Open Access
Add to BibSonomy
Abstract
High-quality ammonium dihydrogen phosphate (NH4H2PO4, ADP) crystals were grown in Z direction and in defined crystallographic direction (θ=90°, φ=45°) by the rapid growth method, respectively. Defect-induced damage behavior in 355 nm of three types of ADP samples cutting in type-II matching and third harmonic generation direction from the as-grown crystals were investigated, including the initial laser induced damage (LID) characteristics and the physical and chemical properties of defects which serve as the damage precursors. The evaluations of damage behaviors include the “sampling” laser induced damage threshold (LIDT) by 1-on-1 and R-on-1 methods, bulk damage growth and bulk damage morphology. UV-visible transmittance spectrum, ultraviolet absorption spectrum, fluorescence spectrum, positron annihilation spectrum and the online light scattering measurements were carried out to investigate the defect-induced damage behavior in ADP crystals. The study will provide a reference for the investigations on laser induced damage properties of ADP crystals in short wavelength.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction:
Ammonium dihydrogen phosphate (NH4H2PO4, ADP) crystals are the typical isomorphs of potassium dihydrogen phosphate (KH2PO4, KDP) crystals, with large nonlinear optical coefficient, high transmittance, high LIDT and the superiority of growing into large sizes [1–3]. Owing to these advantages, ADP crystals are widely utilized in short-wave-length laser technology and high-power laser systems as nonlinear optics, frequency converters and optical parametric oscillators (OPO) et al [4–6]. As reported, laser induced damage (LID) will occur on the surface and bulk of KDP-type crystals under high power laser irradiation, with the laser induced damage threshold (LIDT) far lower than the intrinsic damage threshold in theory [7,8]. Therefore, low LIDT induced by bulk defects as damage precursors severely restrict laser energy output [9,10]. At present, it is generally believed that the defects related to LID performance in (D)KDP crystals mainly include inclusions, impurity ions, intrinsic point defects (clusters) and other kinds of structural defects, etc [11–17]. And as the chemical structures and the physical properties of KDP and ADP crystals are analogous [18–20], the development of damage process dominating by defects in KDP/ADP crystals both exist, while these processes are not thoroughly and profoundly understood . Early considerable effort had been made to establish the mechanisms of LID in ADP crystals and the investigations stated that the earliest damage to ADP crystals is caused by inclusions with the sizes of less than 1 µm [21]. Then the later studies mainly focused on the “sampling” LIDTs (which were relative to the concept of functional LIDT) of ADP crystals [22–24], and few investigations on the relationships of defects and LID in ADP crystals had been done in the past years. Therefore, gaining a detailed investigation to the initial LID properties and defect-induced damage behaviors of ADP crystals is essential for their applications in high power laser systems.
As researched in our earlier works, small size high-quality ADP crystals were grown from aqueous solution by the “point-seed” method in defined crystallographic direction (θ=90°, φ=45°), to impede the dislocations inherited from the seed and enhance the utilization rate in direction of fourth harmonic generation [23,25]. In this work, large size ADP crystals were grown by the rapid growth method in Z direction and in defined crystallographic direction (θ=90°, φ=45°). The samples for the study on defect-induced damage behaviors were cut in type-II matching and third harmonic generation direction from the as-grown large-size ADP crystals. LIDTs by 1-on-1 and R-on-1 methods, bulk damage growth and bulk damage morphology which exhibited the initial LID characteristics in 355 nm were systemically investigated. And UV-visible transmittance spectrum, ultraviolet absorption spectrum, fluorescence spectrum, positron annihilation spectrum and the online light scattering measurements were implemented to investigate the defect-induced damage behavior in ADP crystals. The study will provide a reference for improving the understanding on the laser induced damage properties of ADP crystals in short wavelength.
2. Experimental procedures
High-quality ADP crystals with large size were grown in Z direction and in defined crystallographic direction (θ=90°, φ=45°) by the rapid growth method, respectively, as shown in Fig. 1. The detailed growth procedures were described in Ref [23,25]., while the differences were that the volume of growth tank enlarging to 20 L and the sizes of the as-grown ADP crystals with 95 mm×98 mm×85 mm (Z direction) and 82 mm×78 mm×181 mm (defined direction), respectively. Three types of ADP samples for the study on defect-induced damage behaviors were cut from the as-grown large-size rapid growth ADP crystals as shown in Fig. 1 and conventional growth ADP crystals grown by other ones in our term [26] (C, Z-Pr, Z-Py, D-Pr and D-Py were used to define the ADP crystal grown by conventional method, the prismatic and pyramidal sectors of ADP crystal grown by rapid growth method in Z direction and the prismatic and pyramidal sectors of ADP crystal grown by the rapid growth method in defined direction, respectively). The samples were 50 mm×50 mm×12 mm in type-II matching and third harmonic generation direction, with uncoated, diamond-turned finished surfaces.
The measurements of initial LID characteristics in 355 nm were carried out via a Nd: YAG pulsed laser, with 1Hz repetition and 5.8 ns pulse width. The LID testing system had been described in details elsewhere [27]. The “sampling” LIDTs were conducted by 1-on-1 and R-on-1 methods. In measuring the damage growth properties, 1-on-1 method at a certain laser fluence was implemented to induce initial damage sites, while the damage growth thresholds resting with the LIDTs of ADP crystals were tested by applying the R-on-1 method. Besides, the morphology of LID points of ADP crystal irradiated by 355 nm laser was collected and analyzed by a microscope from the direction perpendicular to laser transmission.
LS5-Lambda 950 ultraviolet-visible spectrophotometer was adopted for the transmittance spectrum and ultraviolet absorption spectrum of ADP crystals, with 1 nm spectral resolution and accuracy of ± 0.5 nm. LS-45/55 fluorescence photometer was employed to the fluorescence spectrum including the excitation and emission spectrums. The positron annihilation lifetime spectrum was measured by fast-slow coincidence measurement technology, with 2 million total lifetime spectrum counts to ensure statistics. The time resolution of the positron annihilation spectrometer was about 195 ps, and the electronics plug-in of the measurement system was the standard NIM plug-in of EG&G Company of the United States. The schematic diagram of in situ scattering system was shown in Fig. 2. A 532nm continuous laser was divided into two laser beams by a polarizing beam splitter prism, which respectively incident from the front surface and the top surface of the crystal. After being reflected by a reflector in the top view direction, the laser entered the crystal again in the reverse direction to ensure the crystal irradiated by laser in the two-dimensional direction. An off-axis high depth-of-field long-distance microscope was arranged on the side surface of the crystal to realize real-time monitoring.
3. Results and discussions
3.1 Initial laser induced damage characteristics
3.1.1 “Sampling” laser induced damage threshold
The LID probability curves of ADP crystals with different growth methods under 355 nm laser irradiation were obtained by 1-on-1 mode, as shown in Fig. 3 and Table 1. The results indicate that the LIDTs of ADP crystals with different growth processes are different. C has the highest LIDT of 7.7 J/cm2, while LIDTs of Z and D are inferior. The types, quantity and density of defects in the pyramidal and prismatic sectors of ADP crystals are in discrepancy, resulting in the significant three times difference in LIDTs of the two growth sectors, namely that the LIDTs of Z-Pr, Z-Py, D-Pr, D-Py are 1.7 J/cm2, 4.9 J/cm2, 1.8 J/cm2 and 5.0 J/cm2, respectively. Therefore, the overall defect distribution and growth quality of ADP crystals grown by different methods can be evaluated from the LIDTs by 1-on-1. The C-ADP crystal displays the best quality, followed by the directional ADP crystal and the ADP crystal grown in Z direction.
Table 1. LIDTs of ADP crystals with different growth methods by 1-on-1 method. (C, D-Pr, D-Py, Z-Pr and Z-Py represent the conventional method grown ADP crystals, the prismatic sector and pyramidal region of defined directional rapid growth ADP crystal, and the prismatic sector and pyramidal region of z-direction rapid growth ADP crystal, respectively)
The LIDTs of ADP crystals with different growth methods under 355 nm laser irradiation were obtained by R-on-1 method, as shown in Fig. 4. As the results point out, the LIDT of C-ADP is the highest with 16.0 J/cm2, indicating that the quantities of the defects inducing initial damage are the smallest in the bulk. The LIDT of Z-Pr, Z-Py, D-Pr and D-Py are 10.8 J/cm2, 10.8 J/cm2, 11.8 J/cm2 and 11.8 J/cm2, respectively. As the data shows that the LIDT between different regions is identical on both samples, which suggest that conditioning suppresses the effects of the lower energy level defects. In addition, comparing the LIDT of Z and D samples obtained by R-on-1 method, it is inferred that the growth quality of defined directional ADP crystal is better than that of Z-ADP crystal in a whole.
3.1.2 Damage growth characteristics
The growth trends of initial bulk damage of ADP crystal samples grown by different methods under 100 laser pulses with different fluences were investigated. In the damage growth test, the initial damage site was obtained by 1-on-1 method, while R-on-1 method was adopted in the damage growth measurement. The typical LID growth images of ADP crystals with different growth methods are shown in Fig. 5.
Fig. 5. The typical laser induced damage growth images of ADP crystals with different growth methods. (The white circle reflects the appearance of the new damage sites.)
For C-ADP crystal, a certain number of initial bulk damage sites were induced after one laser pulse with the laser fluence of 11.5 J/cm2 irradiating. Then, the sample was irradiated with a gradient of 6.9 J/cm2, 8.5 J/cm2, 10.2 J/cm2, 11.3 J/cm2, 12.7 J/cm2 and 14.1 J/cm2 to obtain the damage growth characteristics, as shown in Fig. 5(a). It can be seen that the number, density and size of initial damage points occurring at the fluence of 11.5 J/cm2 did not change after 100 laser pluses with different fluences, until new scattering points appeared at the position marked by the yellow circle in Fig. 5(a) as the 24th pulse laser with an fluence of 14.1J/cm2 irradiated. Therefore, this laser fluence became the LID growth threshold referring to the LID density growth threshold due to none change of the size of initial damage sites. In addition, in order to reduce the experimental errors, the LID damage growth threshold was tested at tens of locations. After taking the average value, the damage growth threshold of C-ADP crystal at the initial damage site induced by the 1-on-1 method is about 14.1 J/cm2 as listed in Table 2. Figure 5(b) and 5(c) are typical laser induced damage growth images of D-Py and D-Pr, respectively. The laser fluence inducing the initial bulk damage and the irradiating laser gradient achieving the damage growth properties are greatly associated with the LIDTs of ADP crystals. Therefore, initial damage sites appeared in the prism and pyramid after one laser pulse with fluence of 3.1 J/cm2 and 9.2 J/cm2, respectively. The laser fluence gradient in prismatic sector ranges by 0.2 J/cm2, 0.8 J/cm2, 1.8 J/cm2, 3.0 J/cm2, 4.4 J/cm2 and 6.1J/cm2, while that in pyramidal region ranges as 4.5 J/cm2, 6.0 J/cm2, 7.6 J/cm2, 9.3J/cm2, 10.8 J/cm2, 12.2 J/cm2 and 13.1 J/cm2. New scattering points appeared at the position marked by the yellow circle in Fig. 5(b) as the 4th pulse laser with an fluence of 6.1 J/cm2 irradiates and in Fig. 5(c) when the 14th pulse laser with 13.1 J/cm2 fluence came. Therefore, the LID growth thresholds of the D-Pr and D-Py are 6.1 J/cm2 and 13.1 J/cm2, respectively. The typical LID growth images of Z-Py and Z-Pr are separately exhibited in Fig. 5(d) and Fig. 5(e). After similar initial damage induction and measurements of LID growth properties of oriented ADP crystals, the LID growth thresholds (the damage density growth thresholds) of Z-Py and Z-Pr are 6.0 J/cm2 and 12.4 J/cm2, respectively, as listed in Table 2.
Table 2. LID growth threshold (density) of ADP crystals with different growth methods.
3.1.3 Damage morphology
The morphologies of LID sites of ADP crystals irradiated by 355 nm pulsed laser were achieved from the direction perpendicular to the laser transmission by a microscope, and the typical morphologies of bulk damage sites were observed in Fig. 6.
As Fig. 6(a) displays, the damage points of C sample are relatively scattered, with the sizes in tens of microns. The appearance includes dot shape or shock wave-like interface extending around the central cavity, etc., indicating the changeless types and small quantity of defects serves as precursors in C sample. The LID site morphologies of D-Py and Z-Py are analogical, with dot shape, filament shape, shock wave-like interfaces extending around the central cavity. The filament shape damage morphology reflects the intrinsic damage caused by nonlinear effects such as self-focusing effect in ADP crystal under laser irradiation, basically crossing samples. There are defects in prismatic region that modify refractive index locally and therefore increases the probability of nonlinear interaction and cause a fraction of beam to collapse into a filament. Therefore, the filament shape damage morphology is excluded of the typical damage sites morphologies induced by precursors. In addition, the typical morphologies of damage sites of D-Pr and Z-Pr shown in Fig. 6(c) and 6(e) are dot shape, line shape (Exclusions are made as described above), shock wave-like interfaces and irregular cracks extending around the central cavity, etc. These results indicate many types of defects exist in the prismatic region of rapid growth ADP crystals, including the low, medium and high damage threshold defects.
3.2 Experiment results
3.2.1 Spectrums measurements
The spectrums measurements of ADP crystals with different growth methods are shown in Fig. 7. As illustrated by the UV-visible transmission spectrum in Fig. 7(a), C-ADP exhibits the highest transmittance in the UV region, indicating the highest growth quality and the smallest quantity of absorption defects in the bulk. Besides, the UV transmittance of D-ADP is slightly superior to that of Z-ADP, deducing that the quality of ADP crystal grown in defined direction slightly precedes over that of ADP crystal grown in Z-direction. The UV absorption spectrums in Fig. 7(b) of ADP crystals with different growth methods are in discrepancy. Separately, the absorptions of D-Pr and Z-Pr greatly differ from those of D-Py and Z-Py, leading to the immense divergence of 1-on-1 LIDTs between the prismatic and pyramidal sectors in rapid growth ADP crystals. In addition, the absorption of the D-ADP is slightly lower than that of Z-ADP, meeting well with the truth that the LIDT of directional ADP crystal somewhat exceeds that of Z-direction ADP crystal. In the range from 240 nm to 280 nm of the absorption spectrum for D-Pr and Z-Pr, a wide absorption peak appears, induced by impurity ions such as Fe3+, Cr3+ etc [28–31]. Moreover, the R-on-1 LIDTs in the prismatic and pyramidal regions of rapid growth ADP crystals are approximately the same, demonstrating laser conditioning can effectively modify or reduce the UV absorption defects in the prismatic sector of ADP crystals. The fluorescence excitation spectrum of ADP crystals at 330 nm is shown in Fig. 7(c) and All samples provide the similar fluorescence excitation spectra with the discrepancies of about ± 1 nm. Therefore, the strongest peak points to 223 nm, selected as the excitation peak for fluorescence spectrum measurement. The emission spectrum ranges from 290 nm to 750 nm, with a scanning rate of about 500 nm/min. In our experiments, over 10 fluorescence spectra are randomly tested for each crystal, and then the average value is taken to obtain the final fluorescence spectrum. The fluorescence emission spectra of ADP crystals with different growth methods are shown in Fig. 7(d). As none literature give out the detailed classification and attribution of fluorescence peaks of ADP crystals, we have not carried out the process of peak-fitting for the fluorescence emission spectra, while the quantity and distribution of defects in ADP crystals can be reflected by the overall fluorescence intensity integrated from Fig. 7(d). In general, the relationship between the two non-parametric variables can be determined by the correlation analysis. Therefore, the correlation coefficient between fluorescence area and LIDT can be quantitatively evaluated by Spearman rank correlation function [32], defined as:
Where di is the difference of rank corresponding to each pair of horizontal and vertical coordinate data pairs, and n is the number of horizontal and vertical coordinate data pairs. The Spearman rank correlation function of the correlation between the overall fluorescence intensity and the LIDT are shown in Table 3. In Table 3, the correlation coefficient r(s) equals to −1, indicating the completely negative correlation between the overall fluorescence intensity and the LIDT. It is further proved that the quantity of defects in C-ADP is far below that of rapid growth ADP crystals and the LIDT is significantly influenced by the precursors in the bulk.
Fig. 7. Spectrums measurements of ADP crystals with different growth methods. (a) Ultraviolet-visible (UV) transmission spectrum; (b) Ultraviolet absorption spectrum; (c) Fluorescence excitation spectrum (All samples provide the similar spectra with the discrepancies of about ± 1 nm); (d) Fluorescence emission spectrum.
Table 3. Overall fluorescence intensity of ADP crystals with different growth methods.
Positron annihilation lifetime spectra of ADP crystals with different growth methods are listed in Table 4, including the lifetime values τ1, τ2 and τ3, corresponding annihilation intensities I1, I2, and I3 and the average annihilation lifetime of positrons τm. According to positron spectroscopy [33], τ1 indicates the existence of point defects such as monovacancy etc., representing the weighted average of positron annihilation lifetime in the perfect region of the crystal and annihilation lifetime after being trapped at the point defects. And I1 represents the content or concentration of point defects in the crystals. What is more, τ2 stands for the weighted average of the positron capture annihilation lifetime and the short lifetime component of the source support where the electron concentration fluctuates greatly, indicating the possessions of defects such as vacancy clusters, dislocations, microcracks in the bulk of ADP crystals. And I2 represents the content or concentration of the microscopic defects in ADP crystals. Moreover, τ3 acts for the weighted average of positron trapping annihilation lifetime and source carrier lifetime at the large-size defects such as holes and cracks in large size, and I3 represents the content or concentration of the large-size defects in the crystal. considering the influence of positron source and surface processing and the low value of I3, τ3 is not taken into consideration. τm is the average annihilation lifetime of positrons, and linearly related to the negative charge density, representing the overall distribution of vacancy defects in ADP crystals. As described in the positron annihilation lifetime spectrum, the values of τ1, I1, τ2 and I2 of C-ADP are all inferior to the lifetime and intensity values of D-ADP. The truth demonstrates larger quantity and density of defects such as monovacancy, vacancy clusters, dislocations and microcracks etc. exist in rapid growth ADP crystals. Further considering, the lifetime and intensity values are in positive correlation with LIDTs by 1-on-1 and R-on-1 modes, proving that the defects such as monovacancy, vacancy clusters, dislocations and microcracks etc. are precursors for the initial laser induced damage. Therefore, from the perspective of defects such as vacancies, it is necessary to further improve the growth technique and improve the quality of rapid growth ADP crystals.
Table 4. Positron annihilation lifetime spectra of ADP crystals with different growth methods.
3.2.2 In-situ scattering measurements
In-situ scattering measurements of ADP crystals grown in different methods were carried out and the typical images of scattering sites related to the initial LID were obtained as shown in Fig. 8. As Fig. 8(a) and 8(b) illustrated, many fine tubular scattering sites existed in initial morphology of C-ADP. Then, the sample was irradiated by 355 nm laser in R-on-1 mode with a gradient of 2.8 J/cm2, 4.2 J/cm2, 5.6 J/cm2, 7.2 J/cm2, 8.9 J/cm2, 10.1 J/cm2, 11.8 J/cm2, 13.4 J/cm2 and 14.5 J/cm2. With the laser fluence arriving at 14.5 J/cm2, LID appeared at the initial defect points in Fig. 8(a), indicating the fine tubular scattering sites were the precursors inducing the initial damage. However, in Fig. 8(b), as the laser with 14.5 J/cm2 irradiated, there was no damage point at the initial defects, while the damage point was induced in the area with none initial defects. This phenomenon demonstrates another type of defect serves as the precursors, which cannot be detected via laser scattering. Through statistical analysis of in-situ scattering of C-ADP, the correlation degree between the tiny tubular scattering sites and initial laser-induced damage is about 10%.
Fig. 8. Typical morphology changes of in situ scattering of ADP crystals with directional seed growth. (a) Scattering sites are related to initial damage in C-ADP; (b) Scattering sites are irrelated to initial damage in C-ADP; (c) Scattering sites are irrelated to initial damage in D-Py; (d) Scattering sites are irrelated to initial damage in D-Pr; (e) Scattering sites are irrelated to initial damage in Z-Py; (f) Scattering sites are irrelated to initial damage in Z-Pr. (The white circle reflects the changes of the typical scattering sites.)
In the in-situ scattering test, Fig. 8(c) and 8(d) are typical images of D-Py and D-Pr. with a laser fluence gradient of 1.1 J/cm2, 2.0 J/cm2, 3.3 J/cm2, 4.8 J/cm2, 6.5 J/cm2, 8.2 J/cm2, 10.0 J/cm2 and 12.7 J/cm2. When the laser fluence was 1.1 J/cm2, laser conditioning effect made the initial fine tubular scattering sites disappear temporarily. As the laser fluence reached to 12.7 J/cm2, none damage points came out at the fine tubular scattering sites, while some damage points at other areas without initial scattering sites appeared, showing the fine tubular scattering sites were not the damage precursors with the lowest LIDT in the pyramidal sector as shown in Fig. 8(c). In the prismatic region, as the laser fluence arrived at 10.0 J/cm2, the initial damage appeared at the areas without fine tubular scattering sites. Therefore, the fine tubular scattering sites are not the precursors with the lowest LIDTs in defined growth ADP crystals. Similarly, the in-situ scattering measurements for Z-directional ADP crystals as shown in Fig. 8(e) and Fig. 8(f) illustrate the truth that the fine tubular scattering sites are also not the precursors with the lowest LIDTs in ADP crystals grown in Z direction.
3.3 Discussions
The laser fluence to induce the initial bulk damage point depends on the precursors with the lowest LIDT in the crystal. Under intense laser irradiation in 1-on-1 method, all the precursors with the lowest LIDTs are “detonated”, resulting in the occurrence of initial damage. While in R-on-1 method, the effect of laser conditioning exists and cannot be neglected. In this case, many precursors with the lowest LIDTs are eliminated or modified to be the precursors with higher LIDTs compared with the initial states, and the initial LID are induced at the positions of other precursors with high LIDTs or the changed precursors. After the initial damage sites come out, as the laser with the same or lower fluence irradiates, none new damage sites appear in the crystal due to the detonation of the defects and the damage density cannot increase. When the laser irradiation fluence is increased and arrived at the threshold induced by other precursors with high LIDTs in the crystal, new damage points begin to appear and the damage density increases. Once one damage point occurs, microscopic cracks may appear around it, and the cracks cleave along the direction with the chemical bonds with weak bonding force in the crystal. Then the weak stress field propagates to the surrounding crystal lattice. If the other chemical bonds are not destroyed in the stress field, the cracks caused by the damage points will not continue extending around. As the laser resumes irradiating the initial damage points, the size of the initial damage points will not increase until the laser fluence reaches the higher laser induced damage threshold of other defects in the crystal, because the surrounding crystal lattice is basically intact and few defects with low LIDT remain. Therefore, the damage growth properties of ADP crystal actually reflect the process of inducing the higher-threshold damage points. In this process, the size of the damage points with low LIDT basically does not change, while the damage density in the crystal increases after the high-threshold damage point occurs. The damage growth properties are irrelated to the “sampling” LIDT, i.e. the initial damage, but related to the distribution of intrinsic defects in ADP crystals. Different types and densities of defects present different intrinsic LIDTs and different corresponding damage growth thresholds. Besides, laser conditioning can greatly affect the damage growth thresholds
As the initial LID properties in section 3.1 present, the overall defect distribution and growth quality of ADP crystals grown by different methods can be evaluated from the LIDTs by 1-on-1 and R-on-1 methods. The ADP crystal grown by conventional method displays the best quality, followed by the directional ADP crystal and the ADP crystal grown in Z direction. Besides, the types, quantity and density of defects in the prismatic and pyramidal regions of rapid grown ADP crystals are quite different, demonstrating that laser conditioning suppresses the effects of all lower energy level defects. It is truth that as the laser fluence is inferior to the critical value when the initial damage occurs, none new damage points crop up in the bulk and the sizes of initial damage sites will not increase in ADP crystals. However, when the laser fluence exceeds another critical value named as the LID growth threshold or the damage density growth threshold, new damage points appear in the bulk, and the sizes of initial damage sites do not increase. As results show in section 3.1, the LIDTs by 1-on-1 and R-on-1 methods of C-ADP are 7.7 J/cm2 and 16.0 J/cm2, respectively, with the LID growth threshold of 14.1 J/cm2; The LIDTs of pyramidal and prismatic sectors via 1-on-1 and R-on-1 methods of directional ADP crystals are 5.0 J/cm2, 1.9 J/cm2 and 11.8 J/cm2, 11.8 J/cm2, respectively, and the LID growth thresholds are 13.1 J/cm2 and 6.1 J/cm2, separately; In addition, the LIDTs of pyramidal and prismatic sectors via 1-on-1 and R-on-1 methods of Z-seed growth ADP crystals are 4.9 J/cm2, 1.7 J/cm2 and 10.8 J/cm2, 10.8 J/cm2, respectively, and the LID growth thresholds are separate 12.4 J/cm2 and 6.0 J/cm2. In a conclusion, these results demonstrate three types of defects associated with their initial LIDTs exist in the bulk of ADP crystals, namely the L defects with low LIDTs ranging from 1 J/cm2 to 5 J/cm2, the M defects with medium LIDTs from 6 J/cm2 to 12 J/cm2 and the H defects with high LIDTs greater than 14 J/cm2. Therefore, M defects and H defects are distributed in C-ADP, while all three kinds of defects exist in Z-directional and defined-directional ADP crystals and the distribution of the three defects in prismatic and pyramidal growth regions is in discrepancy. In addition, laser conditioning can obviously modify or eliminate the L defects and M, resulting in the great increase of LIDT of C-ADP to about 16.0J/cm2 by R-on-1 mode, and the enormous improvement in LIDTs of prismatic and pyramidal sectors in rapid growth ADP crystals.
The damage sites initial from precursors in KDP and DKDP crystals are mostly characterized by multi-directional cracks extending around the central cavity [34–36], compared with the LID morphologies of DKDP crystals, while those of ADP crystals hardly show characteristics of multi-directional obvious cracks extending around the central cavity. This phenomenon is closely related to the differences in chemical bond composition between DKDP and ADP crystals. The binding energies of the three types of chemical bonds in DKDP crystal are quite different with the weakest γ-bond [18,20]. Thus, fractures and LID may come out firstly according to the breaking down of γ-bond, resulting in the damage morphology of multi-directional obvious cracks extending around the central cavity. However, the differences among binding energies of the three types of chemical bonds all related to hydrogen bonds [20] in ADP crystals are not very significant. During the laser irradiation, simultaneous fractures appear in the three kinds of bonds in the precursors. The laser fluence absorbed by the precursors during damage initiation may be the reason for the different damage morphologies. As the energy absorbed, plasma is then generated and subsequent additional laser energy absorption appears. Energy transfer and accumulation from plasma to its surrounding lattice results in heating of the surrounding lattice, leading to the stress generation and accumulation with melting of affected areas [37]. The power of shock waves due to different absorptions result in different morphologies of damage sites in ADP crystals.
The overall quality of ADP crystal and the LIDTs can be preliminarily judged by the spectrum measurements such as the ultraviolet-visible transmittance spectrum, ultraviolet absorption spectrum, fluorescence spectrum and positron annihilation lifetime spectra etc. in our essay. These measurements indicate the different quantities and distributions of L, M and H defects in ADP crystals with different growth methods. In ultraviolet absorption spectrum, the wide absorption peak induced by impurity ions such as Fe3+, Cr3+ etc [28–31]. demonstrates L defects may include the impurity ions. For ADP crystals, the main units during the growth are NH4+and H2PO4- and the distributions of the two units in the prismatic (100) and pyramidal (101) faces are in discrepancy [18,20,38]. As shown in Fig. 9, the structure of the (101) plane of ADP crystal is formed by alternately double-layer NH4+ and double-layer H2PO4-, covered with positive charge. Based on this, the cations such as Fe3+, Cr3+ etc. are hardly incorporated into the (101) plane evolving to the pyramidal sectors [39]. The (100) plane is formed by alternately arranging positive ions NH4+ and negative ions H2PO4-, and the -OH in H2PO4- groups is always exposed on the surface, resulting in the negative charge in (100) plane. Therefore, the cations such as Fe3+, Cr3+ etc. are easily incorporated in the (100) plane evolving into the prismatic sectors [39]. Besides, as positron annihilation lifetime spectra show, the monovacancy, vacancy clusters, dislocations, microcracks et al. all are contained in the L, M and H defects, the quantity and distribution of which affect and even determine the LIDTs of ADP crystals. In-situ scattering tests indicate that only the fine tubular scattering sites exists in ADP crystal grown by different methods. This kind of scattering sites only has a 10% probability of being the precursor with lowest LIDTs in C-ADP. While in rapid growth ADP crystals, it is not the precursor with lowest LIDTs. This phenomenon points out that L defects in ADP crystals may not be detected by the scattering technique in our system and M and H defects are belonging to the fine tubular scattering sites. Therefore, according to the growth threshold, L, M and H defects are collections of many types of defects.
Compared with DKDP crystals in our group with the same conditions such as the growth conditions, the processing and polishing method and the LID testing systems, ADP crystals have higher LIDTs [22,26,40,41]. From a microscopic perspective, the three chemical bonds such as the hydrogen bonds in DKDP crystals are different from those in ADP crystals. The hydrogen bonds are prone to fracture and LID under laser irradiation. Therefore, the damage morphology in DKDP crystal is mostly multi-directional cracks extending around the central cavity, as the extension direction of the cracks is consistent with the hydrogen bonding directions. However, the three kinds of chemical bonds in ADP crystals are all related to hydrogen bonds, with fewer discrepancy, leading to the stronger resistance of the hydrogen bond network. When the laser fluence exceeds the threshold, initial damage points are induced at the defects, and fractures occur around the initial damage points, with the morphology of shock wave-like interfaces or irregular cracks extending around the central cavity. On the other hand, François Guillet et al [42]. found DKDP crystals undergo a phase transformation at about 140°C via the on-line XRD. Under laser irradiation, the interaction between photons and phonons causes thermal vibration and heats up the surrounding crystal lattices. If the temperature in this region reaches the phase transformation temperature of DKDP crystal, the cell parameters in this region will change and a distortion of the crystal lattice occurs, forming a stress field and then resulting in LID. A. Abdel-Kader et al [43]. obtained the decomposition temperature and phase transition temperature of ADP crystal by TG, DTG, DTA and DSC measurements. ADP crystal undergoes phase transition at 175°C, and decomposes into orthophosphoric acid at 210°C with the release of ammonia. The orthophosphoric acid is converted into pyrophosphoric acid and then into metaphosphoric acid. The phase transition and thermal decomposition of ADP crystal were confirmed by X-ray diffraction and infrared absorption spectra [43]. Therefore, the higher phase transition temperature of ADP crystals elucidates one reason for the higher LIDT than that of DKDP crystals.
4. Conclusions
High-quality ammonium dihydrogen phosphate (NH4H2PO4, ADP) crystals were grown in Z direction and in defined crystallographic direction (θ=90°, φ=45°) via the rapid growth method. Mechanism of defect-induced damage behavior in 355 nm of three types of ADP samples cutting in type-II matching and third harmonic generation direction from the as-grown crystals were investigated, including the initial LID characteristics and the physical and chemical properties of defects which serve as the damage precursors. The evaluations of damage behaviors include the “sampling” LIDT by 1-on-1 and R-on-1 methods, bulk damage growth and bulk damage morphology. Three types of defects associated with their initial LIDTs existed in the bulk of ADP crystals, namely the L defects with low LIDTs ranging from 1 J/cm2 to 5 J/cm2, the M defects with medium LIDTs from 6 J/cm2 to 12 J/cm2 and the H defects with high LIDTs greater than 14 J/cm2. UV-visible transmittance spectrum, ultraviolet absorption spectrum, fluorescence spectrum, positron annihilation spectrum and the online light scattering measurements were carried out to investigate the defect-induced damage behavior in ADP crystals, indicating that L, M and H defects were collections of many types of defects according to the growth threshold. The study can provide a reference for the investigations on LID properties of ADP crystals in short wavelength.
Funding
Chinese Academy of Sciences (XDB1603); National Natural Science Foundation of China (11874369, U1831211).
Disclosures
The authors declare no conflicts of interest.
References
1. C. Belouet, “Growth and Characterization of Single-Crystals of Kdp Family,” Prog. Cryst. Growth Charact. 3(2-3), 121–156 (1980). [CrossRef]
2. Z. D. Li, X. J. Huang, D. X. Wu, and K. M. Xiong, “Large crystal growth and measurement of electro-optical coefficients of ADP,” J. Cryst. Growth 222(3), 524–527 (2001). [CrossRef]
3. G. Li, X. Liping, G. Su, X. Zhuang, Z. Li, and Y. He, “Study on the growth and characterization of KDP-type crystals,” J. Cryst. Growth 274(3-4), 555–562 (2005). [CrossRef]
4. S. Chaki, M. P. Deshpande, J. P. Tailor, M. D. Chaudhary, and K. Mahato, “Growth and Characterization of ADP Single Crystal,” AJCMP 2(1), 22–26 (2012). [CrossRef]
5. S. Ji, F. Wang, L. Zhu, X. Xu, Z. Wang, and X. Sun, “Non-critical phase-matching fourth harmonic generation of a 1053-nm laser in an ADP crystal,” Sci. Rep. 3(1), 1605 (2013). [CrossRef]
6. A. Ashkin, G. D. Boyd, and J. M. Dziedzic, “Observation of Continuous Optical Harmonic Generation with Gas Masers,” Phys. Rev. Lett. 11(1), 14–17 (1963). [CrossRef]
7. G. Duchateau, M. D. Feit, and S. G. Demos, “Transient material properties during defect-assisted laser breakdown in deuterated potassium dihydrogen phosphate crystals,” J. Appl. Phys. 115(10), 103506 (2014). [CrossRef]
8. G. Duchateau, “Simple models for laser-induced damage and conditioning of potassium dihydrogen phosphate crystals by nanosecond pulses,” Opt. Express 17(13), 10434–10456 (2009). [CrossRef]
9. C. W. Carr, H. B. Radousky, and S. G. Demos, “Wavelength dependence of laser-induced damage: determining the damage initiation mechanisms,” Phys. Rev. Lett. 91(12), 127402 (2003). [CrossRef]
10. G. Duchateau, G. Geoffroy, A. Dyan, H. Piombini, and S. Guizard, “Electron-hole dynamics in normal and deuterated KH2PO4illuminated by intense femtosecond laser pulses,” Phys. Rev. B 83(7), 075114 (2011). [CrossRef]
11. M. Pommiès, D. Damiani, X. Le Borgne, C. Dujardin, A. Surmin, J. C. Birolleau, F. Pilon, B. Bertussi, and H. Piombini, “Impurities detection by optical techniques in KH2PO4 crystals,” Opt. Commun. 275(2), 372–378 (2007). [CrossRef]
12. M. Pommiès, D. Damiani, B. Bertussi, J. Capoulade, H. Piombini, J. Y. Natoli, and H. Mathis, “Detection and characterization of absorption heterogeneities in KH2PO4 crystals,” Opt. Commun. 267(1), 154–161 (2006). [CrossRef]
13. Z. M. Liao, M. L. Spaeth, K. Manes, J. J. Adams, and C. W. Carr, “Predicting laser-induced bulk damage and conditioning for deuterated potassium dihydrogen phosphate crystals using an absorption distribution model,” Opt. Lett. 35(15), 2538–2540 (2010). [CrossRef]
14. A. K. Burnham, M. Runkel, M. D. Feit, A. M. Rubenchik, R. L. Floyd, T. A. Land, W. J. Siekhaus, and R. A. Hawley-Fedder, “Laser-induced damage in deuterated potassium dihydrogen phosphate,” Appl. Opt. 42(27), 5483–5495 (2003). [CrossRef]
15. Z. L. Wu, M. Reichling, E. Welsch, D. Schafer, Z. X. Fan, and E. Matthias, “Defect Characterization for Thin-Films through Thermal Wave Detection,” Proc. SPIE 1624, 271–281 (1992). [CrossRef]
16. S. G. Demos and M. Staggs, “Application of fluorescence microscopy for noninvasive detection of surface contamination and precursors to laser-induced damage,” Appl. Opt. 41(10), 1977–1983 (2002). [CrossRef]
17. K. Tsuchida, R. Abe, and M. Naito, “Electron spin resonance of γ-irradiated KH2PO4,” J. Phys. Soc. Jpn. 35(3), 806–809 (1973). [CrossRef]
18. X. Ren, C. Sun, K. Li, D. Xu, and D. Xue, “Chemical Bonding and Single Crystal Growth of ADP Crystals,” Mater. Focus 2(4), 309–315 (2013). [CrossRef]
19. C. Sun and D. Xue, “Crystallization behaviors of KDP and ADP,” Opt. Mater. 36(12), 1966–1969 (2014). [CrossRef]
20. D. Xu and D. Xue, “Chemical bond analysis of the crystal growth of KDP and ADP,” J. Cryst. Growth 286(1), 108–113 (2006). [CrossRef]
21. T. M. Christmas and J. M. Ley, “Laser-Induced Damage in Xdp Materials,” Electron. Lett. 7(18), 544 (1971). [CrossRef]
22. Y. Lian, L. Zhu, T. Sui, G. Yu, L. Zhang, H. Zhou, M. Xu, and X. Sun, “The rapid growth of ADP single crystal,” CrystEngComm 18(39), 7530–7536 (2016). [CrossRef]
23. Y. Lian, M. Xu, L. Zhang, D. Cai, T. Sui, X. Sun, and X. Chai, “Rapid growth of ADP crystal in a defined crystallographic direction,” CrystEngComm 20(7), 917–923 (2018). [CrossRef]
24. P. Rajesh, P. Ramasamy, and G. Bhagavannarayana, “Growth of ADP–KDP mixed crystal and its optical, mechanical, dielectric, piezoelectric and laser damage threshold studies,” J. Cryst. Growth 362, 338–342 (2013). [CrossRef]
25. Y. Lian, G. Yu, F. Liu, L. Zhang, M. Xu, H. Zhou, X. Sun, and Q. Gu, “Prismatic morphology of an ADP crystal grown in a defined crystallographic direction,” RSC Adv. 7(83), 52962–52969 (2017). [CrossRef]
26. X. Feng, L. Zhu, F. Li, F. Wang, W. Han, Z. Wang, Q. Zhu, and X. Sun, “Growth and highly efficient third harmonic generation of ammonium dihydrogen phosphate crystals,” RSC Adv. 6(40), 33983–33989 (2016). [CrossRef]
27. J. Huang, Z. Wu, F. Wang, H. Liu, L. Sun, X. Zhou, X. Ye, Q. Deng, X. Jiang, W. Zheng, and D. Sun, “Initial Damage and Damage Growth of KDP Crystals Induced by 355 nm Pulse Laser,” Cryst. Res. Technol. 53(3), 1700269 (2018). [CrossRef]
28. L. Rashkovich and B. Y. Shekunov, “Influence of impurities on growth kinetics and morphology of prismatic faces of ADP and KDP crystals,” in Growth of Crystals (Springer, 1992), pp. 107–119.
29. I. Ogorodnikov, V. Y. Yakovlev, B. Shul’gin, and M. Satybaldieva, “Transient optical absorption of hole polarons in ADP (NH 4 H 2 PO 4) and KDP (KH 2 PO 4) crystals,” Phys. Solid State 44(5), 880–887 (2002). [CrossRef]
30. Y. Asakuma, Q. Li, H. M. Ang, M. Tade, K. Maeda, and K. Fukui, “A study of growth mechanism of KDP and ADP crystals by means of quantum chemistry,” Appl. Surf. Sci. 254(15), 4524–4530 (2008). [CrossRef]
31. X. Chai, Q. Zhu, B. Feng, F. Li, X. Feng, F. Wang, W. Han, and L. Wang, “Nonlinear absorption properties of DKDP crystal at 263 nm and 351 nm,” Opt. Mater. 64, 262–267 (2017). [CrossRef]
32. W. Hoeffding, “A Class of Statistics with Asymptotically Normal Distribution,” Ann. Math. Stat. 19(3), 293–325 (1948). [CrossRef]
33. P. G. Coleman, Positron Spectroscopy (Wiley-VCH Verlag GmbH & Co. KGaA, 2009).
34. C. W. Carr, M. D. Feit, J. Muyco, and A. M. Rubenchik, “Effect on scattering of complex morphology of DKDP bulk damage sites,” in Laser-Induced Damage in Optical Materials: 2004, (International Society for Optics and Photonics, 2005), 532–539.
35. C. Carr, M. Feit, M. Johnson, and A. Rubenchik, “Complex morphology of laser-induced bulk damage in K2H(2− x)DxPO4 crystals,” Appl. Phys. Lett. 89(13), 131901 (2006). [CrossRef]
36. S. Reyné, G. Duchateau, L. Hallo, J.-Y. Natoli, and L. Lamaignère, “Multi-wavelength study of nanosecond laser-induced bulk damage morphology in KDP crystals,” Appl. Phys. A 119(4), 1317–1326 (2015). [CrossRef]
37. P. Demange, R. A. Negres, C. W. Carr, H. B. Radousky, and S. G. Demos, “Laser-induced defect reactions governing damage initiation in DKDP crystals,” Opt. Express 14(12), 5313–5328 (2006). [CrossRef]
38. C. Sun and D. Xue, “In Situ IR Spectral Observation of NH4H2PO4 Crystallization: Structural Identification of Nucleation and Crystal Growth,” J. Phys. Chem. C 117(37), 19146–19153 (2013). [CrossRef]
39. B. Wang, C.-S. Fang, S.-L. Wang, X. Sun, and Y.-P. Li, “Researches on the growth habit and optical properties of Fe3+ ion doped KDP crystal.”
40. B. Liu, G. Hu, Y. Zhao, M. Xu, S. Ji, L. Zhu, L. Zhang, S. Xun, Z. Wang, and X. Xu, “Laser induced damage of DKDP crystals with different deuterated degrees,” Opt. Laser Technol. 45, 469–472 (2013). [CrossRef]
41. M. Xu, X. S. Z. Wang, X. Cheng, S. Sun, L. Ji, Y. Zhao, B. Liu, G. Hui, and X. Xu, “Laser-induced damage of DKDP crystal under different wavelengths,” Cryst. Res. Technol. 45(7), 763–766 (2010). [CrossRef]
42. F. Guillet, B. Bertussi, L. Lamaignère, and C. Maunier, “Effects of thermal annealing on KDP and DKDP on laser damage resistance at 3w,” in Laser-Induced Damage in Optical Materials: 2010, (International Society for Optics and Photonics, 2010), 78421T.
43. A. Abdel-Kader, A. Ammar, and S. Saleh, “Thermal behaviour of ammonium dihydrogen phosphate crystals in the temperature range 25-600° C,” Thermochim. Acta 176, 293–304 (1991). [CrossRef]
