One trityl, two trityl, three trityl groups – Structural differences of differently substituted triethanolamines

2.1 General

NMR spectra were recorded on Bruker Avance(III) 400 and Avance(Neo) 400 instruments. The NMR spectra (¹H, ¹³C) were referenced to the residual solvent peaks (CD₂Cl₂). The signals in the NMR spectrum have been assigned based on the corresponding 2D NMR spectra. The annotation is based on the connectivity of the atoms of the respective signals. These atoms are italicized in the annotation. In addition, the assignment of the aromatic atoms and the oxygen atoms has been indicated by subscripts (o: ortho, m: meta, p: para, ipso: ipso, trityl: trityl ether group, hydroxy: hydroxy group). Elemental analysis was conducted using a Vario EL III CHNS Analyzer. Mass spectra were recorded using a Bruker impact II spectrometer.

Single crystals suitable for X-ray analysis were obtained by slow evaporation of a concentrated solution of the compound in different solvents: For compound 1, petroleum ether:ethylacetate:triethylamine (80:20:3) at room temperature and for compound 3 ethylacetate:triethylamine (100:3) at room temperature were used. Compound 2 was crystallized by slowly cooling a solution in diethyl ether to 4 °C.

2.2 Synthesis of compounds 1 and 2

A solution of triethanolamine (0.99 g, 6.6 mmol, 1.0 eq.) in dichloromethane (40 mL) was prepared, and subsequently, N,N-diisopropylethylamine (2.25 mL, 13.3 mmol, 2.0 eq.) and solid triphenylmethyl chloride (3.70 g, 13.3 mmol, 2.0 eq.) were added to the solution. The reaction mixture was stirred for 4 h at room temperature. Following filtration, the solvent was removed under reduced pressure and the residue purified by column chromatography (silica gel, petroleum ether:ethyl acetate:triethylamine, 80:20:3). Compound 1 (2.23 g, 2.54 mmol) and compound 2 (1.75 g, 2.76 mmol) were obtained sequentially as off-white solids.

Compound 1: Yield: 38 %. R_f = 0.84 (petroleum ether:ethylacetate:triethylamine 80:20:3), ¹H NMR (400 MHz, CD₂Cl₂) δ (ppm): 7.46–7.33 (m, 18H, CH _o), 7.29–7.17 (m, 27H, CH _m, CH _p), 3.06 (t, J = 6.2 Hz, 6H, CCH ₂O), 2.71 (t, J = 6.1 Hz, 6H, NCH ₂C). ¹³C{¹H} NMR (101 MHz, CD₂Cl₂) δ (ppm): 144.9 (C _ipso), 129.0 (C _o), 128.1 (C _m), 127.3 (C _p), 87.0 (OCC₃), 63.4 (CCH₂O), 56.0 (NCH₂C). ESI-MS (m/z): [M+H]⁺ calcd: 876.4411; found: 876.4417. EA (%): C₆₃H₅₇NO₃ calcd: C 86.4, H 6.6, N 1.6, found: C 86.2, H 6.6, N 1.4.

Compound 2: Yield: 42 %. R_f = 0.41 (petroleum ether:ethylacetate:triethylamine 80:20:3), ¹H NMR (400 MHz, CD₂Cl₂) δ (ppm): 7.55–7.37 (m, 12H, CH _o), 7.37–7.17 (m, 18H, CH _m, CH _p), 3.47 (t, J = 5.3 Hz, 2H, CCH ₂O_Hydroxy), 3.17 (t, J = 5.9 Hz, 4H, CCH ₂O_Trityl), 2.75 (t, J = 5.9 Hz, 4H, NCH ₂C_OTrityl), 2.64 (t, J = 5.3 Hz, 2H, NCH ₂C_OHydroxy). ¹³C{¹H} NMR (101 MHz, CD₂Cl₂) δ (ppm): 144.9 (C _ipso), 129.0 (C _o), 128.2 (C _m), 127.4 (C _p), 87.2 (OCC₃), 62.8 (CCH₂O_Trityl), 59.3 (CCH₂O_Hydroxy), 57.3 (NCH₂C_OHydroxy), 54.7 (NCH₂C_OTrityl). ESI-MS (m/z): [M+H]⁺ calcd: 634.3316; found: 634.3322. EA (%): C₄₄H₄₃NO₃ calcd: C 83.4, H 6.8, N 2.2, found: C 83.2, H 6.9, N 2.0.

2.3 Synthesis of compound 3

Triethanolamine (5.97 g, 40.0 mmol, 1.0 eq.) was dissolved in dichloromethane (80 mL) and N,N-diisopropylethylamine (1.7 mL, 10 mmol, 0.25 eq.) was added to the solution. Subsequently, solid triphenylmethyl chloride (2.79 g, 10.0 mmol, 0.25 eq.) was added in four portions to the solution and the reaction mixture was stirred overnight at room temperature. After filtration, the solvent volume was reduced by half and the organic phase was washed with water (5·60 mL). The organic layer was then dried over magnesium sulfate, the solvent removed under reduced pressure and the residue purified by column chromatography (silica gel, ethyl acetate:triethylamine, 100:3). Compound 3 (1.98 g, 5.07 mmol) was obtained as an off-white solid.

Yield: 51 %. R_f = 0.30 (ethylacetate:triethylamine 100:3), ¹H NMR (400 MHz, CD₂Cl₂) δ (ppm): 7.49–7.43 (m, 6H, CH _o), 7.37–7.29 (m, 6H, CH _m), 7.29–7.23 (m, 3H, CH _p), 3.53 (t, J = 5.4 Hz, 4H, CCH ₂O_Hydroxy), 3.20 (t, J = 5.6 Hz, 2H, CCH ₂O_Trityl), 2.77 (t, J = 5.6 Hz, 2H, NCH ₂CO_Trityl), 2.65 (t, J = 5.4 Hz, 4H, NCH ₂CO_Hydroxy). ¹³C{¹H} NMR (101 MHz, CD₂Cl₂) δ (ppm): 144.6 (C _ipso), 129.0 (C _o), 128.3 (C _m), 127.4 (C _p), 87.5 (OCC₃), 62.6 (CCH₂O_Trityl), 60.1 (CCH₂O_Hydroxy), 57.1 (NCH₂CO_Hydroxy), 54.7 (NCH₂CO_Trityl). ESI-MS (m/z): [M+H]⁺ calcd: 392.2220; found: 392.2234. EA (%): C₂₅H₂₉NO₃ calcd: C 76.7, H 7.5, N 3.6, found: C 76.8, H 7.5, N 3.5.

2.4 Single crystal X-ray diffractometry

X-ray diffraction analysis and data collection (Table 2) were performed using Mo-Kα radiation (λ = 0.71073 Å, graphite monochromator) on a Bruker Venture D8 instrument equipped with a microsource and a Photon III CMOS detector. Initial structure solutions were obtained using the SHELXT ¹⁴ package via intrinsic phasing and refined with SHELXL ¹⁵ against all |F²| using initially isotropic and subsequently anisotropic thermal parameters. Anisotropic refinement was applied to all non-hydrogen atoms, while hydrogen atoms were positioned at ideal coordinates based on calculated positions and refined using the riding model for compound 1 and 3. In the case of compound 2 the hydrogen atoms bound to heteroatoms have been refined independently and all hydrogen atoms bound to carbon atoms have been placed at ideal positions and refined using the riding model. Parts of the structure of compound 3 are disordered. The atoms C17A and associated atoms are disordered over two positions in a ratio of 0.54:046, and the atoms C24A and associated atoms are disordered over three positions in a ratio of 0.50:0.22:0.28. The disordered atoms have all been refined anisotropically. The graphical representation and extended evaluation of all structures was carried out using the software Diamond 4.6.1. ¹⁶

Scheme 1:

Protection of triethanolamine: Triple-, double- and single-protected product. a) 2.0 eq. triphenylmethyl chloride, 2.0 eq. DIPEA, DCM, rt, 4 h; b) 0.25 eq. triphenylmethyl chloride, 0.25 eq. DIPEA, DCM, rt, 18 h.

3.1 Synthesis

The protection of the hydroxyl functions of triethanolamine was conducted under basic conditions utilizing trityl chloride (Scheme 1). The ratio of trityl chloride to triethanolamine was found to influence the ratio of the obtained protected products 1, 2, and 3. While both the triple-protected product 1 and the double-protected product 2 were obtained with two equivalents of trityl chloride, no single-protected product could be isolated under these conditions. The targeted synthesis of the single-protected product 3 succeeded in highly dilute reaction conditions and by adding sub-stoichiometric amounts of trityl chloride in several portions.

3.2 Crystal structure description of 1

The triple-protected product 1 crystallizes in the monoclinic space group P2₁/n (Figure 1). The molecule adopts a triskelion-like conformation at the central nitrogen atom. This conformation has been observed for several trialkylamines, consistent with optimal packing conditions for bulky groups. ⁷

Figure 1:

Molecular structure of 1 with corresponding numbering of the atoms (displacement ellipsoids are shown at the 50 % probability level and H atoms are omitted for clarity).

An examination of the structural parameters that describe the pyramidalization of the nitrogen atom reveals that the trigonal pyramid is less elevated (0.3899(14) Å) and that the sum of the C–N–C bond angles (339.59(19)°) is larger than that observed for triethylamine. However, these values remain below those characteristics of a planar nitrogen atom. The average bond length of the C–N bonds of compound 1 (1.47(3) Å) is also similar to that observed in triethylamine.

The oxygen atoms and their corresponding alkyl chains are positioned in an eccentric manner in relation to the central nitrogen atom. In contrast, the oxygen atom O1 and its alkyl chain are oriented at a slight angle, with the oxygen atom directed towards the central nitrogen atom (Figure 2). This is due to the presence of a weak, but stabilizing hydrogen bond interaction with an adjacent methylene group C22 (hydrogen bond geometry (C22–H22A⋯O1) d(D–H): 0.99 Å, d(H⋯A): 2.42 Å, <DHA: 122°, d(D⋯A): 3.055(2) Å, A: O1). In comparison to the typical C–H⋯O interaction observed between a hydrogen atom of a methyl group and an oxygen atom (d(D···A): 3.590(7) Å, <DHA: 137°), ¹⁷ the interaction in the case of 1 is stronger and more directed, which ultimately influences the orientation of the alkyl chain.

Figure 2:

Molecular detail of 1 with highlighted hydrogen bond of C22–H22A and O1 (displacement ellipsoids are shown at the 50 % probability level, except for H22A which is shown at an arbitrary radius, and H atoms, except for H22A, are omitted for clarity).

The intermolecular interaction is primarily facilitated by the trityl groups. The heteroatoms and acidic protons of the molecules are effectively shielded by the protecting groups, thereby preventing any intermolecular interaction. In addition to π-π interactions, sp²-CH/π interactions ¹⁸ occur (Figure 3), which are characterized by the spatial proximity of the C–H protons and the adjacent π-systems (2.7071(10) Å). The distance is not significantly shorter than the average distance of the hydrogen atom and a π-system in typical sp²-CH/π interactions (2.73(13) Å), ¹⁸ indicating a medium-strong interaction in the crystal of 1. Additionally, the nearly perpendicularly (80.85(10)°) edge-to-face oriented aromatic rings interact via their π-systems.

Figure 3:

Partial packing diagram of 1 with highlighted C–H⋯π interactions of the proton H18 and the π-system of an adjacent phenyl ring of another molecule (displacement ellipsoids are shown at the 50 % probability level, except for H18 and H18′ which is shown at an arbitrary radius, and H atoms, except for H18 and H18′, are omitted for clarity). Symmetry code: (´) 1-x, 1-y, 1-z.

3.3 Crystal structure description of 2

The molecular structure of compound 2 exhibits a triskelion-like conformation, analogous to that observed in compound 1. It crystallizes in the monoclinic space group P2₁/n (Figure 4). The trigonal-pyramidal structure of the carbon atoms C15, C16, C24 and the nitrogen atom N1 is more pronounced than in compound 1. The height of the inscribed pyramid is 0.4109(9) Å, which is higher than in compound 1, and the sum of the C–N–C angles is smaller and adds up to 337.44(12)°.

Figure 4:

Molecular structure of 2 with corresponding numbering of the atoms (displacement ellipsoids are shown at the 50 % probability level and H atoms are omitted for clarity).

Similarly to compound 1, one proton of a methylene group located in α-position to the nitrogen atom N1 (Figure 5) engages in an intramolecular hydrogen bond with the trityl-functionalized oxygen atom of an adjacent side chain (Hydrogen bond geometry (C24–H24B⋯O1) d(D–H): 0.99 Å, d(H···A): 2.21 Å, <DHA: 137°, d(D···A): 3.0154(10) Å, A: O1). The hydrogen bond can be classified as a moderate hydrogen bond with a structuring influence, based on directionality and atomic distance. ¹⁹ In addition, the free hydroxyl group on the oxygen atom O2 forms a hydrogen bond with the central nitrogen atom N1 (Hydrogen bond geometry (O2–H1⋯N1) d(D–H): 0.87(2) Å, d(H···A): 2.25(2) Å, <DHA: 121(2)°, d(D⋯A): 2.7951(11) Å, A: N1). The slightly smaller bond angle indicates a weaker hydrogen bond than the previous one, yet it remains moderately strong and causes a twisting of the alkyl chain (C16, C17), thereby facilitating the formation of the hydrogen bond. ¹⁹

Figure 5:

Molecular detail of 2 with highlighted hydrogen bond of C24–H24B and O1 as well as O2–H1 and N1 (displacement ellipsoids are shown at the 50 % probability level, except for H24B which is shown at an arbitrary radius, and H atoms, except for H24B, are omitted for clarity).

In the solid-state structure of 2, the primary intermolecular interactions are mainly mediated via sp²-CH/π interactions (Figure 6). In this instance, the phenyl rings of the trityl groups engage in interactions with a range of other molecules. The interactions of the hydrogen atoms H28’ and H42’ with the corresponding π-system (2.749(1) Å and 2.700(1) Å, respectively) are in the same range as those in related compounds with sp²-CH/π interactions (2.73(13) Å). ¹⁸ Consequently, the bond strengths observed here are consistent with those typically observed in other interactions. Interestingly, the distance between hydrogen atom H31’’’ and the plane of the adjacent π-system is considerably shorter (2.316(1) Å) than would be anticipated for a conventional sp²-CH/π interaction. However, this ostensibly stronger interaction is counterbalanced by the fact that the distance between hydrogen atom H31’’’ and the centroid of the aromatic moiety (C33 – C38) is too long (3.6226(4) Å) for any interaction with the π-system.

Figure 6:

Partial packing of 2 with highlighted short contacts (displacement ellipsoids are shown at the 50 % probability level, except for H31’’’, H42’’ and H28′ which is shown at an arbitrary radius, and H atoms, except for H31’’’, H42’’ and H28′, are omitted for clarity). Symmetry codes: (´) −1+x, y, z, (´´) 1-x, 1-y, 1-z, (´´´) x, −1+y, z.

Furthermore, the unprotected oxygen atom O2 of one molecule is in close proximity to the protons of another molecule (hydrogen bond geometry (C24–H24A···O2′) d(D–H): 0.99 Å, d(H···A): 2.78 Å, <DHA: 125°, d(D···A): 3.6876(13) Å, A: O2′) (Figure 7). The interaction of the unprotected side chain now mediates the intermolecular contacts of the molecules in the solid state, in addition to the sp²-CH/π interactions.

Figure 7:

Partial packing of 2 with highlighted short C–H⋯O contact of an acidified hydrogen atom (displacement ellipsoids are shown at the 50 % probability level, except for H24A which is shown at an arbitrary radius, and H atoms, except for H24A, are omitted for clarity). Symmetry code: (´) 3/2-x, 1/2+y, 3/2-z.

3.4 Crystal structure description of 3 and structural comparison

Compound 3 crystallizes in the orthorhombic space group Pcca (Figure 8). The two alkyl side chains with free hydroxyl functions are both disordered, with one occupying two positions (C17 and O2) and the other three positions (C24, C25 and O3). It is noteworthy that the disorder of these atoms was replicated in additional, independent measurements of other single crystals.

Figure 8:

Molecular structure of 3 with corresponding numbering of the atoms. The site occupation factors for O2B and C17B are 0.54 and for C24A, C25A and O3A are 0.50 (displacement ellipsoids are shown at the 50 % probability level and H atoms are omitted for clarity).

As with the other two compounds, the overall structure of the molecule is triskelion-like. Given the existing disorder, only the average value of the selected parameters (weighted according to the SOF) is discussed in comparison to the other compounds (Table 3).

A comparative analysis of pyramidalization at the nitrogen atom is conducted on the basis of three key parameters: the height of the inscribed trigonal pyramid, the sum of the C–N–C angles and the average of the N–C bonds (Table 3). The average bond length of the N–C bond suggests an interaction between the electron density of the free electron pair at the nitrogen atom and the antibonding orbitals of the side chains, resulting in a shorter than expected bond length. This interaction is favored by the planarization of the nitrogen atom.

Upon examination of the three differently substituted triethanolamines and in comparison to the structure of triethanolamine, no anomaly is discernible. This also applies to the sum of angles at the nitrogen atom. Although the sum decreases in value as the degree of tritylation decreases, the maximum value of 339.59(19)° for compound 1 remains significantly below the ideal 360° for a trigonal-planar structure. In comparison to the angular sum for trimethylamine (336.0°), the increasing values for triple-tritylated triethanolamine 1 demonstrate the elimination of hydrogen bonds and an increase in the steric requirement of the side groups. This increase is also reflected in a decrease in the height of the inscribed trigonal pyramid. As the degree of tritylation increases, the height of the pyramid decreases, resulting in a significantly more planar nitrogen atom for compound 1 compared to triethanolamine and trimethylamine. This planarization appears to be due to the steric influence of the trityl groups, as the bond lengths of the central nitrogen atom to the carbon atoms of the side chain remain unchanged.

3.5 Hirshfeld surface analysis and comparison

Further analysis of the interactions between individual molecules involves a comparison of the d _norm surface plots and two-dimensional fingerprint plots using Hirshfeld surface analysis. Given the highly disordered structure of 3, the Hirshfeld surface analysis will be conducted solely for compounds 1 and 2. The parameter d _norm represents the contact values, which are defined by the distances from the nuclei to the interior (d _i ) and exterior (d _e ) of the Hirshfeld surface, in conjunction with the van der Waals radii of the nuclei. The two-dimensional fingerprint plot is used to integrate and display the three-dimensional d _i and d _e plots, thereby simplifying the complex data set. CrystalExplorer was used to generate and evaluate the d _norm surface plots and two-dimensional fingerprint plots. ²¹

The d _norm surface of compound 1 (Figure 9A) clearly shows the dominant presence of interactions of the aromatic C–H bonds with other aromatic compounds (points i, ii and iii). The shielding of the central nitrogen and oxygen atoms by the trityl protecting groups results in minimal involvement of these atoms in intermolecular interactions.

Figure 9:

d_norm surface of compound 1 (A), and the fingerprint plots illustrating the summarized interactions (B), the C⋯C contacts (C), the H⋯H contacts (D), the reciprocal C⋯H/H⋯C contacts (E) and the reciprocal O⋯H/H⋯O contacts (F). Additionally, the contribution of each interaction to the overall interaction is indicated.

The fingerprint plots of the respective contacts confirm these characteristics. In fact, the plots resemble a combination of the plots of two archetypal compounds, namely ethane and benzene. The majority of the H⋅⋅⋅H contacts (Figure 9D) fall within the range of 1.2 Å to 1.8 Å for both d _i and d _e . The pronounced accumulation of signals along the diagonals in the fingerprint plot can be attributed to head-to-head H⋅⋅⋅H contacts, which are commonly observed in alkanes.

Furthermore, the peak observed at short distances in the fingerprint plot with a strong accumulation at approximately d _e  = d _i  ≈ 1.25 Å shows analogies to the plots of benzene.

In addition to the simple H⋅⋅⋅H contacts, the fingerprint plot of the reciprocal C⋅⋅⋅H/H⋅⋅⋅C contacts (Figure 9E) highlights the occurrence of short C–H⋅⋅⋅π interactions by the two separate scatterings in the fingerprint plot (d _e  + d _i  ≈ 1.8 Å + 1.2 Å). These type of scattering plot is also found in the case of benzene. In this context it is noteworthy that the fingerprint plot of the C⋅⋅⋅C contacts (Figure 9C) does not show any interaction of these atoms, especially not at distances of d _e  = d _i  ≈ 1.8 Å, which is typical for stacked aromatic compounds, implying that π⋅⋅⋅π interactions have no intermolecular significance.

As already seen in the analysis of the d _norm surface, both reciprocal N⋅⋅⋅H and O⋅⋅⋅H interactions (Figure 9F) do not contribute to intermolecular interactions in the solid state.

Similarly, the d _norm surface of compound 2 (Figure 10A) also demonstrates the prevalence of intermolecular interactions via aromatic C–H bonds (points ii and iii in Figure 10A), which is analogous to the d _norm surface of compound 1. In addition, point i (Figure 10A) shows an interaction via the methylene protons, as previously illustrated in the structural discussion (Figure 7).

Figure 10:

d_norm surface of compound 2 (A), and the fingerprint plots illustrating the summarized interactions (B), the C⋯C contacts (C), the H⋯H contacts (D), the reciprocal C⋯H/H⋯C contacts (E) and the reciprocal O⋯H/H⋯O contacts (F). Additionally, the contribution of each interaction to the overall interaction is indicated.

Moreover, the fingerprint plots of compound 2 exhibit numerous similarities with those of compound 1. The H⋅⋅⋅H contacts (Figure 10D) are dominated by head-to-head contacts, with a notable accumulation of points in the fingerprint plot at approximately d _e  = d _i  ≈ 1.25 Å, which is characteristic of simple aromatic compounds. The reciprocal C⋯H/H⋯C contacts (Figure 10E) also indicate the occurrence of C–H⋯π interactions. However, the absence of a trityl group renders the corresponding alkyl chain more accessible, thereby facilitating further C–H interactions by the methylene groups. This is reflected in the fingerprint plot by the absence of two distinct scatterings, resulting in a single, connected scattering.

The high flexibility of the aforementioned side chain is also reflected in the occurrence of intermolecular H⋅⋅⋅O interactions (Figure 10F), which are primarily mediated by methylene protons which are polarized by the adjacent nitrogen atom, and the oxygen atom of the free hydroxyl function. Moreover, the absence of a sterically demanding group permits the aromatic moieties of the remaining trityl groups to interact more strongly. However, π⋯π interactions play a secondary role in the intermolecular interactions, as evidenced by the fingerprint plot of the C–C contacts (Figure 10C), which indicates interactions at slightly longer distances than the ideal distance of d _e  = d _i  ≈ 1.8 Å.

The relative contributions of different interactions (Figure 11) in the intermolecular interactions are only marginally affected by the degree of tritylation when comparing compounds 1 and 2. Although compound 2 exhibits augmented proportions of both C⋅⋅⋅C and O⋅⋅⋅H interactions, which can be attributed to the increased flexibility of the unprotected alkyl chain, H⋅⋅⋅H and C⋅⋅⋅H contacts persist in dominant structural roles. Nevertheless, it can be postulated that the impact is likely to be even more pronounced for compound 3, given that the side chains display enhanced flexibility.

Figure 11:

Summary of the contribution of each intermolecular interaction for each compound (violet: Other contacts 0.2 % for compound 2).

The results demonstrate that the structural environment surrounding the central nitrogen atom of trialkyl-functionalized molecules can be influenced by sterically demanding side groups, even when these side groups are not in close proximity to the nitrogen atom in question. Nevertheless, this study also demonstrates that the effect is considerably lower compared to groups directly connected to the nitrogen atom. In conclusion, this study contributes to the understanding of factors influencing the planarization of nitrogen atoms and reveals additional control mechanisms beyond the orbital interactions of the lone pair of the nitrogen atom.