Abstract
Tetrahedra‐based nitrides with network structures have emerged as versatile materials with a broad spectrum of properties and applications. Both nitridosilicates and nitridophosphates are well‐known examples of such nitrides that upon doping with Eu2+ exhibit intriguing luminescence properties, which makes them attractive for applications. Nitridosilicates and nitridophosphates show manifold structural variability; however, no mixed nitridosilicatephosphates except SiPN3 and SiP2N4NH have been described so far. The compounds AESiP3N7 (AE=Sr, Ba) were synthesized by a high‐pressure high‐temperature approach using the multianvil technique (8 GPa, 1400–1700 °C) starting from the respective alkaline earth azides and the binary nitrides P3N5 and Si3N4. The latter were activated by NH4F, probably acting as a mineralizing agent. SrSiP3N7 and BaSiP3N7 were obtained as single crystals. They crystallized in the barylite‐1O (M=Sr) and barylite‐2O structure types (M=Ba), respectively, with P and Si being occupationally disordered. Cation disorder was further supported by solid‐state NMR spectroscopy and energy‐dispersive X‐ray spectroscopy (EDX) mapping of BaSiP3N7 with atomic resolution. Upon doping with Eu2+, both compounds showed blue emission under UV excitation.
Keywords: disordered compounds, high-pressure chemistry, nitrides, phosphorus, silicon
The first nitridosilicatephosphates are synthesized using high‐pressure high‐temperature reactions. The incorporation of P and Si as well as their occupational disorder in the compounds AESiP3N7 (AE=Sr, Ba) is proven by solid‐state NMR spectroscopy, STEM‐EDX mapping, and single‐crystal diffraction. Eu2+‐doped samples show blue luminescence under irradiation with UV light and therefore expand the range of possible host lattices for luminescent materials.

Introduction
Emerging environmental consciousness has pushed the development of solid‐state lighting solutions forward. The invention of efficient InGaN‐based blue LEDs (light‐emitting diodes) enabled the development of pc‐LEDs (phosphor‐converted) with remarkable properties in terms of color temperature, color rendition, and efficacy. Significant improvements in the aforementioned properties were possible due to nitride compounds such as M 2Si5N8:Eu2+ (M=Sr, Ba), MSi2O2N2:Eu2+ (M=Ca, Sr, Ba), SrLiAl3N4:Eu2+, and MAlSiN3:Eu2+.[ 1 , 2 , 3 , 4 ]
Materials properties concerning solid‐state lighting can be tuned by dopant concentration to a limited extent, affecting Stokes shifts in emission spectra, or by a variation of the size of coordination polyhedra by substitution such as introducing Sr on Ba sites. Completely shifted emission properties, however, can only be achieved by a fundamental alteration of the host lattice.[ 5 , 6 ]
The main goal of this work was to expand the compositional and structural diversity of tetrahedra‐based luminescent materials. Thus, discovery of the title compounds SrSiP3N7 and BaSiP3N7 opens up the novel compound class of mixed nitridosilicatephosphates, which can now be further explored as innovative host lattices. While nitridosilicates have been investigated thoroughly and nitridophosphates show similar promising structures and properties, only two compounds that contain both SiNx (x=4, 6) and PN4 units have been reported so far, that is, SiPN3 and SiP2N4NH.[ 7 , 8 , 9 ] The crystal structure of SiPN3 corresponds to a defect wurtzite‐type arrangement with mixed occupation of Si and P at the tetrahedral sites. The crystal structure of SiP2N4NH is related to sillimanite‐type Al2SiO5. It is built up from edge‐sharing SiN6 octahedra interconnected by all‐side vertex‐sharing PN4 tetrahedra. A possible explanation for the challenges concerning syntheses that are involved in the preparation of mixed nitridosilicatephosphates could be the chemical inertness of Si3N4, while P3N5 already decomposes at temperatures above 850 °C if no external pressure is applied. As shown in previous publications, according to Le Chatelier's principle, the decomposition of P3N5 under the formation of N2 is suppressed by external pressure. [10]
NH4Cl has been successfully employed as a mineralizer facilitating crystal growth of nitridophosphates. HCl formed in situ most likely leads to reversible P−N bond formation and cleavage.[ 11 , 12 ] After nitridosilicatephosphates proved to be not accessible with the help of NH4Cl, changing the mineralizing agent to NH4F afforded the title compounds AESiP3N7 (AE=Sr, Ba). This may be explained by the fact that HF cannot only reversibly cleave and form P−N bonds, but also Si−N bonds. The surface of Si3N4 features SiNH2 groups that can be attacked by F− in a nucleophilic substitution. [13]
Results and Discussion
The nitridosilicatephosphates AESiP3N7 (AE=Sr, Ba) were synthesized by high‐pressure high‐temperature (HP/HT) reactions at 8 GPa and 1400 °C (Ba) and 1700 °C (Sr), respectively, using a modified Walker‐type multianvil apparatus. [14] The synthesis of SrSiP3N7 at temperatures below 1700 °C resulted in samples with significant amounts of unknown side phases. Reactions followed the so‐called azide route using P3N5, Si3N4, and the respective metal azide as starting materials with additional NH4F (≈5 wt %) as a mineralizing agent [Eq. 1]. To investigate luminescence properties, samples with the addition of approximately 1 mol % of EuF3 (concerning AE 2+) to the starting mixture were prepared.
| (1) |
The title compounds were obtained as colorless powders (Eu2+‐doped samples of SrSiP3N7 show a yellow tint) and showed no sensitivity to air or moisture. More detailed information on the HP/HT synthesis is given in the Supporting Information.
The crystal structures were elucidated by single‐crystal X‐ray diffraction (SCXRD) using direct methods and least‐squares refinement. SrSiP3N7 crystallizes in space group Pmn21 (no. 31) with Z=2. BaSiP3N7 crystallizes in space group Pnma (no. 62) with Z=4); details are given in Tables 1 and S2–7. In addition, Rietveld refinements indicate the presence of BaSiP3N7 crystallizing in space group Pmn21 (no. 31) with Z=2 as a side phase. Both compounds are isotypic to the two polymorphs of barylite BaBeSi2O7. SrSiP3N7 corresponds to the barylite‐1O polymorph, whereas BaSiP3N7 features the structure of barylite‐2O. The structures of barylite‐1O and barylite‐2O represent the maximum degree of order (MDO) polytypes of their polytype family. Both structures consist of a network of all‐vertex‐sharing PN4 and (Si0.5P0.5)N4 tetrahedra and elongated square pyramid (J8) AEN9 (AE=Sr, Ba) polyhedra (Figure S1). [15]
Table 1.
Crystallographic data of the single‐crystal structure refinements of AESiP3N7 (AE=Sr, Ba). Standard deviations are given in parentheses.
|
Formula |
SrSiP3N7 |
BaSiP3N7 |
|---|---|---|
|
molar mass [g mol−1] |
306.69 |
356.41 |
|
crystal system |
orthorhombic |
orthorhombic |
|
space group |
Pmn21 (no. 31) |
Pnma (no. 62) |
|
lattice parameters [Å] |
a=11.979(2) |
a=9.9048(3) |
|
|
b=4.9040(10) |
b=12.1858(3) |
|
|
c=4.6870(9) |
c=4.73580(10) |
|
cell volume [Å3] |
275.34(10) |
571.60(3) |
|
formula units/unit cell |
2 |
4 |
|
density [g cm−3] |
3.699 |
4.142 |
|
μ [mm−1] |
10.807 |
7.927 |
|
temperature [K] |
296(2) |
297(2) |
|
absorption correction |
semiempirical |
|
|
radiation |
Mo‐Kα (λ=0.71073 Å) |
|
|
F(000) |
292 |
656 |
|
θ range [°] |
3.4≤θ≤44.09 |
3.34≤θ≤38.44 |
|
total no. of reflections |
9890 |
10 523 |
|
Independent reflections [I≥2σ(I)/all] |
1253/1379 |
959/1128 |
|
R σ, R int |
0.0393, 0.0952 |
0.0234, 0.0481 |
|
refined parameters |
60 |
58 |
|
goodness of fit |
1.100 |
1.044 |
|
R‐values [I≥2σ(I)] |
R1=0.0293 wR2=0.0731 |
R1=0.0261 wR2=0.0576 |
|
R‐values (all data) |
R1=0.0343 wR2=0.0750 |
R1=0.0340 wR2=0.0599 |
|
Δρ max, Δρ min [e Å−3] |
2.18, −1.49 |
0.81, −1.53 |
The main difference between the two polymorphs concerns the arrangement of tetrahedra. While in SrSiP3N7 all tetrahedra vertices point in the same direction, those in BaSiP3N7 alternate, which results in a doubled unit cell with 2b (SrSiP3N7)=a (BaSiP3N7) (Figure 1). Although tetrahedra orientation differs in both compounds, the environment of AE atoms is strikingly similar. The tetrahedra connection patterns show that both compounds consist of dreier, vierer, and sechser rings that, apart from slight distortions, are arranged and distributed in the same manner (Figure 2), leading to the same topology point symbol {32.43.5.64}{34.45.54.62}. [16] Both compounds exhibit one tetrahedrally coordinated site shared by Si and P, while the other site is solely occupied by P. Further details of the crystal structure investigations may be obtained from the joint CCDC/FIZ Karlsruhe online deposition service by quoting the deposition numbers CSD‐2050660 and 2050661. In the case of SrSiP3N7, potential ordering of Si and P was considered by symmetry reduction and refinement of the structure against SCXRD data in the subgroups P21, Pn, and Pm of space group Pmn21. However, no indications of complete ordering were found. In all structure models, the volumes of the resulting four symmetrically independent tetrahedra were compared as (P,Si)−N bond lengths were not sufficiently meaningful for discrimination. [17] This investigation led to two different kinds of tetrahedra.
Figure 1.

Crystal structures of SrSiP3N7 (top) and BaSiP3N7 (bottom) both along [001]. For SrSiP3N7 PN4 tetrahedra (red) and (Si0.5P0.5)N4 (orange) all vertices point in the same direction (behind the plane of projection). For BaSiP3N7 the orientation of tetrahedra vertices alternates.
Figure 2.

Tetrahedra connection patterns of SrSiP3N7 (left) and BaSiP3N7 (right). Patterns of both compounds show dreier (red), vierer (lilac), and sechser rings (green).
Structure models in subgroups P21, Pn, and Pm reveal two tetrahedra exhibiting a volume of 2.21–2.24 Å3 that coincides with the volume of PN4 tetrahedra from known nitridophosphates in literature and corresponds to one site in the final structure model in Pmn21. The other two tetrahedra have a volume of 2.37–2.40 Å3, which lies between the volumes of PN4 (2.13–2.28 Å3) and SiN4 tetrahedra (2.52–2.76 Å3) (Tables S10 and 11).[ 8 , 9 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 ] BVS (bond valence sum) calculations performed on all structure models revealed two tetrahedral sites fully occupied by P and two tetrahedral sites occupied by Si and P in a 1:1 ratio (Tables S13–15). [29]
In the case of BaSiP3N7, the ordering of tetrahedra was considered by symmetry reduction and refinement of the structure against SCXRD data in subgroups Pna21, P21 ma, and P21/c of space group Pnma. Only subgroups retaining the extinction condition of the a glide planes present in Pnma were taken into account because electron diffraction parallel to [001] showed no violation of the extinction conditions. Additional electron diffraction patterns parallel to [100] showed no violation of the extinction conditions of the n glide, too, further supporting the structure model in space group Pnma (a comparison of experimental diffraction patterns with simulated ones based on the structure model in space group Pnma is given in Figure S5).
The comparison of resulting tetrahedral volumes showed the same features as for SrSiP3N7 (Table S12). BVS calculations performed for the different structure models again suggested two sites completely occupied by P and two sites shared by Si and P (Tables S16–18). The simple approach of comparing tetrahedra volume as a tool for assigning Si and P, which lack scattering contrast, to the respective sites was indeed confirmed by scanning transmission electron microscope energy‐dispersive X‐ray spectroscopy (STEM‐EDX) mapping with atomic resolution for BaSiP3N7. These data support the model in space group Pnma, showing two sites with mixed Si/P occupation (Figure 3, enlarged version see Figure S6). This result is also corroborated by 31P solid‐state magic angle spinning (MAS)‐NMR spectra, which show a broad signal [full width at half‐maximum (fwhm)=19.7 ppm] that is consistent with a disordered model (Figure 4, Figure S4). Line broadening in the NMR spectrum is probably due to disorder in the second coordination sphere of P atoms. In contrast, ordered nitrides like BP3N6 or Li12P3N9 show very narrow signals in their 31P NMR spectra.[ 8 , 30 ]
Figure 3.

Atomic‐resolution STEM‐EDX of BaSiP3N7 along [001]. STEM high‐angle annular dark‐field (HAADF) image (top) with structure overlay (Ba: cyan, P: red, Si: yellow). The inset shows the corresponding area for EDX maps (bottom) with a combined color map, Ba map (cyan), Si map (yellow), and P map (red).
Figure 4.

31P NMR spectrum showing one broad signal for BaSiP3N7 instead of two signals for the different crystallographic sites, most likely due to the occupational disorder.
In both structures, the connectivity of the tetrahedra via their vertices can explain the presence of different tetrahedral volumes. The smaller tetrahedra have three vertices occupied by twofold bridging nitrogen atoms N [2] and one vertex occupied by a threefold bridging nitrogen atom N [3] . The larger tetrahedra, in contrast, feature two vertices occupied by N [2] and two vertices occupied by N [3] . Chemical analysis by EDX agrees with the sum formulas. Due to ambiguous O contents (as indicated by EDX measurements) either surface hydrolysis or slight compositional variations cannot completely be ruled out so that a phase width according to AESi1+xP3−xN7−xOx (AE=Sr, Ba) with x<1 could also be considered (Table S8) even though some analyses show no O.
Upon doping with Eu2+, both compounds emit blue light under UV excitation. Luminescence spectra show emission maxima of λ max=430 nm for SrSiP3N7:Eu2+ and λ max=424 nm for BaSiP3N7:Eu2+ upon excitation with λ exc=400 nm. The emission curves were extrapolated to give an estimate of the fwhm, which amount to 45 nm (2404 cm−1) for SrSiP3N7:Eu2+ and 53 nm (2731 cm−1) for BaSiP3N7:Eu2+ (Figure 5). The corresponding Stokes shifts are 38 nm (2254 cm−1) for SrSiP3N7:Eu2+ and 32 nm (1925 cm−1) for BaSiP3N7:Eu2+. The presence of a single narrow emission band for both phosphors can be explained by the emission properties of Eu2+and the presence of a single crystallographic site for the alkaline earth ions suitable for doping with Eu2+ with AE−N distances ranging from 2.696(3)–3.270(3) Å (SrSiP3N7) to 2.872(3)–3.230(3) Å (BaSiP3N7). The similarity of emission properties in terms of fwhm are most likely to be explained by the P‐Si0.5P0.5 „cages“ around the AE position, which are very similar. Thus, only minute deviations are caused by different AE cation sizes even though both structures differ with respect to their space groups, unit cell volumes, and tetrahedra orientations.
Figure 5.

Emission spectra of (a) SrSiP3N7:Eu2+ (blue) and (b) BaSiP3N7:Eu2+ (black); measured data in solid lines and extrapolation in dotted lines; respective excitation spectra (red) (insets: micrographs of luminescent particles).
Conclusions
High‐pressure high‐temperature synthesis with the addition of NH4F is a suitable approach to the synthesis of mixed nitridosilicatephosphates. The compounds AESiP3N7 (AE=Sr, Ba) adopt the structure types of the two polymorphs of the mineral barylite. This structure type has not been observed for nitride compounds so far. Although silicon and phosphorus exhibit little contrast in X‐ray diffraction, the comparison of polyhedra volumes led to structure models with an occupationally disordered site that also persists if potential ordering is considered by symmetry reduction. The disordered model for BaSiP3N7 is further supported by solid‐state NMR spectroscopy. Scanning transmission electron microscopy energy‐dispersive X‐ray spectroscopy (STEM‐EDX) mapping with atomic resolution enables to directly observe said disorder, which is additionally in accordance with systematic absences observed in electron diffraction patterns. Therefore, nitridosilicatephosphates have the potential to significantly diversify the structural chemistry of nitrides. Their suitability as host lattices for rare‐earth activator ions seems especially intriguing considering the emission properties of other compounds with multiple tetrahedra centers like CaAlSiN3:Eu2+, Sr[Li2Al2O2N2]:Eu2+, or Sr[LiAl3N4]:Eu2+.[ 2 , 31 , 32 ]
Conflict of interest
The authors declare no conflict of interest.
Supporting information
As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.
Supplementary
Acknowledgements
Financial support by the Deutsche Forschungsgemeinschaft DFG (projects SCHN 377/18‐1 and OE 513/6‐1) is gratefully acknowledged. We thank Dr. Peter Mayer for single‐crystal data collection, Dr. Thomas Bräuniger, Christian Minke, and Dr. Otto Zeman for NMR measurements, Lisa Gamperl (all at Department of Chemistry at LMU Munich) for SEM investigations and Dr. Philipp Strobel (Lumileds Phosphor Center Aachen) for luminescence measurements and helpful discussions. Open access funding enabled and organized by Projekt DEAL.
L. Eisenburger, O. Oeckler, W. Schnick, Chem. Eur. J. 2021, 27, 4461.
Dedicated to Professor Thomas M. Klapötke on the occasion of his 60th birthday
Contributor Information
Prof. Dr. Oliver Oeckler, Email: oliver.oeckler@gmx.de.
Prof. Dr. Wolfgang Schnick, Email: wolfgang.schnick@uni-muenchen.de.
References
- 1. Zeuner M., Pagano S., Schnick W., Angew. Chem. Int. Ed. 2011, 50, 7754–7775; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2011, 123, 7898–7920. [Google Scholar]
- 2. Pust P., Schmidt P. J., Schnick W., Nat. Mater. 2015, 14, 454–458. [DOI] [PubMed] [Google Scholar]
- 3. Uheda K., Hirosaki N., Yamamoto Y., Naito A., Nakajima T., Yamamoto H., Electrochem. Solid-State Lett. 2006, 9, H22. [Google Scholar]
- 4. Wang L., Xie R. J., Suehiro T., Takeda T., Hirosaki N., Chem. Rev. 2018, 118, 1951–2009. [DOI] [PubMed] [Google Scholar]
- 5. Li Y. Q., Delsing A. C. A., De With G., Hintzen H. T., Chem. Mater. 2005, 17, 3242–3248. [Google Scholar]
- 6. Wang F., Wang W., Zhang L., Zheng J., Jin Y., Zhang J., RSC Adv. 2017, 7, 27422–27430. [Google Scholar]
- 7. Baldus H. P., Schnick W., Lücke J., Wannagat U., Bogedain G., Chem. Mater. 1993, 5, 845–850. [Google Scholar]
- 8. Vogel S., Buda A. T., Schnick W., Angew. Chem. Int. Ed. 2018, 57, 13202–13205; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2018, 130, 13386–13389. [Google Scholar]
- 9. Wendl S., Eisenburger L., Strobel P., Günther D., Wright J. P., Schmidt P. J., Oeckler O., Schnick W., Chem. Eur. J. 2020, 26, 7292–7298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kloß S. D., Schnick W., Angew. Chem. Int. Ed. 2019, 58, 7933–7944; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2019, 131, 8015–8027. [Google Scholar]
- 11. Marchuk A., Pucher F. J., Karau F. W., Schnick W., Angew. Chem. Int. Ed. 2014, 53, 2469–2472; [DOI] [PubMed] [Google Scholar]; Angew. Chem. 2014, 126, 2501–2504. [Google Scholar]
- 12. Baumann D., Schnick W., Inorg. Chem. 2014, 53, 7977–7982. [DOI] [PubMed] [Google Scholar]
- 13. Martin Knotter D., (Dee) Denteneer T. J. J., J. Electrochem. Soc. 2001, 148, F43. [Google Scholar]
- 14. Huppertz H., Z. Kristallogr. Cryst. Mater. 2004, 219, 330–338. [Google Scholar]
- 15. Merlino S., Biagioni C., Bonaccorsi E., Chukanov N. V., Pekov I. V., Krivovichev S. V., Yakovenchuk V. N., Armbruster T., Mineral. Mag. 2015, 79, 145–155. [Google Scholar]
- 16. Blatov V. A., Shevchenko A. P., Proserpio D. M., Cryst. Growth Des. 2014, 14, 3576–3586. [Google Scholar]
- 17. Momma K., Izumi F., J. Appl. Crystallogr. 2011, 44, 1272–1276. [Google Scholar]
- 18. Schlieper T., Milius W., Schnick W., Z. Anorg. Allg. Chem. 1995, 621, 1380–1384. [Google Scholar]
- 19. Yamane H., Nagura T., Miyazaki T., Acta Crystallogr. Sect. E 2014, 70, i23–i24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Schmolke C., Bichler D., Johrendt D., Schnick W., Solid State Sci. 2009, 11, 389–394. [Google Scholar]
- 21. Pilet G., Höppe H. A., Schnick W., Esmaeilzadeh S., Solid State Sci. 2005, 7, 391–396. [Google Scholar]
- 22. Woike M., Jeitschko W., Inorg. Chem. 1995, 34, 5105–5108. [Google Scholar]
- 23. David J., Laurent Y., Charlot J.-P., Lang J., Bull. Soc. Fr. Mineral. Cristallogr. 1973, 96, 21–24. [Google Scholar]
- 24. Karau F. W., Schnick W., J. Solid State Chem. 2005, 178, 135–141. [Google Scholar]
- 25. Kloß S. D., Weidmann N., Niklaus R., Schnick W., Inorg. Chem. 2016, 55, 9400–9409. [DOI] [PubMed] [Google Scholar]
- 26. Landskron K., Schnick W., J. Solid State Chem. 2001, 156, 390–393. [Google Scholar]
- 27. Marchand R., L'Haridon P., Laurent Y., J. Solid State Chem. 1982, 43, 126–130. [Google Scholar]
- 28. Landskron K., Irran E., Schnick W., Chem. Eur. J. 1999, 5, 2548–2553. [Google Scholar]
- 29. Brese N. E., O'Keeffe M., Acta Crystallogr. Sect. B 1991, 47, 192–197. [Google Scholar]
- 30. Bertschler E., Niklaus R., Schnick W., Chem. Eur. J. 2017, 23, 9592–9599. [DOI] [PubMed] [Google Scholar]
- 31. Piao X., Machida K. I., Horikawa T., Hanzawa H., Shimomura Y., Kijima N., Chem. Mater. 2007, 19, 4592–4599. [Google Scholar]
- 32. Hoerder G. J., Seibald M., Baumann D., Schröder T., Peschke S., Schmid P. C., Tyborski T., Pust P., Stoll I., Bergler M., Patzig C., Reißaus S., Krause M., Berthold L., Höche T., Johrendt D., Huppertz H., Nat. Commun. 2019, 10, 1824. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.
Supplementary
