Powder X-ray diffraction as a powerful tool to exploit in organic electronics: shedding light on the first N,N′,N′′-tri­alkyl­diindolocarbazole

The first crystal structure of a fully N-alkyl­ated diindolocarbazole derivative, namely, 5,8,14-tri­butyl­diindolo[3,2-b;2′,3′-h]carbazole (1, C₃₆H₃₉N₃), has been determined from laboratory powder X-ray diffraction (PXRD) data. A com­plex trigonal structure with a high-volume unit cell of 12987 ų was found, with a very long a(=b) [52.8790 (14) Å] and a very short c [5.36308 (13) Å] unit-cell parameter (hexa­gonal setting). The detailed analysis of the inter­molecular inter­actions ob­served in the crystal structure of 1 highlights its potential towards the imple­mentation of this core as a semiconductor in organic thin-film transistor (OTFT) devices. Since the mol­ecule has a flat con­fig­uration reflecting its π-conjugated system, neighbouring mol­ecules are found to stack atop each other in a slipped parallel fashion via π–π stacking inter­actions between planes of ca 3.30 Å, with a centroid–centroid distance between the aromatic rings corresponding to the shortest axis of the unit cell (i.e. c). The alkyl­ation of the three N atoms proves to be a decisive feature since it favours the presence of C—H⋯π inter­actions in all directions, which strengthens the crystal packing. As a whole, PXRD proves to be a valuable option for the resolution of otherwise inaccessible organic crystal structures of inter­est in different areas.

1. Introduction

Organic semiconductors have become, in the optoelectronic field, potential alternatives to the conventional inorganic materials. Some of the enticing advantages associated with them are the easy modulation of their properties, their lower cost and the feasibility to fabricate large-area, lightweight and flexible displays (Bronstein et al., 2020; Yao et al., 2016). In fact, their applicability in devices such as organic thin-film tran­sistors (OTFTs) and organic light-emitting diodes (OLEDs) have been extensively demonstrated (Wang et al., 2017; Huang et al., 2019). Nevertheless, the demand for im­proved materials, especially in terms of charge transport and air stability, dictates further investigation.

The inter­molecular disposition of the semiconductor within the active layer is crucial in determining the performance of the device, but this can be challenging to predict beforehand in organic structures. Indeed, an appropriate arrangement that facilitates an effective charge transport between neighbouring mol­ecules is highly desired (Campbell et al., 2017). Thus, an exhaustive analysis of the crystal structure improves the com­prehension of such a process and paves the way towards subsequent structural optimizations or new enhanced designs.

Ladder-type mol­ecules, which feature large and planar aromatic cores, are particularly desirable for high-performing OTFTs because of their ease of establishing inter­molecular π–π stacking inter­actions (Chen et al., 2018; Cai et al., 2018; Bujaldón et al., 2021). A prime example of this type of material is the diindolo[3,2-b:2′,3′-h]carbazole core. In fact, similar structures, like indolo[3,2-b]carbazole and tri­indole, have excelled as air-stable semiconductors in OTFTs as a result of their advantageous inter­molecular dispositions (Reig et al., 2015). The diindolocarbazole core features an ionization potential of ∼5.1 eV that closely fits the Au work function, and an electron affinity of ∼2.5 eV (Srour et al., 2016). As reported, this preliminary data agrees with the possibility of a suitable hole injection and transport while also anti­cipating considerable long-term stability. However, the properties of this core integrated in optoelectronic devices remain unexplored. In terms of crystallographic studies, only two derivatives have been reported to date (Kawano et al., 2018; Wrobel et al., 2017), which were elucidated via conventional single-crystal X-ray diffraction (SXRD). Unfortunately, the aforementioned examples possess one or more unprotected N atoms, which is typically unfavourable in OTFTs (Reig et al., 2018). The herein studied diindolocarbazole derivative, 1, on the other hand, possesses a promising homogeneous alkyl­ation pattern with three butyl chains. Fig. 1 displays its chemical structure, as well as those of the two com­pounds already published, for com­parison. It should be mentioned also that the two previously described crystal structures incorporate different solvent mol­ecules into the packing as a result of the formation of the respective single crystals. This is inconvenient, since final devices usually require the absence of solvent within their active layers to function properly. Consequently, the acquisition of solvent-free crystal structures is crucial to compile realistic and unbiased information about the intermolecular disposition of the material for subsequent optoelectronic applications. The elucidation of the crystal structure of 1 was therefore sought-after information in moving towards the implementation of this core in OTFT devices.

Figure 1 The chemical structures of the elucidated diindolocarbazole derivative 1 and the two derivatives reported previously in the literature. The grey region delimitates the diindolocarbazole core, whereas the colour of the N atoms indicates whether they are alkyl­ated (green) or unprotected (red).

SXRD is the most widely used method for the determination of crystal structures due to the high success of this approach. In spite of numerous attempts, however, our efforts to produce single crystals of 1 suitable for analysis were fruitless. The use of a polycrystalline material for the diffraction experiment was consequently inevitable. Structure determination from powder diffraction (SDPD), while by no means routine, has emerged as a realistic option for determining crystal structures and has become an active research area (Meden et al., 2015). Besides, elucidation through this procedure does not involve the crystallization of the material in a solvent as opposed to SXRD. Hence, the resulting crystal structure is less prone to embed solvent molecules, which is a mandatory factor in this study.

Although single-crystal and powder diffraction patterns contain essentially the same information, the diffraction data are distributed in three-dimensional (3D) space or com­pressed into one dimension, respectively. As a consequence, there is usually considerable overlap of the peaks in the powder diffraction pattern, resulting in a loss of information. Whereas reflections are well resolved and can be measured individually in the single-crystal case, they can overlap partially or com­pletely in the observed powder diffraction pattern. The problem increases with the size of the unit cell, in the absence of strong scatterers and is generally greater when the symmetry is lower. In recent years, developments in high-resolution PXRD instrumentation, as well as com­putational resources and well-developed algorithms, have reached a level where SDPD has become a powerful tool for structural characterization (Černý, 2017).

Indeed, the number of organic structures solved annually from PXRD data has increased during the last decade. Fig. 2 shows the number of structures determined from PXRD data deposited in the Cambridge Structural Database (CSD; Groom et al., 2016) from 2010 to 2019. The criteria used in this search are the following: 3D coordinates determined, any R factor, not disordered, no errors and only organics. Although there is a growing number of crystal structures solved using PXRD, they represent less than 1% of the structures deposited in the CSD. Regarding the application of PXRD in organic electronics, the elucidation of the well-established asymmetrically-alkylated C₈-BTBT ([1]benzothieno[3,2-b]benzo­thio­phene) by Gbabode et al. (2014) and the recent advances from Ishii et al. (2020) towards a more straightforward approach to evaluate organic semiconductors are prime examples of its usefulness in this topic. SDPD, far from being automated, is an active research area where different software has been developed.

Figure 2 The number of organic crystal structures determined from PXRD data deposited in the CSD during the last decade.

Therefore, in the present study, we explore the crystal structure determination of 1 from PXRD data and analyse its inter­actions in order to gain new insight into this core as a promising semiconductor for OTFT devices.

2. Experimental

2.1. Sample preparation

Compound 1 was synthesized from 2,7-di­bromo-9H-carbazole following a reported method (Srour et al., 2016). Characterization data of 1, ¹H NMR (500 MHz, C₆D₆): δ (ppm) 8.32 (d, J = 7.8 Hz, 2H), 8.31 (s, 2H), 8.17 (s, 2H), 7.53 (ddd, J = 8.2, J = 7.1, J = 1.2 Hz, 2H), 7.34 (m, 2H), 7.30 (d, J = 8.1 Hz, 2H), 4.21 (t, J = 7.1 Hz, 2H), 4.07 (t, J = 7.1 Hz, 4H), 1.78 (m, 2H), 1.66 (m, 4H), 1.34–1.10 (m, 6H), 0.76 (t, J = 7.4 Hz, 3H), 0.70 (t, J = 7.3 Hz, 6H). ¹³C NMR (125 MHz, C₆D₆): δ (ppm) 142.4, 138.2, 136.3, 126.0, 123.8, 123.8, 123.7, 120.8, 118.5, 108.9, 99.3, 99.2, 43.5, 43.2, 31.3, 31.1, 21.0, 20.9, 14.1, 14.0. HRMS (ESI–MS) (m/z): calculated for C₃₆H₃₉N₃ M^+., 513.3138; found 513.3139. The subsequent PXRD measurements were carried out by introducing some of the powder material in a 0.5 mm diameter Hilgenberg glass capillary.

2.2. Instrument and experimental conditions

PXRD patterns were obtained on a PANalytical X'Pert PRO MPD diffractometer of radius 240 mm in a transmission con­fig­uration with a spinner glass capillary sample holder, using Cu K_α1+2 radiation (λ = 1.5418 Å) with a focalizing elliptic mirror and a PIXcel detector working at a maximum detector active length of 3.347 Å. Incident and diffracted beam 0.01 radians soller slits and incident beam slits defining a beam height of 0.4 mm were used with the sample placed in a glass capillary. 22 consecutive 2θ scans were measured and added from 2 to 70° in 2θ, with a step size of 0.013° and a measuring time of 700 s per step (total measuring time 88 h).

2.3. Structure determination from the powder X-ray diffraction approach

The SDPD procedure, irrespective of the software used, always consists of the steps shown in Fig. 3 (David & Shankland, 2008).

Figure 3 Structure determination from the powder X-ray diffraction (PXRD) procedure.

For a high quality data collection it is vital to ensure that the experimental conditions yield angular accuracy and a proper choice of slits, control over the morphology and the size of the sample, and the choice of transmission instead of reflection geometry avoid preferential orientation problems (Bergese et al., 2001).

The indexing process, i.e. analysis of the peak positions in the powder diffraction pattern in order to detect the correct unit-cell parameters and crystal system, is still a bottleneck, especially in cases of low-resolution powder patterns or for triclinic symmetry. Some of the programs used include DICVOL, TREOR (Werner et al., 1985), ITO (Visser, 1969), CRYSFIRE (Shirley, 2004) and DAJUST (Vallcorba et al., 2012). Our personal choice is the dichotomy algorithm, nicely implemented in the program DICVOL04 (Boultif & Louër, 2004). This allows automatic searching for cubic to monoclinic systems with the volume range estimated. The indexing of triclinic systems is also possible but more difficult. The final results are listed using the standard M20 and F20 figures of merit.

Firstly, the powder X-ray diffractogram of 1 was in principle well indexed to a monoclinic cell with the unit-cell parameters a = 5.36, b = 26.40, c = 15.35 Å, β = 96.6° and V = 2156 ų. Taking into account the unit-cell volume, the mol­ecular weight of 1 and an estimated density value of 1.2 Mg m⁻³, the number of mol­ecules in the unit cell was calculated to be Z = 3.

Indexing is followed by space-group determination, which is based on systematic extinction analysis. In the present case, the unit cell was primitive and a binary helicoidal axis was discarded due to the presence of the 010 reflection. Moreover, a-, c- and n-glide planes were also discarded as reflections such as 100, 001 and 20 were present. Therefore, the possible space groups for which structure solution should be attempted were P2, Pm and P2/m. Considering the multiplicity of these space groups, 1.5 would be the number of mol­ecules in the asymmetric unit for P2 and Pm, and 0.75 for P2/m.

At this stage, we recognized the possibility that this monoclinic cell could be a subcell of a higher symmetry and larger-volume cell. Thus, we went back to the indexing step and increased the maximum unit-cell parameters to 60 Å and the maximum volume to 15000 ų. The PXRD diffractogram was perfectly indexed to a hexa­gonal cell with the unit-cell parameters a = 52.80, c = 5.36 Å and V = 12933 ų. In fact, the largest parameter of this higher-symmetry cell was twice the largest parameter of the monoclinic cell and the shortest unit-cell parameters are the same. Now, given the new volume, the unit-cell content was set to 18 mol­ecules. On the basis of systematic absences, a rhombohedral R lattice fitted perfectly but the space group could not be assigned unambiguously since in principle all the possible trigonal groups, i.e. R3, R, R32, R3m, R3c, Rm and Rc, were com­patible with the observed reflections.

The aim of the next step, namely, so-called pattern matching or profile fitting, is to fit the com­plete experimental PXRD pattern by refinement of variables. These variables describe the peak positions (unit-cell parameters and zero-point shift parameter), the background intensity distribution and the peak intensities, widths and shapes. No structural model is used. The goal is to obtain reliable values of the variables that describe different features of the powder diffraction pattern in preparation for the subsequent stages of the structure determination process. Traditional approaches use peak intensities but many direct-space methods are based on com­parison of the com­plete powder diffraction pattern using a whole profile figure of merit, such as R_wp, and, in this case, the intensity data extracted from the powder diffraction pattern are not used. The two most common applied techniques for pattern matching are those developed by Pawley and Le Bail. In the present case, Pawley (1981) refinement by means of the TOPAS software (Version 6; Coelho et al., 2011; Coelho, 2018) was performed. The background was modelled with a 20th-order Chebyschev polynomial and a 1/X background function was applied. The instrumental contribution to the diffraction profile was calculated with the Fundamental Parameters Approach (Cheary et al., 2004). The peak width was modelled with the double-Voigt approach by considering only the Lorentzian contribution of the crystallite size and only the Gaussian contribution to the microstrain. Moreover, the Stephens (1999) model for anisotropic line broadening was applied. Fig. 4 shows a com­parison of the Pawley fits for the monoclinic [Fig. 4(a)] and for the trigonal [Fig. 4(b)] cells found in the previous indexing process. Both pattern matchings fit very well the com­plete experimental PXRD pattern, with the agreement factor for the trigonal cell being slightly better (1.53%) than the R_wp value for the monoclinic cell (1.87%).

Figure 4 Pattern matching Pawley fit plots for the (a) monoclinic and (b) trigonal cells; agreement factors: Rwp = 1.87% and Rwp = 1.53%, respectively. Both plots show the experimental PXRD profile (blue solid line), the calculated PXRD profile (red solid line) and the difference profile (grey, lower line). Black tick marks indicate the peak positions.

Structure solution procedures can be divided into two groups: reciprocal space methods and direct space methods (Černý & Favre-Nicolin, 2007). On the one hand, the traditionally used reciprocal space methods involve intensity extraction algorithms working on the reciprocal space. On the other hand, the so-called direct space methods or global optimization are based on pattern modelling algorithms working on the direct space and using the chemical knowledge from that space. Hybrid methods iterating between both spaces are known as well.

Nowadays, the most popular procedure to solve organic crystal structures from powder diffraction data is the direct space (also called real space) method. The space group, unit-cell parameters and an initial mol­ecular geometry of the com­ponents in the asymmetric unit are given as input. Starting from a random arrangement of mol­ecules, trial structures are generated in direct space by translating and rotating the mol­ecules within the unit cell, and the suitability of each trial structure is assessed by direct com­parison between the calculated and observed diffraction patterns (Harris et al., 2001). Intra­molecular degrees of freedom (DOF), such as rotations around single bonds, are also adjusted. Several global optimization techniques are used in direct space SDPD, such as Monte Carlo/Simulated Annealing or Parallel Tempering and Genetic Algorithm techniques (Černý, 2017). In powder diffraction, the most often used cost function is the weighted profile R_wp factor calculated over the whole powder pattern.

In the present case, the solution of the crystal structure of 1 was attempted using the direct space methodology for each of the possible space groups. The starting model for the crystal structure determination was optimized previously with the program SPARTAN (Young, 2001) and some constraints were introduced, considering the mol­ecule as a rigid body using the Z-matrix notation, which was allowed to rotate and translate in the three directions (3 orientational + 3 positional DOFs) within the cell. Planarity restraints were applied to the aromatic rings and nine torsion angles of the flexible aliphatic chains were refined as is shown in Fig. 5. H atoms were not included in the model at this stage.

Figure 5 Starting model of 1 used in the Simulated Annealing solution process in TOPAS. The torsion angles refined are indicated by arrows and three of them are specified in one of the alkyl chains.

The crystal structure was solved using the Global Optimization Simulated Annealing approach integrated in TOPAS. The best fit to the experimental PXRD data was found for the space group R; this solution was also chemically sensible (Table 1).

Table 1
Structure solution for the possible space groups

Space group	Multiplicity	Z/Z′	Rwp	Chemical sense
R3	9	18/2	47.9	No
R	18	18/1	15.4	Yes
R32	18	18/1	44.9	No
R3m	18	18/1	37.7	No
R3c	18	18/1	37.9	No
Rm	36	18/0.5	37.4	No
Rc	36	18/0.5	34.2	No

The crystal structure was subsequently refined by the Rietveld (1969) method, also by means of TOPAS software, giving satisfying results with low R_wp values (Toby, 2006). In this final step, the variables that define both the structural model and the powder diffraction profile are adjusted by least-squares methods in order to obtain the optimal fit between the experimental and calculated powder X-ray diffractograms. H atoms attached to C atoms were placed at calculated positions.

Several attempts were made to refine the isotropic displacement parameters. Two different B_eq values were defined, one for the 27 core non-H atoms plus three attached to core C atoms, and another for the remaining nine alkyl C atoms. After refinement of these parameters, either by leaving them free or by setting them to specific values, worse R_wp and R_Bragg values were always obtained. Finally, B_eq values were set to 3 Ų for C and N, and 10 Ų for H atoms, to obtain the lowest R_wp value, with a meaningful B_eq value (i.e. not a negative value). Furthermore, a preferred orientation correction, the eighth-order spherical harmonics function (Whitfield, 2009), was applied, as implemented in the TOPAS software.

During the Rietveld process, the checkCIF report (Spek, 2020; https://checkcif.iucr.org/) was used to examine the consistency and integrity of the proposed crystal structure. Some inter­molecular inter­actions involving the central alkyl chain were determined to be too short, specifically those labelled in Fig. 6. Thus, a feasible correction was to rotate the angle involving the three terminal C atoms of the alkyl chain.

Figure 6 The crystal structure determined for 1 after initial Rietveld refinement but before angle refinement. The labelled atoms indicate the angle that should be modified.

Therefore, the opposite angle involving C56—C59—C62 was calculated by subtracting its initial value from 360°, giving a value of 246.91°, which was introduced in the corresponding Z-matrix used to define the rigid body. Fig. 7 shows the new conformation obtained after the angle rotation which eliminates the checkCIF alerts derived from the exceedingly short inter­molecular contacts.

Figure 7 The crystal structure of 1 after angle rotation during Rietveld refinement.

3. Results and discussion

Compound 1 crystallizes in the trigonal space group R with a high-volume unit cell of 12987.1 (8) ų and 18 mol­ecules inside it. The final Rietveld plot of the experimental and calculated data is shown in Fig. 8. The TOPAS Rietveld refinement gives an R_wp value of 4.82% that matches very favourably with the Pawley R_wp value of 1.53%. Table 2 summarizes the most relevant parameters of the crystal structure determination and refinement of 1.

Table 2
Crystal data and structure refinement parameters of 1

Empirical formula	C36H39N3
Formula weight	513.71
Temperature (K)	298
Wavelength (Å)	1.54180
Crystal system	Trigonal
Space group	R
a, b, c (Å)	52.8790 (14), 52.8790 (14), 5.36308 (13)
Volume (Å^3)	12987.1 (8)
Z, Z′	18, 1
Density (calc.) (Mg m^−3)	1.182
Measured 2θ range	2.007 to 69.9840
Stepsize (°)	0.013
Measured data points	5230
Rietveld refinement details
Profile function	Double-Voigt
2θ range used (°)	3.2 to 69.98
No. reflections	1257
Data points	5138
Parameters	62
Rwp (%)	4.82
Rp (%)	3.54
RBragg (%)	2.11
GoF	4.05
CCDC deposition No.	2115611

Figure 8 (a) Final Rietveld plot for the crystal structure refinement of 1; agreement factors: Rwp = 4.82% and Rp = 3.54%. The plot shows the experimental PXRD profile (blue solid line), the calculated PXRD profile (red solid line) and the difference profile (grey, lower line). Black tick marks indicate the peak positions.

Fig. 9 shows the packing of the crystal structure of com­pound 1. Compound 1 lacks hydrogen-bond donor and acceptor groups. Instead, its main backbone consists of an aromatic system com­posed by four benzene rings fused with three inter­calated pyrrole heterocycles. Thus, the stabilizing impact of the aromatic inter­actions on this structure is expected to be maximal, and certainly structure-directing. Fig. 10 shows the π–π and C—H⋯π inter­actions established between the aro­matic systems of contiguous mol­ecules along the c axis. Since the mol­ecule has a flat con­fig­uration derived from its π-con­jugated system, neighbouring mol­ecules assemble atop each other in a parallel-displaced fashion via π–π stacking inter­actions. Indeed, the shortest π–π stacking distance found corresponds to 3.30 Å, which should grant an adequate overlap and charge transport along the material layers. The centroid–centroid distance between molecules of 5.36 Å corresponds to the length of the c axis. The angle between the planes of the aromatic rings of 0° and the lateral offset relative to one another of 4.23 Å leads to a favourable offset π-stacked geometry as electronic repulsion dominates in a face-to-face π-stacked geometry (Hunter & Sanders, 1990). Considering the packing motifs typically seen in polycyclic aromatic hydro­carbons (Desiraju & Gavezzotti, 1989), the arrangement of 1 can be related to a γ type, which is mainly governed by aromatic C⋯C (and C⋯N in this case) inter­actions between parallel-translated mol­ecules. On the other hand, materials such as penta­cene (Campbell et al., 1962) or indolo[3,2-b]carbazole (Reig et al., 2015), which is a less π-extended analogue of 1, arrange in a herringbone packing characterized by edge-to-face C—H⋯π inter­actions. Despite their excellence in organic electronics, the herringbone motif appears as non-optimal for charge transport (Campbell et al., 2017) because of the generally smaller direct π–π orbital overlap (Yao et al., 2018). Consequently, the elongation of the aromatic system from indolo[3,2-b]carbazole to diindolocarbazole seemingly results in a favourable, more flattened, γ disposition. From a structural point of view, these results support the potential of com­pound 1 as a semiconductor for OTFT devices.

Figure 9 Packing of 1 along the c axis showing the 18 mol­ecules in the unit cell.

Figure 10 Strong π–π parallel-displaced stackings (shortest inter­molecular π–π distance = 3.30 Å) and C—H⋯π inter­actions (C⋯centroid distances in blue: 3.39, 3.42 and 3.45 Å) between parallel mol­ecules of 1 along the c axis.

Another inter­esting point relates to the role of the alkyl chains, which frequently determine the inter­molecular arrange­ment. In fact, alkyl­ation is often found to suppress the herringbone packing in favour of a more π-stacked one (Klues & Witte, 2018). Even in the above-mentioned example of indolo[3,2-b]carbazole, the variation of the N-alkyl­ation pattern is able to modulate and shift the inter­molecular packing of the core (Zhao et al., 2012). In the case of 1, the presence of C—H⋯π inter­actions is detected between the alkyl chains and the aromatic systems of contiguous diindolocarbazole mol­ecules along the c axis. Specifically, all three methyl­ene groups adjacent to the N atoms provide C—H⋯π inter­actions, thus reinforcing the packing in this direction.

On the other hand, additional C—H⋯π inter­actions are also formed between mol­ecules belonging to the same plane. In this case, the inter­actions involve two of the three butyl chains: specifically, two methyl­ene and one terminal methyl fragment establish inter­molecular inter­actions with the aro­matic rings of surrounding mol­ecules (Fig. 11). Thus, the alkyl­­ation of all three N atoms increases the number of inter­actions in all directions and strengthens the crystal packing. As a whole, it can be concluded that this packing satisfies the required aromatic inter­actions particularly well for the subsequent application of 1 in optoelectronic devices.

Figure 11 C—H⋯π inter­actions (H⋯centroid distances in Å) observed among various mol­ecules of 1 in the (001) plane.

4. Conclusions

Compound 1 has a ladder-type construction based on the diindolocarbazole core, which features a highly π-conjugated system and three butyl chains bonded to the nitro­gen heteroatoms. Its crystal structure, which was inaccessible through conventional solution methods, has been finally determined from PXRD data. The structural solution proved to be a challenging process, mainly due to the large-volume cell range needed to find the proper trigonal cell, i.e. the limiting step in the SDPD process. The space group R afforded the best fit to the experimental PXRD data, as determined using the Global Optimization Simulated Annealing approach integrated in TOPAS. Finally, Rietveld refinement provided a good agreement with an R_wp value of 4.82%. The analysis of the crystal structure of 1 shows that this linear coplanar mol­ecular structure adopts an inter­esting π-stacked arrangement along the c axis with a small π–π stacking distance as close as 3.30 Å, anti­cipating great potential with respect to organic electronics. The presence of the three N-alkyl chains also increases the number of inter­molecular inter­actions in all directions, strengthening the packing. Overall, the structural design of 1 leads to a very appropriate mol­ecular arrangement that should grant an effective charge transport. Therefore, the elucidation of organic crystal structures provides crucial information regarding the potential of such materials. In this way, the solution via PXRD is confirmed as a helpful alternative towards the understanding, evaluation and eventual application of materials of inter­est in organic electronics in which conventional SXRD is not available.

Supporting information

Computing details

Data reduction: Fullprof; program(s) used to solve structure: Topas 6; program(s) used to refine structure: Topas 6.

5,8,14-Tributyldiindolo[3,2-b;2',3'-h]carbazole

Crystal data

C36H39N3	Z = 18
Mr = 513.71	Dx = 1.182 Mg m^−3
Trigonal, R3	Cu Kα radiation
a = 52.8790 (14) Å	T = 298 K
c = 5.36308 (13) Å	yellow
V = 12987.1 (8) Å^3	Specimen preparation: Prepared at 298 K

Data collection

Panalytical X'pert Pro MPD diffractometer	Data collection mode: transmission
Radiation source: sealed X-ray tube	2θmin = 3.2°, 2θmax = 69.984°, 2θstep = 0.013°
Specimen mounting: glass capillary

Refinement

Least-squares matrix: full	62 parameters
Rwp = 0.048	H atoms treated by a mixture of independent and constrained refinement
RBragg = 0.035	1/Yobs
5138 data points	(Δ/σ)max = 0.001
Profile function: Double-Voigt

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Biso*/Beq
C4	0.25884 (10)	0.28262 (16)	1.0467 (13)	3
C2	0.28231 (12)	0.30912 (19)	0.9647 (14)	3
C7	0.24198 (12)	0.28231 (18)	1.2545 (14)	3
C15	0.24649 (13)	0.25272 (19)	0.9638 (13)	3
C5	0.28879 (16)	0.3348 (3)	1.0886 (14)	3
C3	0.24841 (16)	0.3081 (2)	1.3789 (14)	3
N19	0.22020 (14)	0.2537 (2)	1.3012 (15)	3
C16	0.22258 (15)	0.2354 (2)	1.1263 (15)	3
C12	0.25384 (17)	0.2398 (2)	0.7710 (14)	3
C6	0.27194 (18)	0.3342 (3)	1.2936 (14)	3
C40	0.19824 (19)	0.2439 (4)	1.4966 (19)	3
C13	0.2057 (2)	0.2052 (4)	1.1024 (17)	3
C14	0.2369 (2)	0.2096 (3)	0.7458 (14)	3
C43	0.16733 (16)	0.2349 (3)	1.404 (3)	3
C17	0.2132 (2)	0.1923 (4)	0.9102 (16)	3
N38	0.2391 (3)	0.1913 (4)	0.5703 (15)	3
C46	0.1665 (2)	0.2539 (6)	1.195 (5)	3
C25	0.2011 (3)	0.1622 (4)	0.8313 (18)	3
C24	0.2180 (3)	0.1626 (4)	0.6217 (17)	3
C53	0.2617 (3)	0.2004 (4)	0.3819 (17)	3
C49	0.1361 (3)	0.2507 (7)	1.160 (6)	3
C22	0.1781 (4)	0.1359 (6)	0.921 (2)	3
C21	0.2125 (4)	0.1371 (6)	0.4993 (18)	3
C56	0.2897 (4)	0.2014 (6)	0.474 (2)	3
C26	0.1726 (4)	0.1105 (6)	0.798 (2)	3
C23	0.1894 (5)	0.1108 (6)	0.589 (2)	3
C59	0.3153 (4)	0.2178 (6)	0.298 (3)	3
N39	0.1514 (5)	0.0815 (7)	0.850 (3)	3
C32	0.1774 (5)	0.0807 (7)	0.511 (2)	3
C62	0.3287 (8)	0.1997 (9)	0.211 (7)	3
C66	0.1289 (5)	0.0724 (8)	1.040 (3)	3
C33	0.1539 (6)	0.0635 (8)	0.676 (3)	3
C29	0.1842 (6)	0.0671 (8)	0.320 (2)	3
C69	0.1046 (5)	0.0791 (8)	0.973 (3)	3
C30	0.1372 (6)	0.0331 (9)	0.650 (3)	3
C31	0.1678 (7)	0.0369 (9)	0.295 (3)	3
C72	0.0893 (5)	0.0647 (9)	0.729 (4)	3
C34	0.1445 (7)	0.0201 (10)	0.458 (3)	3
C75	0.0562 (5)	0.0483 (15)	0.751 (6)	3
H1	0.0952 (8)	0.0801 (13)	0.601 (4)	10
H2	0.0956 (9)	0.0507 (12)	0.678 (6)	10
H3	0.1334 (7)	−0.0007 (10)	0.437 (3)	10
H4	0.0500 (8)	0.033 (2)	0.878 (7)	10
H5	0.0475 (6)	0.0395 (15)	0.587 (7)	10
H6	0.0495 (7)	0.0620 (15)	0.800 (8)	10
H7	0.1724 (7)	0.0274 (9)	0.164 (3)	10
H8	0.1213 (7)	0.0216 (10)	0.760 (3)	10
H9	0.29382 (13)	0.3096 (2)	0.8245 (15)	10
H10	0.3049 (2)	0.3531 (3)	1.0336 (16)	10
H11	0.1975 (3)	0.2268 (4)	1.580 (2)	10
H12	0.2043 (2)	0.2600 (4)	1.618 (2)	10
H13	0.23695 (19)	0.3079 (3)	1.5187 (16)	10
H14	0.26989 (18)	0.2512 (2)	0.6608 (14)	10
H15	0.2767 (2)	0.3521 (3)	1.3769 (14)	10
H16	0.1807 (3)	0.2746 (7)	1.233 (8)	10
H17	0.1722 (2)	0.2483 (8)	1.037 (4)	10
H18	0.1996 (6)	0.0781 (8)	0.212 (2)	10
H19	0.1580 (2)	0.2146 (4)	1.344 (4)	10
H20	0.1560 (2)	0.2361 (5)	1.547 (3)	10
H21	0.1223 (2)	0.2309 (6)	1.121 (4)	10
H22	0.1368 (4)	0.2634 (9)	1.022 (9)	10
H23	0.1303 (4)	0.2565 (6)	1.316 (8)	10
H24	0.0899 (5)	0.0717 (9)	1.108 (4)	10
H25	0.1132 (5)	0.1005 (8)	0.956 (4)	10
H26	0.1897 (2)	0.1937 (4)	1.2130 (19)	10
H27	0.2538 (4)	0.1866 (6)	0.240 (2)	10
H28	0.2669 (4)	0.2202 (5)	0.324 (3)	10
H29	0.1201 (6)	0.0512 (9)	1.065 (3)	10
H30	0.1382 (5)	0.0828 (8)	1.197 (3)	10
H31	0.3084 (4)	0.2238 (7)	0.150 (4)	10
H32	0.3305 (4)	0.2354 (10)	0.384 (5)	10
H33	0.2240 (4)	0.1375 (6)	0.3593 (19)	10
H34	0.1665 (4)	0.1354 (6)	1.061 (2)	10
H35	0.2949 (4)	0.2110 (7)	0.639 (3)	10
H36	0.2858 (6)	0.1811 (7)	0.491 (5)	10
H37	0.3359 (10)	0.1937 (11)	0.357 (9)	10
H38	0.3452 (8)	0.2115 (11)	0.097 (6)	10
H39	0.3138 (11)	0.1822 (14)	0.122 (9)	10

Geometric parameters (Å, º)

C4—C2	1.40 (1)	C49—H22	0.99 (6)
C4—C7	1.422 (9)	C49—H23	0.99 (5)
C4—C15	1.45 (1)	C22—C26	1.39 (3)
C2—C5	1.39 (1)	C22—H34	0.96 (2)
C2—H9	0.96 (1)	C21—C23	1.40 (4)
C7—C3	1.40 (1)	C21—H33	0.96 (2)
C7—N19	1.39 (1)	C56—C59	1.52 (3)
C15—C16	1.43 (1)	C56—H35	0.99 (3)
C15—C12	1.40 (1)	C56—H36	0.99 (4)
C5—C6	1.41 (1)	C26—C23	1.43 (2)
C5—H10	0.96 (2)	C26—N39	1.40 (4)
C3—C6	1.39 (1)	C23—C32	1.45 (4)
C3—H13	0.96 (1)	C59—C62	1.52 (3)
N19—C16	1.40 (1)	C59—H31	0.99 (3)
N19—C40	1.45 (1)	C59—H32	0.99 (4)
C16—C13	1.39 (2)	N39—C66	1.45 (3)
C12—C14	1.39 (2)	N39—C33	1.39 (4)
C12—H14	0.96 (1)	C32—C33	1.42 (3)
C6—H15	0.96 (2)	C32—C29	1.40 (3)
C40—C43	1.54 (1)	C62—H37	0.99 (5)
C40—H11	0.99 (2)	C62—H38	0.99 (6)
C40—H12	0.99 (2)	C62—H39	0.99 (7)
C13—C17	1.40 (2)	C66—C69	1.54 (3)
C13—H26	0.96 (1)	C66—H29	0.98 (6)
C14—C17	1.43 (1)	C66—H30	0.99 (3)
C14—N38	1.40 (1)	C33—C30	1.40 (6)
C43—C46	1.52 (3)	C29—C31	1.39 (6)
C43—H19	0.99 (3)	C29—H18	0.93 (4)
C43—H20	0.99 (2)	C69—C72	1.53 (3)
C17—C25	1.45 (2)	C69—H24	0.99 (3)
N38—C24	1.39 (3)	C69—H25	0.99 (5)
N38—C53	1.45 (2)	C30—C34	1.40 (4)
C46—C49	1.54 (2)	C30—H8	0.96 (4)
C46—H16	0.99 (4)	C31—C34	1.41 (4)
C46—H17	0.99 (4)	C31—H7	0.96 (3)
C25—C24	1.43 (1)	C72—C75	1.52 (3)
C25—C22	1.40 (3)	C72—H1	0.99 (6)
C24—C21	1.39 (3)	C72—H2	0.99 (6)
C53—C56	1.54 (2)	C34—H3	0.96 (6)
C53—H27	0.99 (2)	C75—H4	0.99 (9)
C53—H28	0.99 (3)	C75—H5	0.99 (5)
C49—H21	0.95 (4)	C75—H6	0.99 (9)
C2—C4—C7	119.7 (6)	H22—C49—H23	110 (4)
C2—C4—C15	133.5 (6)	C25—C22—C26	117 (2)
C7—C4—C15	106.9 (6)	C25—C22—H34	121 (2)
C4—C2—C5	119.3 (7)	C26—C22—H34	121 (2)
C4—C2—H9	120.3 (8)	C24—C21—C23	117 (2)
C5—C2—H9	120.4 (9)	C24—C21—H33	122 (2)
C4—C7—C3	121.1 (6)	C23—C21—H33	121 (2)
C4—C7—N19	109.1 (6)	C53—C56—C59	113 (1)
C3—C7—N19	129.7 (7)	C53—C56—H35	109 (2)
C4—C15—C16	106.5 (6)	C53—C56—H36	109 (2)
C4—C15—C12	132.8 (7)	C59—C56—H35	109 (2)
C16—C15—C12	120.8 (7)	C59—C56—H36	109 (2)
C2—C5—C6	120.4 (8)	H35—C56—H36	109 (3)
C2—C5—H10	120 (1)	C22—C26—C23	122 (2)
C6—C5—H10	120 (1)	C22—C26—N39	129 (2)
C7—C3—C6	118.0 (7)	C23—C26—N39	109 (2)
C7—C3—H13	121.0 (9)	C21—C23—C26	121 (2)
C6—C3—H13	121 (1)	C21—C23—C32	133 (2)
C7—N19—C16	108.6 (7)	C26—C23—C32	107 (2)
C7—N19—C40	126.7 (8)	C56—C59—C62	113 (2)
C16—N19—C40	124.7 (8)	C56—C59—H31	108 (2)
C15—C16—N19	108.9 (7)	C56—C59—H32	109 (3)
C15—C16—C13	122.2 (8)	C62—C59—H31	109 (3)
N19—C16—C13	128.9 (9)	C62—C59—H32	109 (3)
C15—C12—C14	117.1 (8)	H31—C59—H32	109 (3)
C15—C12—H14	121.4 (9)	C26—N39—C66	125 (2)
C14—C12—H14	121 (1)	C26—N39—C33	109 (2)
C5—C6—C3	121.5 (8)	C66—N39—C33	126 (2)
C5—C6—H15	119 (1)	C23—C32—C33	107 (2)
C3—C6—H15	119 (1)	C23—C32—C29	134 (2)
N19—C40—C43	114 (1)	C33—C32—C29	120 (2)
N19—C40—H11	108 (1)	C59—C62—H37	110 (4)
N19—C40—H12	108 (1)	C59—C62—H38	110 (4)
C43—C40—H11	108 (1)	C59—C62—H39	109 (4)
C43—C40—H12	108 (1)	H37—C62—H38	109 (5)
H11—C40—H12	110 (2)	H37—C62—H39	110 (5)
C16—C13—C17	117 (1)	H38—C62—H39	109 (5)
C16—C13—H26	122 (1)	N39—C66—C69	114 (2)
C17—C13—H26	121 (1)	N39—C66—H29	108 (3)
C12—C14—C17	122.0 (9)	N39—C66—H30	108 (3)
C12—C14—N38	129.1 (9)	C69—C66—H29	109 (3)
C17—C14—N38	108.8 (9)	C69—C66—H30	108 (3)
C40—C43—C46	114 (1)	H29—C66—H30	110 (3)
C40—C43—H19	109 (2)	N39—C33—C32	109 (2)
C40—C43—H20	108 (2)	N39—C33—C30	129 (2)
C46—C43—H19	108 (2)	C32—C33—C30	121 (2)
C46—C43—H20	108 (2)	C32—C29—C31	119 (2)
H19—C43—H20	109 (2)	C32—C29—H18	120 (3)
C13—C17—C14	121 (1)	C31—C29—H18	120 (3)
C13—C17—C25	132 (1)	C66—C69—C72	112 (2)
C14—C17—C25	107 (1)	C66—C69—H24	109 (3)
C14—N38—C24	109 (1)	C66—C69—H25	109 (3)
C14—N38—C53	125 (1)	C72—C69—H24	108 (3)
C24—N38—C53	125 (1)	C72—C69—H25	108 (3)
C43—C46—C49	113 (2)	H24—C69—H25	110 (3)
C43—C46—H16	109 (3)	C33—C30—C34	118 (3)
C43—C46—H17	109 (2)	C33—C30—H8	121 (3)
C49—C46—H16	109 (3)	C34—C30—H8	121 (3)
C49—C46—H17	109 (2)	C29—C31—C34	121 (3)
H16—C46—H17	109 (3)	C29—C31—H7	120 (3)
C17—C25—C24	106 (1)	C34—C31—H7	120 (3)
C17—C25—C22	133 (1)	C69—C72—C75	113 (3)
C24—C25—C22	121 (1)	C69—C72—H1	108 (3)
N38—C24—C25	109 (1)	C69—C72—H2	108 (3)
N38—C24—C21	129 (1)	C75—C72—H1	109 (4)
C25—C24—C21	122 (1)	C75—C72—H2	108 (4)
N38—C53—C56	114 (1)	H1—C72—H2	110 (4)
N38—C53—H27	108 (1)	C30—C34—C31	121 (3)
N38—C53—H28	108 (1)	C30—C34—H3	119 (3)
C56—C53—H27	108 (2)	C31—C34—H3	119 (3)
C56—C53—H28	108 (2)	C72—C75—H4	110 (5)
H27—C53—H28	109 (2)	C72—C75—H5	109 (5)
C46—C49—H21	109 (3)	C72—C75—H6	110 (5)
C46—C49—H22	110 (3)	H4—C75—H5	110 (6)
C46—C49—H23	109 (3)	H4—C75—H6	109 (6)
H21—C49—H22	109 (4)	H5—C75—H6	109 (6)
H21—C49—H23	110 (3)