The reflectivity of Al/Zr multilayers is enhanced by the use of a novel structure. The Al layers are divided by insertion of Si layers. In addition, Si barrier layers are inserted at the Al/Zr interfaces (Zr-on-Al and Al-on-Zr). As a result, crystallization of the Al layer is inhibited and that of Zr is enhanced. In grazing incidence x-ray reflectometry, x-ray diffraction, and extreme ultraviolet measurements, the novel multilayers exhibit lower interfacial roughness compared with traditional multilayer structures, and their reflectivity is increased from 48.2% to 50.0% at a 5° angle of incidence. These novel multilayers also have potential applications in other multilayer systems and the semiconductor industry.
© 2013 OSA
The multilayer systems used in extreme ultraviolet (EUV) applications are alternately layered structures consisting of two materials of different scattering powers. In particular, Al-based multilayers have potential applications in the construction of mirrors for instruments to detect solar coronal or transition-region emission lines in the wavelength region of 17–19 nm, since a number of Al-based multilayer combinations (such as Al/Zr systems) have significant reflectivity in this region [1–8].
In order to obtain the highest optical performance in Al/Zr systems, it is necessary to eliminate, as far as possible, any factors that can lead to loss of reflectivity. On comparing the theoretical reflectivity with experimental data for Al/Zr multilayers , it is found that four factors are responsible for reduced reflectivity––inhomogeneous crystallization of Al, interdiffusion between Al and Zr layers, the presence of an oxidized surface layer, and contamination. There are a variety of approaches that can be adopted to reduce the influence of these factors. For example, inserting a third material [4,9] or a buffer layer  into the multilayer structure can inhibit crystallization of the metal layer and lower the interfacial roughness. Using thermal treatments [11,12] and a barrier layer [13,14] can smooth the interfacial boundary and prevent interdiffusion at the interfaces. The simplest approach is to focus on the multilayer structure.
In the present work, we fabricate a multilayer structure to reduce crystallization of the Al layer and interdiffusion at the Al/Zr interfaces (Zr-on-Al and Al-on-Zr), and thus improve reflectivity. From previous studies [1,8] it is known that the Al layer is not highly oriented in the Al<111> phase when the thickness is kept below 3.0 nm, and the presence of Si in the Al/Zr system not only inhibits crystallization of the Al layer but also enhances crystallization of the Zr layer. Based on these considerations, we have devised a novel multilayer structure in which Si layers are inserted within the Al layer to influence the growth of the layer and Si barrier layers are inserted at the Al and Zr interfaces to smooth the interfacial boundary.
We describe our novel multilayer structure in Section 2 and our experimental procedure in Section 3. In Section 4.1, we use grazing incidence x-ray reflectometry (GIXR) to compare the different multilayer types (traditional Al/Zr multilayers and novel Al/Zr multilayers with and without Si barrier layers, with different Al layer thicknesses). In Section 4.2, we use near-normal-incidence EUV reflectance measurements to investigate the enhancement of the reflectivity of Al/Zr with Si barrier layers. In Section 4.3, we further characterize the performance of the novel multilayer structure by x-ray diffraction (XRD). In Section 4.4, we discuss the results of these investigations as well as the possibility of extending the use of this novel multilayer structure to other applications. Finally, we conclude in Section 5 with comments regarding the performance of novel Al/Zr multilayers.
2. The novel multilayer structure
It is known from previous work that the variable interfacial and surface roughness in Al/Zr multilayers are mainly caused by inhomogeneous crystallization of Al [1–3]. In order to prevent crystallization of Al, the Al layer in our design is itself divided into a number of thin layers along the out-of-plane direction by the insertion of Si layers. In addition, Si barrier layers can be inserted between the Al and Zr layers in an attempt to smooth the boundaries at the interfaces. Therefore, we designed two kinds of structures to compare the performances of the novel multilayers with and without the Si barrier layers. Schematic diagrams of these structures are shown in Fig. 1.
The advantages of these designs over traditional multilayer structures are as follows:
- ● Prevention of Al crystallization—Doping of the Al layers with Si helps to prevent crystallization of the Al. In Al/Zr multilayers in the previous study, 1 wt.% Si has been added, although it is possible that 1.5 wt.% would have a stronger effect . Thus, more Si layers could be inserted between the thin Al layers, which would not only keep the thin Al layer thickness below the critical value, but also would maintain the total thickness of the whole Al layer (thin Al layers plus Si layers) at around 6.5 nm. As a result of inter-diffusion between the Al and Si layers, the Si could penetrate into the Al crystal lattice with the formation of a novel Al–Si alloy, and this could also help to inhibit crystallization of the Al layer . The number of thin Al layers will vary depending on the intended practical application, and the number of Si layers will also vary accordingly. For example, in Fig. 1(b), if the Al layer is divided into x thin layers, there will be x−1 Si layers.
- ● Smoothing of the interfaces between the Al and Zr layers—There is considerable interdiffusion among the Si, Al, and Zr layers. We have found that Si not only inhibits the crystallization of Al but also enhances that of Zr, which helps to provide a smooth interface between the layers . The function of the Si barrier layer is thus not to prevent interaction between the Al and Zr layers but rather to smooth the interfaces between them.
- ● Maintenance of the optical contrast between the materials constituting the multilayer—The atomic numbers of Si and Al are similar (13 and 14, respectively), and when an Si layer is inserted between two Al thin layers it does not affect the electron standing-wave field, and so the optical constants of the Al are unchanged. However, it should be noted that the addition of too many Si layers will affect the optical constants of Al and thus alter the optical contrast between the Al and Zr layers.
- ● Improvement in lateral uniformity—If the thickness of the Al layers is less than a critical value (3.0 nm), they do not exhibit a high orientation . The Al<111> of face-centered cubic Al (Al-FCC) could not be observed in the Al layers, and there is no specific orientation of Al with respect to the direction perpendicular to the layers. Thus, the lateral uniformity of the Al layers in the multilayers is improved. The surface and interfacial roughness is also lowered.
Generally speaking, these properties of the novel multilayers result in enhanced reflectivity . In order to simplify the discussion in this paper each complex multilayer structure is represented by the thickness of the respective thin Al layers. For example, by “Al=0.6 without Si barrier layers” we denote the Al/Zr multilayer without Si barrier layers in which the Al layer is divided into eight thin layers of 0.6 nm thickness each, with seven Si layers of 0.4 nm thickness each inserted between each thin Al layer. Similarly, by “Al=0.6 with Si barrier layers” we denote the same multilayer except with the addition of Si barrier layers between the Al and Zr layers. The traditional Al/Zr multilayers with 40 periods are simply denoted by Al/Zr.
The Al/Zr multilayers were all prepared using a direct-current magnetron sputtering system [1–3]. The sputtering targets with diameters of 100 mm were Zr (99.5%) and Si-doped Al [which is represented by the symbol “Al” to replace the symbol “Al(1wt.%Si)” in this paper]. The base pressure was 4.0×10−5 Pa and the samples were deposited onto polished Si wafers under an atmosphere of 0.16 Pa Ar (99.9999% purity). In order to investigate the potential advantages of this novel multilayer system, we fabricated a number of different samples and used the symbol to replace each sample––the novel multilayer structure without Si barrier layers (the samples Al=0.6-S1, Al=1.4-S2, Al=2.8-S3, and Al=3.5-S4), the novel multilayer structure with Si barrier layers (the samples Al=0.6-S5 and Al=2.8-S6), and traditional Al/Zr multilayers (Al/Zr-S7). For the samples S1 and S5, the x value is 8, which means that the Al layer is divided into 8 thin layers, with 7 thin Si layers inserted between them. The whole Al layer structure can thus be described as [Al/Si]7/Al. The x values of the samples S2, S3, S4, and S6 are 4, 1, 1, and 1, respectively. The symbols and periodic thicknesses of the samples are shown in Table 1.
To characterize the interfacial structure, GIXR was performed using a Cu Kα source (λ=0.154 nm), and the data fitting was done with Bede Refs software (a genetic algorithm) . The XRD measurements identified crystalline phases present in the modified layer along with structural changes. EUV reflectivity measurements were made at a 5° angle of incidence in the wavelength region from 16.5 to 20.5 nm at the ELETTR Synchrotron Light Laboratory (Spectromicroscopy Beamline 3.2L) in Italy.
4. Results and discussions
4.1 Grazing incidence x-ray reflectometry
To illustrate the advantages of the novel multilayer structure, all samples were characterized by GIXR measured over the angular range θ=0°–3°. Because of the different amounts of Si penetrating into the Al layers, the GIXR data were fitted by different models, which are shown in Table 2. Examples of GIXR spectra and fitted curves for the samples of S5, S6, and S7 are shown in Fig. 2.
From the fitting parameters in Tables 2–4 (not all fitting data are presented in Fig. 2), it can be seen that all the models represent a complete period, with the periodic thicknesses ranging from 9.5 to 11.9 nm. In Tables 2 and 3, the complete Al layer contains different numbers of thin Al layers and inserted Si layers in different samples. For example, in sample S1, the whole Al layer ([Al/Si]7/Al) contains eight Al layers (0.6 nm) and seven Si layers (0.4 nm) (these values correspond to the intended thickness of the Al layer in the experiment). Because the thicknesses of Si and Al are too small, they could not easily be separated, and so we used an Al(42wt.%Si) layer to represent the whole Al layer in the fitting model. For sample S2, the thickness of the Al layer is 1.4 nm, which could be considered as an intact layer. Thus, we used the whole Al layer (four thin Al layers and three Si layers) in the fitting process. However, the Si material could not be considered as an intact layer in the experimental samples owing to the large amount of interdiffusion between Al and Si. An Si layer inserted into an Al layer can fully intermix with the layer owing to the small thickness (0.4 nm) of the Si layer and finally transform to an Al(51wt.%Si) layer.
Based on the different properties  and critical thicknesses  of the Al–Si alloy layers with variable proportions of Si, the Al layers with different Si contents can have different critical thicknesses. It is possible that at a thickness at which Al layers would have crystallized, the new Al(51wt.%Si) materials might still have an amorphous structure in the multilayers. Thus, the Al(51wt.%Si) alloy could limit the crystallization of the Al. If this is the case, then the model will differ from the original one of Fig. 1(a), as shown in Fig. 3. The Si layer inserted into the Al layers has transformed into an Al(51wt.%Si) layer. In addition, the Si barrier layers added at the Al-on-Zr and Zr-on-Al interfaces can interdiffuse with the Al and Zr layers and produce an Al–Zr–Si alloy layer at the interfaces.
From Table 2, the novel multilayers without Si barrier layers may exhibit interactions between the Al and Zr layers in which asymmetrical interlayers  Al-on-Zr and Zr-on-Al can appear between the Al and Zr layers. The Al and Si layers do not form intact layers in sample S1. Because of considerable interdiffusion between Al and Si, the Si penetrates the Al crystal lattices to form a novel alloy Al(42wt.%Si). The percentage by weight of Si is calculated from the quantity of Si in the whole Al layer. The roughnesses of the Zr, Zr-on-Al, Al(42wt.%Si), and Al-on-Zr layers are 0.90, 0.32, 0.88, and 0.28 nm, respectively. With increasing thickness of the Al layer in samples S2 and S3, the Al layer can form an intact layer in the multilayers.
An interesting feature of the GIXR measurements of S2 and S3 is that the curves of the two samples are similar (not shown in Fig. 2) and there is little difference between the roughness of different layers [Zr, Zr-on-Al, Al, Al(51wt.%Si), and Al-on-Zr]. When the Al layer thickness is above 3.0 nm in the sample Al=3.5 without Si barrier layers, the roughness of Al(51wt.%Si) and the two symmetrical interlayers Al-on-Zr and Zr-on-Al are almost the same as those in samples S2 and S3. However, the roughness of Zr and Al in samples S3 and S4 are increased from 0.70 to 0.92 nm and from 0.72 to 0.95 nm, respectively. From the results for the samples without Si barrier layers, we deduce that the roughness of different layers in the novel multilayers is influenced by the formation of Al- FCC. The Si can interact with Al and Zr layers, in particular, forming an Al(51wt.%Si) layer and preventing crystallization of the Al layer. The Al layer thickness in the novel multilayers is below the critical thickness (3.0 nm). In the absence of crystallization of Al-FCC, the surface and interfacial roughness is lower, and the lateral uniformity of the Al layer is also improved in the novel multilayers.
To examine the performance of Si barrier layers, we compare S1-S7 (in Tables 2-4). In Table 3, the novel multilayers with Si barrier layers can have symmetrical interfaces in which the Si barrier layers are about 0.4 nm thick. With Si barrier layers in S5, the roughness of Zr and Al(42wt.%Si) are 0.75 and 0.73 nm, respectively, which are lower than those (0.90 and 0.88 nm) in sample S1 (Table 2), implying that an Si barrier layer (Al–Zr–Si alloy) can smooth the interfacial boundary. Similarly, compared to S3, the roughness in the Zr and Al layers of S6 are lowered. We find that the Si barrier layers influence the formation of Zr-on-Al and Al-on-Zr interlayers, which can reduce the interfacial roughness (Table 2).
It can be seen from Fig. 2 that the novel multilayers S5 and S6 show significantly enhanced reflectivity at the high-order Bragg peaks (especially those of the fifth and sixth orders) compared with S7 and also exhibit lower roughness of the Al layers (Tables 3 and 4).
From the above results, we see that the Si barrier layers and the Si layers within the Al layers have different roles in the novel multilayers. The Si barrier layers reduce the interaction between the Al and Zr layers and transform them to Al–Si–Zr alloy layers. The other Si layers penetrate the Al layers to form Al(51wt.%Si) layers, inhibiting crystallization of Al and lowering the roughness of the Al layers. As a result of the combined effects of the two types of Si layers, the reflectivity is improved at a grazing angle of incidence. Based on previous works [1,3], the multilayers with lower interfacial roughness could have better optical performance in EUV reflectivity. Based on this reason, sample S7 might have the lowest reflectivity, and the other samples (S1-S4) could not have better optical performances than the results of S5 and S7. Therefore, when investigating EUV reflectivity at a near-normal angle of incidence, we decided to focus on the samples S5, S6, and S7.
4.2 Near-normal-incidence EUV reflectance
In order to confirm the GIXR results shown in Tables 2–4 and Fig. 2, the EUV reflectivity of the multilayers was measured at near-normal incidence in the wavelength region 16.5–20.5 nm. The results are shown in Fig. 4(a), together with curves fitted using IMD software , with the same parameters and models as for the fitting of the GIXR data [Tables 3 and 4]. The reflectivity of S7, S5, and S6 are 48.2% at 18.6 nm, 48.7% at 18.2 nm, and 50.0% at 19.0 nm, respectively. Because the optical constants of Al vary in the wavelength range 18.2–19.0 nm, we can compare the reflectivity of each pair of multilayers. S6 should have a much higher reflectivity than that of S5 when the thickness of the S6 sample could be controlled down to 9.5 nm. The S5 and S7 samples have similar thicknesses and similar optical constants of Al in these multilayers. The reflectivity of S5 is much better than that of S7.
Although in a theoretical model the Al layers could remain amorphous and the roughness of each layer is not too large in the data fit from GIXR (Table 3), as can be seen from Fig. 4(a), the observed reflectivity of sample S5 is still lower than that of sample S6. Because too many Si layers are inserted into the Al layers in sample S5 to keep the total periodic thickness around 9.8 nm (the theoretical periodic thickness), the optical constants of the Al are changed, which affects the optical performance of the multilayer in the EUV measurements. Based on the results, we can find that S6 has better optical performance. The lower reflectivity of S7 could be influenced by the different layer structure, which may hinder the optical performance, even though the crystallization of the Al layers could be the major contributor.
4.3 X-ray diffraction
To further confirm the GIXR and EUV reflectivity results, the novel multilayer structures with Si barrier layers were characterized by XRD measurements. Figure 4(b) shows the diffraction curves of S5, S6, and S7. There are three phases for the sample of S7, at 38.79° (Al<111>), 35.30° (Zr<002>), and 36.50° (Zr<101>), but just one obvious peak for each of the novel multilayers at 38.64° and 38.67° for S5 and S6, respectively. The difference in diffraction peak positions for S5, S6, and S7 indicates that there are different Al–Si alloys present in the novel multilayers with Si barriers.
From Fig. 4(b), it can be seen that the peak height of Al<111> in the sample of S7 is higher than those for the novel multilayers (S5 and S6). The Al layers in S7 are highly oriented in Al<111> with complete crystallization, but the crystallization of the Al layer in S5 and S6 is inhibited owing to the presence of the insertion of an Si layer.
It appears that the novel materials could differ in their Al layers depending on the number of Si layers present. New Al–Si alloys [such as Al(51wt.%Si) in S6] could influence the crystallization of the original Al layer, which would be in accord with the results of GIXR and EUV. The Zr<002> and Zr<101> phases are present in S7 but almost nonexistent in the novel multilayers. This may be related to large interdiffusion between the Si barrier layer and Zr layer.
Our analysis of the XRD results suggests that the insertion of Si layers into the Al layers inhibits the crystallization of Al and lowers the interfacial roughness. The Si barrier layers not only decrease the roughness of the interfacial boundary but also enhance the crystallization of Zr. These properties all indicate that the novel structures can improve the reflectivity of Al/Zr multilayers.
To discuss the potential further applications of novel multilayer structures, we first consider Al/Zr systems. In our previous work [1–3,7,8], we found that crystallization of Al could adversely affect the optical and structural performance of Al/Zr multilayers. In order to reduce this effect, we have designed novel multilayers in which the Al layers are divided into many layers by Si layers. In addition, Si barrier layers can be inserted at the interfaces between the Al and Zr layers.
Based on experimental measurements and data fitting from GIXR, EUV, and XRD investigations, we found that the best structural performance is achieved when the Al layers are not divided into too many layers; otherwise, the Si layers influence the optical constants of the Al and lower the reflectivity. Except for the addition of two Si barrier layers at Al-on-Zr and Zr-on-Al interfaces in one period, the optimal approach is to use only one Si layer inserted into the Al layer, which allows the thickness of the layer to be kept below 3.0 nm.
Depending on the intended practical application, the optimal multilayer structure will be different in other systems. The Al/Zr multilayers investigated here represent a novel approach to the design of multilayer structures with the aim of limiting crystallization of the metal layer. We believe that this approach is also suitable for other types of multilayers intended for EUV and soft-X-ray applications in which crystallization of material layers could affect optical and structural properties such as reflectivity and stress. This is the case for Al/Mo multilayers [18,19], where reflectivity is adversely affected by crystallization of Al. We can use our novel multilayer structure to prevent the formation of Al crystallites and improve the performance of Al/Mo multilayers in practical applications. For some other multilayers, the main problem limiting their performance is the presence of residual stresses. A number of methods to reduce residual stresses in multilayers have been tried, such as using buffer layers  or sputtering the material layers with different gases (Ni2/Ar  or Ar/air  mixtures and low Ar pressure ). Although these methods can be useful, we suggest a simpler approach based on the use of novel multilayer structures. Stress is influenced by the preferred orientation of crystallites (texture), the number of layers, and the layer thickness, with the principal problem being the texture of the material layer . We believe that the novel multilayer design could effectively prevent the crystallites adopting a preferred orientation in the material layers. For example, in Mo/Si and Mo/Be multilayers , crystallization of Mo is the main cause of increased residual stress. It has been found experimentally that using buffer layers could solve this problem, but at the cost of decreased reflectivity. On the other hand, the novel multilayer structure proposed here could inhibit crystallization of Mo and reduce the residual stress without affecting optical performance.
The novel multilayer structures could also have applications in the semiconductor industry. In particular, crystallization of the hole transport layer (HTL) can decrease the lifetime of organic light-emitting diodes (LEDs). A layer based on aromatic hydrocarbon compounds can be inserted into the HTL to prevent its crystallization and thereby increase the LED lifetime from 10 000 hours to 50 000 hours .
In a word, the design of the novel multilayer structures is flexible, depending on the intended application, but in each case it is possible to effectively solve the problems caused by crystallization of material layers without adverse effects on other performance characteristics.
We have demonstrated that novel Al/Zr multilayer structures can enhance reflectivity at both grazing and near-normal incidence. Based on the results of GIXR, EUV, and XRD measurements, the roughness of the Zr and Al layers of the novel multilayers are lower than those of traditional Al/Zr multilayers. The inserted Si layers can interact with the Al and Zr layers, with the formation of different alloys, Al(51wt.%Si) and Al–Zr–Si alloy. Thus, Si has two functions in the multilayers. The Si layers inserted into the Al layers inhibit crystallization of Al and influence the position of the diffraction peak of Al<111>. The Si barrier layers inserted between the Al and Zr layers smooth the interfacial boundary and enhance the crystallization of Zr. Although it is possible to divide the Al layers into many thin layers, too many inserted Si layers can adversely affect optical and structural performance. The optimal multilayer structure is obtained by keeping the thickness of Al below 3.0 nm and inserting only one Si layer into the Al layers, and by inserting two Si barrier layers between the Al and Zr layers in a single period. Finally, we believe that the novel multilayer structures could not only enhance the reflectivity of EUV and soft-x-ray multilayers over a wide range of angles of incidence but also could solve other problems caused by the crystallization of the material layers in other applications of multilayers.
This work is supported by the National Basic Research Program of China (No. 2011CB922203) and the National Natural Science Foundation of China (Nos. 10978002 and 11027507). Part of this work was done in the framework of the COBMUL project funded by both Agence National de la Recherche in France (No. 10-INTB-902-01) and the Natural Science Foundation of China (No. 11061130549). The authors would also like to acknowledge the essential contributions of Maria Guglielmina Pelizzo, who performed the EUV reflectometry measurements using synchrotron radiation that we have reported here.
References and links
1. Q. Zhong, W. B. Li, Z. Zhang, J. T. Zhu, Q. S. Huang, H. Li, Z. S. Wang, P. Jonnard, K. Le Guen, J.-M. André, H. J. Zhou, and T. L. Huo, “Optical and structural performance of the Al/Zr reflection multilayers in the 17–19 nm region,” Opt. Express 20(10), 10692–10700 (2012). [CrossRef] [PubMed]
2. Q. Zhong, Z. Zhang, J. T. Zhu, Z. S. Wang, P. Jonnard, K. Le Guen, and J.-M. André, “The chemical characterization and reflectivity of the Al(1.0%wtSi)/Zr periodic multilayer,” Appl. Surf. Sci. 259, 371–375 (2012). [CrossRef]
3. Q. Zhong, Z. Zhang, J. T. Zhu, Z. S. Wang, P. Jonnard, K. Le Guen, Y. Y. Yuan, J.-M. André, H. J. Zhou, and T. L. Huo, “The thermal stability of Al(1%wtSi)/Zr EUV mirrors,” Appl. Phys. A: Mater. Sci. Process. 109(1), 133–138 (2012). [CrossRef]
4. E. Meltchakov, C. Hecquet, M. Roulliay, S. D. Rossi, Y. Menesguen, A. Jérome, F. Bridou, F. Varniere, M.-F. Ravet-Krill, and F. Delmotte, “Development of Al-based multilayer optics for EUV,” Appl. Phys. A: Mater. Sci. Process. 98(1), 111–117 (2010). [CrossRef]
5. M.-H. Hu, K. Le Guen, J.-M. André, P. Jonnard, E. Meltchakov, F. Delmotte, and A. Galtayries, “Structural properties of Al/Mo/SiC multilayers with high reflectivity for extreme ultraviolet light,” Opt. Express 18(19), 20019–20028 (2010). [CrossRef] [PubMed]
7. Q. Zhong, Z. Zhang, S. Ma, R. Q. Ze, J. Li, Z. S. Wang, P. Jonnard, K. Le Guen, and J.-M. André, “Thermally-induced structural modification in the Al/Zr multilayers,” Appl. Surf. Sci. (submitted), http://hal.archives-ouvertes.fr/hal-00744814.
8. Q. Zhong, Z. Zhang, S. Ma, R. Q. Ze, J. Li, Z. S. Wang, K. Le Guen, J.-M. André, and P. Jonnard, “The transition from amorphous to crystalline in Al/Zr multilayers,” J. Appl. Phys. 113(13), 133508 (2013). [CrossRef]
9. A. Aquila, F. Salmassi, Y. W. Liu, and E. M. Gullikson, “Tri-material multilayer coatings with high reflectivity and wide bandwidth for 25 to 50 nm extreme ultraviolet light,” Opt. Express 17(24), 22102–22107 (2009). [CrossRef] [PubMed]
10. H.-J. Stock, F. Hamelmann, U. Kleineberg, D. Menke, B. Schmiedeskamp, K. Osterried, K. F. Heidemann, and U. Heinzmann, “Carbon buffer layers for smoothing superpolished glass surfaces as substrates for molybdenum /silicon multilayer soft-x-ray mirrors,” Appl. Opt. 36(7), 1650–1654 (1997). [CrossRef] [PubMed]
11. H. L. Bai, E. Y. Jiang, C. D. Wang, and R. Y. Tian, “Enhancement of the reﬂectivity of soft-x-ray Co/C multilayers at grazing incidence by thermal treatment,” J. Phys. 8, 8763–8776 (1996).
12. A. Kloidt, K. Nolting, U. Kleineberg, B. Schmiedeskamp, U. Heinzmann, P. Müller, and M. Kühne, “Enhancement of the reflectivity of Mo/Si multilayer x-ray mirrors by thermal treatment,” Appl. Phys. Lett. 58(23), 2601–2603 (1991). [CrossRef]
14. S. Braun, H. Mai, M. Moss, R. Scholz, and A. Leson, “Mo/Si multilayers with different barrier layers for applications as extreme ultraviolet mirrors,” Jpn. J. Appl. Phys. 41(Part 1, No. 6B), 4074–4081 (2002). [CrossRef]
15. P. A. Totta and R. P. Sopher, “SLT device metallurgy and its monolithic extension,” IBM J. Res. Develop. 13(3), 226–238 (1969). [CrossRef]
16. M. Wormington, C. Panaccione, K. Matney, and D. Bowen, “Characterization of structures from X-ray scattering data using genetic algorithms,” Philos. Trans. R. Soc. Lond. A 357(1761), 2827–2848 (1999). [CrossRef]
17. D. L. Windt, “IMD—software for modeling the optical properties of multilayer films,” Comput. Phys. 12(4), 360–370 (1998). [CrossRef]
19. H. Nii, M. Miyagawa, Y. Matsuo, Y. Sugie, M. Niibe, and H. Kinoshita, “Control of roughness in Mo/Al multilayer film fabricated by DC magnetron sputtering,” Jpn. J. Appl. Phys. 41(Part 1, No. 8), 5338–5341 (2002). [CrossRef]
20. P. B. Mirkarimi, “Stress, reflectance, and temporal stability of sputter-deposited Mo/Si and Mo/Be multilayer films for extreme ultraviolet lithography,” Opt. Eng. 38(7), 1246–1259 (1999). [CrossRef]
21. M. S. Kumar, P. Böni, S. Tixier, and D. Clemens, “Stress minimization in sputtered Ni/Ti supermirrors,” Physica B 241–243, 95–97 (1997). [CrossRef]
22. K. MacArthur, B. Shi, R. Conley, and A. T. Macrander, “Periodic variation of stress in sputter deposited Si/WSi2 multilayers,” Appl. Phys. Lett. 99(8), 081905 (2011). [CrossRef]
23. L. Lutterotti, D. Chateigner, S. Ferrari, and J. Ricote, “Texture, residual stress, and structural analysis of thin films using a combined x-ray analysis,” Thin Solid Films 450(1), 34–41 (2004). [CrossRef]
24. Z. D. Popovic, S. Xie, N. Hu, A. Hor, D. Fork, G. Anderson, and C. Tripp, “Life extension of organic LEDs by doping of a hole transport layer,” Thin Solid Films 363(1-2), 6–8 (2000). [CrossRef]