Introduction

Continuous scaling of silicon-based transistors has led the rapid growth of the semiconductor industry over the last 40 years. During this period, high-quality interfaces between the Si substrate and SiO2 gate dielectrics facilitated the steady downscaling of Si devices, which accelerated the operation speed while reducing the power consumption1. However, as the thickness of SiO2 is decreased to less than a few nanometers, the traditional fabrication process faced with significant leakage currents that originate from tunneling through ultrathin SiO2 dielectrics1,2,3. This was resolved by incorporating high dielectric constant (high-k) materials such as HfO2 and ZrO23,4,5,6,7,8. These high-k oxides can reduce leakage currents by increasing the physical thickness of insulating layers while enhancing capacitive coupling between the channel layer and gate electrode.

Currently, the rapid expansion of mobile devices and high-performance computing markets are driving further development of transistors towards higher performance and lower power consumption. This in turn necessitates dielectric materials with higher-k than those of HfO2 or ZrO2; according to the International Roadmap for Devices and Systems (IRDS)9, dielectrics with k of 50~100 will be required in transistors or capacitors by 2024. Among the oxides, rutile TiO2 or SrTiO3 with k >100 are attracting interests as next-generation gate dielectrics but their small band gaps cause significant leakage currents7,10,11,12,13. On the other hand, new channel materials such as Ge, InSb and InGaAs are considered for the next-generation semiconducting devices because intrinsic carrier mobilities are higher in these materials than in Si9. However, the interface between these materials and oxides, for instance Ge/GeO2 and InGaAs/HfO2, are more defective than the Si/SiO2 interface, degrading the carrier mobility in actual devices14,15,16,17. The foregoing discussions indicate that a more diverse library of high-k materials will be beneficial in coping with the challenges in next-generation semiconducting devices. In particular, non-oxide dielectrics may provide solutions to issues that occur with conventional oxide dielectrics. For example, a recent study showed that CaF2 is superior to Al2O3 as a gate dielectric layer in the p-GaN device by reducing interface trap densities18. It was also reported that CaF2 forms a stable interface with the GaAs substrate19. Therefore, the property database of dielectric constants and band gaps (Eg’s) covering both oxides and non-oxides will be useful in selecting optimal high-k dielectrics.

Considering the huge material space of known dielectrics, it is not feasible to build a big database on Eg and k experimentally. Recently, owing to development of the density-functional theory (DFT) and exponentially growing computational speed, it becomes possible to conduct massive calculations on dielectric and electronic properties of crystals. In a previous study20, we carried out high-throughput DFT screening over ~1,800 oxides and identified new candidate high-k oxides such as c-BeO whose figure of merit far exceeding that of industry-standard HfO2. However, discussions in the above imply that non-oxide dielectrics would be also valuable in view of exploiting diverse chemistry. Recently, Petousis et al. performed high-throughput screening on ~1,000 inorganic compounds including both oxide and non-oxide compounds21. However, the search space in ref.21. was limited to stable or metastable phases with hull energies in the phase diagram less than 20 meV/atom, and relatively small primitive cells containing less than 20 atoms. However, such restrictive conditions could miss promising high-k materials. For example, the industry-standard HfO2 thin film includes a significant portion of tetragonal or cubic phases that exhibit high k values22. However, these are high-temperature phases and their energies are higher than for monoclinic HfO2 by more than 50 meV/atom23. In addition, c-BeO which was suggested to be a promising high-k candidate in ref.20 is also a metastable phase with the energy of 483 meV/atom with respect to the stable wurtzite BeO. Therefore, a more extensive table of Eg-k relations for non-oxide compounds is in demand. We note that the open material database such as Materials Project24 or AFLOW25 provides Eg computed by the semilocal functional. However, the semilocal functional severely underestimates Eg such that it is not appropriate in screening dielectric materials, which requires accurate band gaps. In addition, these open databases do not provide dielectric constants. (Materials Project seems to open dielectric constants only for materials studied in ref.21.).

In this study, by conducting high-throughput ab initio calculations on accurate Eg and k, we screen high-k non-oxide materials such as carbides, nitrides, sulfides, phosphides, chloride and fluorides. Since the number of these compounds amounts to ~30,000 entries in the current the Inorganic Crystal Structure Database (ICSD)26,27, the computation on the whole materials is not feasible within our computational resource. Therefore, we first limit the screening to binary phases (869 structures) and compare Eg-k relations depending on the anion species. We find that the Eg-k relation of binary fluorides looks most promising and so extend the screening space to ternary fluorides (415 structures). Consequently, we identify candidate fluorides that are suitable for high-k dielectrics.

Results

Automation workflow

Figure 1 shows the overall workflow of the automated computations. First, from ICSD (the 2015 version), we garner structural information on ordered binary crystals that were experimentally identified and contain only one of carbon, nitrogen, fluorine, phosphorus, sulfur or chlorine atoms (non-oxide groups hereafter). We exclude compounds including 3d transition metal elements with partially occupied d orbitals (V ~ Cu) because their band gaps are usually smaller than 3 eV, and so they are not suitable for high-k applications. In addition, DFT + U methods that are necessary for 3d orbitals can significantly underestimate dielectric constants by hardening phonon modes28. We also omit large primitive cells that contain more than 50 atoms in the unit cell due to a sheer computational cost. (The total number of such structures is 53, and their dielectric constants are typically small.) After these pre-screening steps, we perform the structural relaxation and calculate Eg and k using the in-house automation package (Automated Ab initio Modeling of Materials Property Package (AMP2))29 for high-throughput calculations. (See the Methods section for further computational details.).

Figure 1
figure 1

The workflow of calculating Eg and k for non-oxide compounds.

Validation of automatic calculations

In the present work, the band gap is calculated within the hybrid functional (HSE06) with structural parameters (lattice vectors and atomic coordinates) fixed to those obtained using the generalized gradient approximation (GGA) functional. To reduce the computational cost, we employ the HSE@GGA scheme in which the HSE calculation is performed on the band edge points identified by GGA. In ref.20, the HSE@GGA scheme was validated by comparing with experimental and other theoretical band gaps for selected oxides. Similarly, Fig. 2(a) tests this approach against various non-oxide compounds. The estimated band gaps are in good agreement with the experiment except for compounds with Eg > 8 eV that show noticeable discrepancies30. This is caused by the fixed fraction of the exact exchange term; it is known that materials with large Eg require higher fractions of the Fock term due to the weak electronic screening31. For a comparison purpose, we also present in Fig. 2(a) results with G0W0@HSE in which one-shot GW calculation is performed based on the HSE result32. It is seen that Eg’s from G0W0@HSE are in very good agreement with experimental values. However, the GW method is too expensive to be used in the high-throughput screening.

Figure 2
figure 2

Comparison of experimental57,58,59,60,61,62,63 and theoretical data for (a) the band gap (Eg) and (b) the static dielectric constant (k). HSE@GGA means a scheme applying the hybrid-functional calculations on the band-edge points identified by GGA. For the comparison purpose, results with G0W0@HSE are also presented.

For dielectric constants, we compare GGA and LDA results of selected compounds with experimental values (see Fig. 2(b)). The estimated dielectric constants are in good agreements with experiment regardless of the functional, though results show that GGA tends to give higher dielectric constant compared to LDA. The mean average deviation (MAD) is slightly lower with LDA than GGA (0.70 and 0.98, respectively). Considering that GGA tends to overestimate dielectric constants for high-k materials20, we employ the LDA scheme in evaluating dielectric constants. To note, we do not consider the HSE functional for evaluating k because HSE tends to underestimate the optical dielectric constant32. Furthermore, HSE is not superior to LDA for the static dielectric constant, as was demonstrated for TiO228.

High-throughput screening for binary non-oxides

For binary dielectrics, we consider 869 compounds (76 carbides, 123 nitrides, 132 fluorides, 205 sulfides, 194 phosphides and 139 chlorides). Among them, 20 carbides, 49 nitrides, 86 fluorides, 82 phosphides, 114 sulfides, and 98 chlorides are found to be insulators. The Eg-k relation of each non-oxide group is presented in Fig. 3. All the numerical data are provided in Tables S1S6 in Supplementary Information. For the comparison purpose, Eg-k of oxides from ref.20 are also plotted as gray points in Fig. 3(a). The metallic compounds are not displayed since their dielectric constants are ill defined. We also exclude unstable structures that show imaginary optical modes in the unit cell calculation.

Figure 3
figure 3

Band gap versus dielectric constants for (a) carbides and oxides20 (grey), (b) nitrides, (c) fluorides, (d) phosphides, (e) sulfides, and (f) chlorides. The number of data points are 20 (carbides), 49 (nitrides), 86 (fluorides), 82 (phosphides), 114 (sulfides), and 98 (chlorides).

In Fig. 3, it is seen that the inverse relationship between Eg and k persist for every compound. However, the detailed distributions are distinct between the material groups. Especially, it is noticeable that fluorides have wide band gaps than other non-oxide groups. The band gap is related to the energy splitting between bonding and anti-bonding orbitals. Therefore, a large difference in electronegativity, small ionic size, and a high degree of orbital overlap contribute to strong bond strength and large band gap. Fluorine has the highest electronegativity in the Periodic Table and its ionic size is the smallest among considered anions. This results in a much broader Eg range than those of other non-oxides. The maximum band gap follows the order of fluorides > oxides > chlorides > nitrides > sulfides > carbides > phosphides, which is also in line with the order of electronegativity. The correlation between Eg and electronegativity was also discussed previously33,34.

Distribution of dielectric constants

In Fig. 3, the inverse relationship between Eg and k consistently appear, which puts a fundamental limitation on the existence of ideal high-k dielectrics that have large Eg and k simultaneously. For detailed analysis on this, we divide the dielectric constant into electronic and ionic contributions (kel and kion, respectively), and plot Eg-kel and Eg-kion relations separately for all the data on binary compounds (see Figs. 4(a) and 4(b), respectively.) The density of data points is drawn in contours by representing each point with a Gaussian. In Fig. 4(a), a clear inverse relation is found between Eg and kel. This can be rationalized by the fact that the band gap reflects the bonding-antibonding separation and so a small Eg means higher electronic polarizability associated with the facile excitation into antibonding states, leading to higher kel. On the other hand, it is seen in Fig. 4(b) that the data points in Eg-kion are more scattered such that the inverse relationship between Eg and kion is weaker than for between Eg and kel. The ionic dielectric constant is dictated by off-centering of the cations with respect to the anions (and vice versa) under electric fields, the degree of which depends on the bond strength. Compared to the band gap, the bond strength or phonon frequency is highly sensitive to the bond length. Therefore, various bond lengths among similar compositions result in a wide variation of kion. Since the dielectric constant of high-k materials is mostly contributed by kion, this weak correlation between Eg and kion increases a chance of finding new high-k materials by expanding the search space.

Figure 4
figure 4

Dielectric constant versus band gap relation for binary non-oxides. The dielectric constant is split into the (a) electronic part (kel) and (b) ionic part (kion). The black dots indicate each material. The density distribution of data points is visualized in contour plots.

Figure 5
figure 5

Eg-k map for 86 binary and 415 ternary fluorides. Each material is color coded according to the figure of merit that is the product of Eg and k. The candidate fluorides for high-k dielectrics are marked in red (stable phase) or green (metastable phase) circles. As a reference, monoclinic and tetragonal HfO2 (m-HfO2 and t-HfO2, respectively) are also noted.

High-throughput screening for ternary fluorides

Figure 3 indicates that band gaps of carbides, nitrides, phosphides and sulfides are distributed mostly over medium to small values and therefore, these material groups might not be appropriate for high-k applications. Both fluorides and chlorides follow similar Eg-k relations but fluorides show a broader distribution. Considering the above discussion on a loose relation between Eg and kion, it might be worthwhile to extend the search space to ternary fluorides. There are 644 ternary fluorides reported in ICSD and the calculated Eg-k relations for 415 ternary fluorides with finite gaps are provided in Fig. 5, together with the binary fluorides presented in Fig. 3(c). The numerical data are compiled in Table S7 in Supplementary Information. To rank the candidate materials, we assign Eg·k as the figure of merit (FOM) because Eg and k are approximately proportional to the logarithm of the leakage current density20,35,36. Data points in Fig. 5 are colored according to FOM.

In Fig. 5, we identify candidate fluorides that merit consideration as high-k dielectrics. For the candidate materials, we confirm the dynamical stability with phonon analysis and neglect dynamically unstable structures. (See the Methods section for details on the phonon calculations.) As a reference, monoclinic and tetragonal HfO2 (m-HfO2 and t-HfO2, respectively) are also marked. It is seen that no fluorides outperform tetragonal HfO2 (t-HfO2) that is currently industry-standard high-k dielectrics. However, t-HfO2 is a high-temperature phase and hence its stabilization at the room temperature requires strain engineering or dopants during device fabrication37,38. In contrast, many candidate materials in Fig. 5 are stable phases (9 stable and 3 metstable) and so synthesis would be more straightforward than HfO2.

Discussions

In Table 1, we enlist candidate fluorides that were identified in Fig. 5. Besides Eg and k values, relative energies with respect to the most stable phase at ambient conditions are also provided. Some materials in Table 1 were also noted in ref.21 but the dielectric constants were larger than the present results because of the functional difference (see above). Among the binary phases, BiF3 (see Fig. 6(a)) looks promising with Eg and k values close to those of t-HfO2. BiF3 was used as dielectric buffer layer for surface plasmon resonance39. However, unlike t-HfO2 that is metastable, BiF3 is the stable phase and therefore we expect that the fabrication would be easier, and the film quality would be more uniform than HfO2. The polymorphs of LaF3 are also intriguing as they possess large band gaps of 9~10 eV. LaF3 was used as dielectric buffer layer39 or UV coating40. There are several polymorphs in LaF3 and their dielectric constants range over 11~17 with metastable phases showing larger values (see Table S3). The Pmmn structure with the largest k is shown in Fig. 6(b).

Table 1 Candidate fluorides for high-k dielectrics with Eg > 4 eV and figure of merit (FOM) > 120. ∆E means the energy difference with respect to that of the stable phase.
Figure 6
figure 6

Crystal structures of (a) BiF3, (b) LaF3, (c) BaBeF4, and (d) TlAlF4.

There are several Ge-F compounds such as Ge3F8, Ge5F12, GeF2, and GeF4. The band gap of these compounds range over 5~7.5 eV. (See Table S3) This is in contrast with GeO2 whose band gap is only 2.8 eV. The small band gaps of GeO2 and suboxides result in high-density defect states, degrading the interface quality in Ge transistors41. In this respect, the large band gap of germanium fluorides may contribute to forming stable interfaces in Ge devices by playing as the passivation layer that removes interface defects.

Among the ternary fluorides, BaBeF4 is noticeable as the material possess a large band gap of 9.7 eV and dielectric constant of 17, surpassing that of LaF3. (See Fig. 6(c).) We note that tetragonal TlAlF4, which has the highest FOM in ref.21, is found to be dynamically unstable in the present calculation (both LDA and GGA) while dynamically stable monoclinic TlAlF4 (see Fig. 6(d)) has k of 27. Nevertheless, the dielectric property of monoclinic TlAlF4 approaches that of t-HfO2.

In summary, we conducted high-throughput calculations of Eg and k for 449 binary non-oxides and 415 ternary fluorides. We confirmed that inverse relationships between Eg and k are present in non-oxide compounds like in oxide compounds. Among the different anion groups, binary fluorides are the most promising as they show a wide distribution of Eg. By further screening over ternary fluorides, we identified fluorides such as BiF3, LaF3, and BaBeF4 that could serve as useful high-k dielectrics. We believe that the suggested fluoride compounds may contribute to resolving various issues in microelectronic devices that is caused by using only oxide dielectrics.

Methods

Computational details

The DFT calculations were performed using Vienna ab initio simulation package (VASP)42,43,44 based on the projector augmented wave (PAW) pseudopotential45,46. For the exchange-correlation functional, we employ the generalized gradient approximation (GGA) in the form of the Perdew-Burke-Ernzerhof (PBE)47. Compounds bearing Zn, La, and Ce are performed with the GGA + U method48. The effective U parameter of 7.5 eV is used for Zn d and La f orbitals49,50 while 4.5 eV is used for Ce f orbitals51. We also carry out the hybrid functional (HSE06) calculation52 to overcome underestimation of the band gap in GGA. The k-point meshes are selected by ensuring that the total energy, stress tensor components, and all forces are converged within 10 meV/atom, 10 kbar, and 0.02 eV Å−1, respectively. The atomic positions and lattice parameters are relaxed until the total energy, atomic forces, and stress tensors are reduced to within the same criteria.

Band gap

We adopt an automated approach in estimating the band gap as was detailed in our former high-throughput study20. To be brief, we first identify k points corresponding to the valence band maximum and conduction band minimum by sweeping along the lines connecting high-symmetry points53 using GGA(+U) functional. Then, we perform the one-shot hybrid functional calculation on these band-edge points.

Dielectric constant

The density-functional perturbation theory (DFPT)54,55 implemented in VASP is used to estimate Born effective charges, phonon frequencies, and dielectric constants. In calculating the dielectric constants, the k-point density is doubled because DFPT is sensitive to this computational parameter.

Phonon analysis

Phonon bands are calculated using the PHONOPY package56.