ns and scan speed ≥ 58 mm/s.Otherwise, the HAZ Exit is smaller than the HAZ Entry .Figure 9a shows that the HAZ Exit rst increases and then decreases with increasing laser power, with the maximum HAZ Exit (441.163µm) being obtained at 96 W laser power.When the laser power exceeds 102 W, the HAZ Exit increases again, so that the minimum HAZ Exit (377.022µm) for this single dependent variable is obtained at 102 W. Figure 9b shows that increasing pulse width leads to a decrease and then an increase in the HAZ Exit , with a minimum HAZ Exit (325.942µm) obtained at a pulse width of 13 ns.The increase slows down when the pulse width is greater than 20 ns.
Figure 9c shows that the HAZ Exit reaches a maximum at the laser frequency of 1500 kHz.As the frequency increases, there is a signi cant decrease in the HAZ Exit .When the frequency reaches 2000 kHz, the HAZ Exit is 312.765µm.The effect of the scanning speed is similar to that of the pulse width, and Fig. 9d shows that the smallest HAZ (268.366µm) is obtained at a scanning speed of 50 mm/s.The variation of HAZ Exit is mainly related to whether the energy absorbed by the material reaches the carbon ber ablation condition after changing the laser parameters, while the relationship between each parameter and the energy is consistent as described in section 3.1.When the energy reaches the carbon ber ablation condition, the energy utilization increases and the energy used for heat transfer dissipation decreases, resulting in a decrease in HAZ Exit .However, if the absorbed energy is not su cient to remove the carbon bers, the heat will evaporate the resin through the carbon ber heat transfer, increasing the HAZ Exit .In addition, if the energy density is too high, the epoxy resin matrix will be removed by both direct vaporization and indirect ablation by heat conduction, resulting in an even greater HAZ Exit .
The range of the HAZ shown in Fig. 7 and Fig. 9 uctuates considerably.It is because the thermal conductivity of carbon bers along the radial direction is 10 times higher than in the axial direction [19].The heat is mainly transmitted along the radial direction and diffuses into the interior of the material, but the total energy input remains constant, forming an elliptical heat-affected zone.The different contact locations of the laser beam with the material and the anisotropy of the carbon bers, when the laser irradiates the carbon bers with different weave orientations, lead to a difference in the extent of the HAZ, resulting in a large variation in the HAZ of the micro-holes obtained for the same parameters.

Micro-hole sidewall morphology and defects
Figure 10 shows the morphology of the micro-hole sidewall obtained at different laser parameters.It can be seen that striations, voids, ber pull-out, rings, ber fractures, tip swelling, ber delamination, epoxy resin coverage, and fragments occur within the micro-holes.It also leads to multi-physical complexities such as the formation of rough surfaces, ber damage, and interlayer cracks on the sidewall surface.
The striations and voids shown in Fig. 10a appear on the sidewall of the micro-holes for each combination of laser parameters.Negarestani et al. [30] attributed the striations to the unsteady motion of pyrolysis products of the epoxy, such as light gases, various hydrocarbons, and carbon.It was found that the number of striations and voids in Fig. 10f and i increased signi cantly when the laser frequency was less than 1000 kHz and the scanning speed was greater than 66 mm/s.This is attributed to the low overlap rate between adjacent pulse spots at these laser parameters, resulting in the presence of an incomplete removal area between the materials such that striations are visible.Voids are the thermal defects in the ablation of the resin matrix that resulting in matrix loss [26].During laser irradiation, less energy per unit of time is used to reach the carbon ber ablation threshold, more energy is dissipated by heat conduction rather than material removal, and the resin matrix is ablated over a wider temperature range, resulting in an increased number of surface voids.It has been found that the number of voids decreases with increasing hole depth.There are two main reasons for this phenomenon.One is that more laser energy is absorbed near the entrance of the micro-hole, the vaporization temperature of the carbon ber and resin is inconsistent, and the heat accumulation effect between the unremoved carbon ber is signi cant, resulting in more ablation of the epoxy resin matrix.Secondly, the shrinkage of the surface ber ends increased the formation of voids.Meanwhile, varying degrees of rings in the center of the ber end are also observed in Fig. 10a and other gures.This is related to the radial thermal melting of the carbon ber [31].It is just that the degree of thermal melting varies with different energy parameters and the formation of rings varies.
In addition, loosening occurs in the area near the surface of the micro-hole shown in Fig. 10b, and the outer carbon bers are easily pulled out.This phenomenon becomes more pronounced as the laser power, pulse width, and frequency increase and the scanning speed decreases.This is related to the thermal deformation of the carbon ber between the layers and the use of suction at the entrance.The suction pulls the heat out of the hole and carries it away to the entrance surface, thus exacerbating the heat accumulation effect within the entrance region.In addition, Fig. 10c shows that the laser scanning direction removes more material parallel to the carbon ber direction than perpendicular to it, with more fractured and damaged carbon bers.This is due to the higher heat conductivity along the carbon ber direction than in the vertical direction.
As the power, pulse width and frequency increase and the scanning speed decreases, the number of delamination and cracks in Fig. 10d-h increases, and the epoxy resin covering the processing section and wrapping the carbon ber gradually disappears.This is related to the increase of laser energy and heat accumulation.When the laser frequency reaches 2000 kHz, the processed cross-section shows a rough and uneven area shown in Fig. 10g, and the ber ends are already slightly swollen.This may be related to thermal damage caused by shortened cooling time or may be caused by incomplete thermal fusion occurring between carbon bers [31].At scanning speed below 34 mm/s, signi cant ' sh scale' peel damage and micro-cracks appear on the surface of the carbon ber as shown in Fig. 10h, where Fig. 10j shows a magni ed image of the box in Fig. 10h.The thermal damage defects caused by the laser scanning of the CFRP surface over a long time are the main cause of these phenomena.
Based on the morphological analysis of the micro-hole sidewalls, it can be concluded that when the combination of 99 W, 50 mm/s, 13 ns, and 1500 kHz is used to drill the holes, the carbon ber arrangement is not altered and leads to less thermal damage and machining defects.In addition, the t and HAZ are smaller for this combination of parameters.

Orthogonal test results and analysis
Table 3 shows the experimental results obtained from the four-factor, four-level L16 orthogonal design, and the statistical results shown in Table 4 are obtained based on the experimental data in this table.When D Entry , D Exit , HAZ Entry , HAZ Exit and t are used as indicators to evaluate the hole quality, laser power is the main factor affecting D Entry and HAZ Entry .Laser frequency has no signi cant effect on either indicator.The effect of scanning speed on D Entry and D Exit is more pronounced, while the pulse width only had a signi cant effect on D Exit , HAZ Exit and t.It is inconsistent with the conclusion of reference [19] that pulse width has the most signi cant effect on each quality factor.It may be related to the small range of pulse width in this orthogonal experiment.Meanwhile, Zhou et al. [19] used a pulse width of 0.1 ~ 0.25 ms, which is much larger than the pulse width used in this work.A larger pulse width of the laser inevitably leads to a larger thermal effect.In addition, the laser drilling of CFRP in this work uses outward pumping along the entrance of the sample, which interferes with the direction of heat conduction and affects the quality of holes at the entrance.
In summary, the effect of laser parameters on D Entry is: laser power > scanning speed > pulse width > laser frequency.The effect of laser parameters on D Exit is: pulse width > scanning speed > laser power > laser frequency.
The effect of laser parameters on HAZ Entry is: laser power > pulse width > laser frequency > scanning speed.The effect of laser parameters on HAZ Exit and t is: pulse width > laser power > scanning speed > laser frequency.4 Conclusions A drilling study was carried out on the surface of carbon ber-reinforced resin matrix composites using a 1064 nm nanosecond pulsed ber laser based on the univariate variables and orthogonal design experiments.The effects of laser parameters such as laser power, pulse width, laser frequency and scanning speed on the processing quality were analyzed.The conclusions are as follows: (1) The variation of micro-hole diameter and heat-affected zone width is positively proportional to the variation of laser power, pulse width and frequency, inversely proportional to the scanning speed.The hole surface without smoke accumulation at the entrance can be obtained by using the front inspiratory method, except that the taper Figures Figure 1 The     F=1500 kHz; e P=99 W, V=50 mm/s, τ=6 ns, F=1500 kHz; f P=99 W, V=50 mm/s, τ=13 ns, F=1000 kHz; g P=99 W, V=50 mm/s, τ=13 ns, F=2000 kHz; h P=99 W, V=34 mm/s, τ=13 ns, F=1500 kHz; (i)P=99 W, V=66 mm/s, τ=13 ns, F=1500 kHz; j is an enlarged image of the boxed area in h

Figure
Figure 7c and d show that laser frequency and scanning speed have different effects on the HAZ.It is due to the two parameters affecting the laser spot overlap rate.According to the formula:

Figure 5 Hole
Figure 5

Table 3
Orthogonal test results of laser drilling CFRP