Synthesis of Guerbet ionic liquids and extractants as β-branched biosourceable hydrophobes

Article abstract of DOI:10.1039/c9ob02237f

This study investigates the synthesis of β-branched amines and β-branched quaternary ammonium chloride ionic liquids as novel extractants. The synthesis methodology was tailored to facilitate the reaction scale-up and the use of biorenewable starting materials. The developed process is an overall green, easy and straightforward synthesis of β-branched amines, and ammonium salts, starting from linear aldehydes. In order to evaluate the potential of the synthesised materials in applications, the rheology, density, thermal stability, chemical stability, phase transitions, and mutual solubility with water of the novel extractants was studied.

Full text of DOI:10.1039/c9ob02237f

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Synthesis of Guerbet ionic liquids and extractants
as β-branched biosourceable hydrophobes†
Cite this: Org. Biomol. Chem., 2019,
17, 9778
Giacomo Damilano,
Koen Binnemans
and Wim Dehaen
*
This study investigates the synthesis of β-branched amines and β-branched quaternary ammonium chlor-
ide ionic liquids as novel extractants. The synthesis methodology was tailored to facilitate the reaction
scale-up and the use of biorenewable starting materials. The developed process is an overall green, easy
and straightforward synthesis of β-branched amines, and ammonium salts, starting from linear aldehydes.
In order to evaluate the potential of the synthesised materials in applications, the rheology, density,
thermal stability, chemical stability, phase transitions, and mutual solubility with water of the novel extrac-
tants was studied.
Received 16th October 2019,
Accepted 28th October 2019
DOI: 10.1039/c9ob02237f
rsc.li/obc
Quaternisation of these amines could lead to new quaternary
ammonium ionic liquids (ILs).9,10
Introduction
In 2012, the European Union plan ‘towards a sustainable
A recent market analysis described the industrial under-
European future’ supported a transition towards a more sus- usage of Guerbet additives in fine products, giving evidence of
tainable development.1 Shifting from a fossil-based to a bio- the potential economic worth of these compounds.11
based circular economy is key to this transition.2 Because cost Currently, many chemical manufacturers (i.e. BASF, SASOL,
efficiency is the driving force for industry, the exploitation of Evonik) are commercialising Guerbet alcohols for industrial
cheap non-food biomass is the key for accessing a circular applications.12 This highlights the commercial relevance of
bioeconomy. The efficient use of platform biochemicals these compounds and interest in devising a sustainable syn-
happens through reactions with high atom economy (i.e. con- thesis towards them.
densation, ketonisation, esterification and etherification reac-
The Guerbet condensation can be rationalised as a three-
tions). These reactions are means to a competitive synthesis of step one-pot reaction. It comprises an alcohol oxidation, an
renewable building blocks.3 A successful example of sustain- aldol condensation and a reduction step (Scheme 1). Although
able biosourcing is ethanol. Nowadays, over 90% of ethanol Guerbet condensation produces β-branched alcohols in a neat
derives from biomass fermentation, which overcame petro- and straightforward way, the reaction has a low efficiency.
chemical synthesis in economic and environmental Harsh reaction conditions are required and cause reduced
sustainability.3,4
selectivity towards the desired product.
Employing biomaterials gathered the interest of both aca-
A growing number of studies is aiming at devising suitable
demic and industrial researchers.5 This shifted the focus back conditions to promote Guerbet condensation at an industrial
to condensation reactions as a means to upgrade low-weight scale. Yet, most of the discovered pathways still require harsh
alcohols. The Guerbet reaction, discovered in 1909, is the con- reaction conditions and suffer either from a narrow reaction
densation of linear alcohols to β-branched alcohols (i.e. scope, use of scarce and noxious catalysts, or a limited
Guerbet alcohols). Guerbet alcohols are surfactants with selectivity.13–16 The limitations of the reported procedures are
reduced viscosity, low irritation profile, and high thermal due to an elusive optimisation of reaction conditions, which at
stability.6 Because of these features, Guerbet alcohols, acids the same time would have to favour the alcohol oxidation and
and esters find wide use in detergents, cosmetics, emulsifiers, the unsaturated aldehyde reduction.14 A lack of optimal reac-
antistatic additives, inks, lubricants, and waxes.7 By the con-
version of Guerbet alcohols to amines new extractants for
solvent extraction of metals could be designed.8
KU Leuven, Department of Chemistry, Celestijnenlaan 200F – P. O. Box 2404, B-3001
Leuven, Belgium. E-mail: wim.dehaen@kuleuven.be
†Electronic supplementary information (ESI) available: Characterisation and
material and equipment specifications. See DOI: 10.1039/c9ob02237f
Scheme 1 General reaction scheme of Guerbet condensation.
9778 | Org. Biomol. Chem., 2019, 17, 9778–9791
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
tion conditions favours the formation of by-products (i.e.
resulting from Tishchenko, Cannizzaro, Dieckmann reactions),
limiting the overall yield.13,17–19
In 2016, Biermann et al. suggested an alternative approach,
based on splitting up the reaction in its composing steps (i.e.
oxidation, condensation and reduction).13,20 Although
effective, it relies on biocatalysis, which is a valuable approach
but limited by the enzyme turnover number and substrate
affinity. Because of the reported slow conversion and poor
affinity of most enzymes for fatty alkyl chains, this approach
was not suited to our goal.21,22 For use in the solvent extraction
of metal ions, the amines and the ionic liquids have to be
hydrophobic. Hydrophobicity is observed in compounds with
a high carbon content. Therefore, the use of long alkyl chains
could serve this goal. However, this lengthening of the alkyl
chains comes at a price, namely a sharp increase in viscosity
due to the extensive dispersive interactions. The introduction
of branching in the chains lowers the efficacy of these inter-
actions, hence limiting the increase in viscosity.
Moreover, branching in the β-position was designed to
support ammonium ion base stability. The base instability of
ammonium compounds is reported to be mainly due to
Hofmann elimination. Hofmann elimination is an E2 degra-
dation process caused by the concerted deprotonation in β to
the nitrogen atom of an ammonium ion together with the clea-
vage of the nitrogen–carbon bond, resulting in the formation
of a tertiary amine and an alkene (Scheme 2).23–25 Base-stable
hydrophobic ionic liquids are of interest for the solvent extrac-
tion of metal ions from alkaline media.26
The β-branched structure hinders Hofmann elimination by
reducing the number of β-protons and, more importantly, by
increasing the steric hindrance around the remaining
Scheme 3 General schematic representation of the studied synthetic
β-protons. This approach was effective in a former research on pathway.
di(2-ethylhexyl)dimethylammonium (BEDMA) ionic liquids.27
The structural similarity of our compounds to methyl-tri
(octyl)-ammonium chloride (main component of Aliquat
336®), tri(octyl)amine (TOA, main component of Alamine
336®), and tri(2-ethylhexyl)amine (TEHA) suggests that they
could be potential competitors in extractive metallurgical pro-
cesses, while being stable at a broader pH range.
were able to synthesise compounds on a multigram scale,
allowing for the physical characterisation of the produced
materials.27
We report herein an adaptation of the Guerbet synthetic
pathway to achieve a green and scalable procedure for the syn-
thesis of longer-chain Guerbet aldehydes (Scheme 3). Then,
Experimental
Chemicals and solvents
the β-branched aldehydes were used in the synthesis of nitro- L-Lysine (98%, 56-87-1) and tri-N-octylamine (97%, 1116-76-3)
gen derivatives (i.e. tertiary amine and ammonium ionic were purchased from Acros Organics (Geel, Belgium).
liquids), which are unreported, to the best of our knowledge, N-Methylformamide (99%, 123-39-7) was purchased from Alfa
in the literature. Under the optimised reaction conditions, we Aesar (Karlsruhe, Germany). Dimethyl carbonate (≥99.8%, 616-
38-6) was purchased from Carl-Roth (Karlsruhe, Germany).
Formamide (99.5%, 75-12-7), calcium chloride anhydrous
(95%+, 10043-52-4), and formic acid (99–100%, 64-18-6) were
purchased from Chem-Lab (Zedelgem, Belgium). Palladium on
activated carbon (10% Pd, unreduced, 7440-05-3) was purchased
from J&K Scientific (Overpelt, Belgium). 1-Hexanal (99.9%, 66-
25-1), 1-heptanal (98.3%, 111-71-7), 1-octanal (99.8%, 124-13-0),
1-nonanal (99.2%, 124-19-6), 1-decanal (97.8%, 112-31-2), deu-
Scheme 2 General reaction scheme of Hofmann elimination.
terochloroform (99.8 atom % D with 0.03% (v/v) TMS, 865-49-6)
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9778–9791 | 9779

View Article Online
Paper
Organic & Biomolecular Chemistry
and RANEY®-nickel 2800, slurry in water (≥89%, 7440-02-0) 2 mol), L-lysine (14.7 g, 0.5 mmol), isopropanol (156 g,
were purchased from Sigma-Aldrich (Diegem, Belgium). Sodium 260 cm3, 2.6 mol) and palladium over carbon (10% w/w
hydrogen carbonate (>99%, 144-55-8) was purchased from loading, 7 g). Isolated yield 173.74 g, 68%. Colourless clear oil.
Fisher Scientific (Merelbeke, Belgium).
Boiling point: (109
1) °C, (0.29
0.029) mbar. Density:
Petroleum ether (technical grade), methanol (technical 0.83144 g cm−3 (20.02 °C). Dynamic viscosity: 4.085 mPa s−1
grade), isopropanol (technical grade) and sodium hydroxide (20.00 °C), kinetic viscosity: 4.913 mm2 s−1 (20.00 °C). 1H NMR
(Normapur) were obtained from VWR (Heverlee, Belgium). (400 MHz, CDCl3, δ/ppm): 0.73–1.03 (6 H, m), 1.19–1.36 (16 H,
Petroleum ether was distilled prior to use to remove high m), 1.36–1.51 (2 H, m), 1.51–1.79 (3 H, m), 2.22 (1 H, ttd, J
boiling residues; all other chemicals were used without any 8.39, 5.49, 5.45, 3.17), 9.55 (1 H, d, J 3.18). 13C NMR (100 MHz,
further purification.
CDCl3, δ/ppm): 205.8, 52.0, 31.9, 31.8, 29.7, 29.1, 29.0, 28.9,
27.1, 26.8, 22.7, 22.5, 14.1, 14.0. IR (Diamond-ATR, neat,
cm−1): 2956, 2925, 2855, 2693, 1725, 1459, 1378, 1266, 1042,
Instrumentation
Nuclear magnetic resonance spectra were recorded on a 893, 821, 737, 724, 571, 473, 450, 406. HRMS calcd for
Bruker Avance 400 spectrometer. Fourier transform infrared C14H28O: 212.2140 found: [M + 1] 213.2208.
spectra were recorded in ATR mode (attenuated total reflec-
2-Hexyl-1-decanal
tance) on a Bruker Vertex 70 spectrometer equipped with a
Bruker Platinum ATR module and a diamond crystal. HRMS The synthesis was carried out following the ESI General pro-
spectra were acquired on a quadrupole orthogonal acceleration cedures 9 and 10† with 1-octanal (99.8%, 256.93 g, 314 cm3,
time-of-flight mass spectrometer (Synapt G2 HDMS, Waters, 2 mol), L-lysine (14.7 g, 0.5 mmol), isopropanol (156 g,
Milford, MA). Samples were infused at 3 μL min−1 and spectra 260 cm3, 2.6 mol) and palladium over carbon (10% w/w
were obtained in positive (or negative) ionization mode with a loading, 8 g). Isolated yield 171.80 g, 67%. Colourless clear oil.
resolution of 15 000 (FWHM) using leucine enkephalin as the Boiling point: (110
lock mass. The water content in samples was measured using 0.83162 g cm−3 (20.02 °C). Dynamic viscosity: 5.409 mPa s−1
Mettler-Toledo V30S volumetric Karl Fischer titrator. (20.00 °C), kinetic viscosity: 6.504 mm2 s−1 (20.00 °C). 1H NMR
1) °C, (0.40
0.04) mbar. Density:
a
Viscosities and densities were measured on an Anton-Paar (400 MHz, CDCl3, δ/ppm): 0.77–0.95 (6 H, m), 1.12–1.36 (20 H,
Lovis 2000 M/ME rolling-ball viscometer and DMA 4500 M m), 1.36–1.53 (3 H, m), 1.53–1.75 (2 H, m), 2.16–2.27 (1 H, m),
densitometer. TGA and DSC analyses were carried out on a TA 9.55 (1 H, d, J 3.18). 13C NMR (100 MHz, CDCl3, δ/ppm): 205.8,
Instruments TGA Q500 and a TA Instruments DSC Q2000 52.0, 31.9, 31.7, 29.7, 29.4, 29.4, 29.3, 29.0, 27.1, 27.1, 22.7,
respectively, with a heating/cooling rate of 10 °C min−1. Ionic 22.6, 14.1, 14.1. IR (Diamond-ATR, neat, cm−1): 2955, 2923,
liquids were pre-dried for 120 min at 80 °C before TGA 2854, 2690, 1726, 1466, 1378, 1303, 1144, 962, 783, 722, 665,
measurement. Vacuum measurements were carried out with a 565, 506, 484, 473, 462, 448, 437, 429, 419, 413. HRMS calcd
Pfeiffer TPG 201 digital manometer. An Ismatec Reglo ICC for C16H32O: 240.2453 found: [M − 44] 197.1035.
peristaltic pump was used for the dropping of the reagents.
2-Heptyl-1-undecanal
2-Butyl-1-octanal
The synthesis was carried out following the ESI General pro-
The synthesis was carried out following the ESI General pro- cedures 9 and 10† with 1-nonanal (99.2%, 286.77 g, 346 cm3,
cedures 9 and 10† with 1-hexanal (99.9%, 200 g, 200 cm3, 2 mol), L-lysine (14.7 g, 0.5 mmol), isopropanol (156 g,
2 mol), L-lysine (14.7 g, 0.5 mmol), isopropanol (156 g, 260 cm3, 2.6 mol) and palladium over carbon (10% w/w
260 cm3, 2.6 mol) and palladium over carbon (10% w/w loading, 9 g). Isolated yield 179.21 g, 63%. Colourless clear oil.
loading, 6 g). Isolated yield 122.04 g, 66.3%. Colourless clear Boiling point: (128
oil. Boiling point: (100 5) °C, (1.5
1) °C, (0.42
0.042) mbar. Density:
0.15) mbar. Density: 0.83306 g cm−3 (20.02 °C). Dynamic viscosity: 7.201 mPa s−1
0.83050 g cm−3 (20.02 °C). Dynamic viscosity: 2.432 mPa s−1 (20.00 °C), kinetic viscosity: 8.645 mm2 s−1 (20.00 °C). 1H NMR
(20.00 °C), Kinetic viscosity: 2.929 mm2 s−1 (20.00 °C). 1H (400 MHz, CDCl3, δ/ppm): 0.81–0.95 (7 H, m), 1.18–1.37 (29 H,
NMR (400 MHz, CDCl3, δ/ppm): 0.84–0.93 (6 H, m), 1.20–1.38 m), 1.37–1.52 (3 H, m), 1.52–1.69 (3 H, m), 2.22 (1 H, ttd,
(13 H, m), 1.38–1.54 (2 H, m), 1.56–1.68 (2 H, m), 2.22 (1 H, J 8.41, 5.49, 5.44, 3.15), 9.55 (1 H, d, J 3.09). 13C NMR
ttd, J 8.36, 5.49, 5.45, 3.16), 9.56 (1 H, d, J 3.20). 13C NMR (100 MHz, CDCl3, δ/ppm): 205.8, 52.0, 31.9, 31.8, 29.7, 29.7,
(100 MHz, CDCl3, δ/ppm): 205.8, 52.0, 31.7, 29.4, 29.3, 29.0, 29.5, 29.5, 29.3, 29.1, 29.0, 27.1, 22.7, 22.6, 14.1, 14.1. IR
28.6, 27.1, 22.8, 22.6, 14.1, 13.9. IR (Diamond-ATR, neat, (Diamond-ATR, neat, cm−1): 2955, 2922, 2853, 2689, 1727,
cm−1): 2957, 2927, 2857, 2692, 1725, 1688, 1640, 1465, 1378, 1651, 1465, 1377, 1069, 892, 809, 721, 662, 558, 526, 503, 497,
1132, 1076, 1029, 936, 913, 896, 797, 725, 649, 565, 499, 488, 473, 455, 441, 430, 420, 412. HRMS calcd for C18H36O:
459, 446, 429, 420, 409. HRMS calcd for C12H24O: 184.1827 268.2766 found: [M − 44] 193.1949.
found: [M − 44] 137.0826.
2-Octyl-1-dodecanal
2-Pentyl-1-nonanal
The synthesis was carried out following the ESI General pro-
The synthesis was carried out following the ESI General pro- cedures 9 and 10† with 1-decanal (97.8%, 318.38 g, 387 cm3,
cedures 9 and 10† with 1-heptanal (98.3%, 232.33 g, 284 cm3, 2 mol), L-lysine (14.7 g, 0.5 mmol), isopropanol (156 g,
9780 | Org. Biomol. Chem., 2019, 17, 9778–9791
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
260 cm3, 2.6 mol) and palladium over carbon (10% w/w (49 H, m), 1.39–1.50 (3 H, m), 2.04 (4 H, d, J 7.02), 2.09
loading, 10 g). Isolated yield 202.38 g, 68.6%. Colourless clear (3 H, s). 13C NMR (100 MHz, CDCl3, δ/ppm): 63.7, 43.4, 35.9,
oil. Melting point: (18.5 1) °C, 1003 mbar. Boiling point: (145 32.0, 32.0, 30.3, 30.0, 29.7, 29.4, 26.7, 26.6, 22.8, 22.7, 14.1,
1) °C, (0.31
0.031) mbar. Density: 0.84467 g cm−3 32.3. IR (Diamond-ATR, neat, cm−1): 2955, 2921, 2852, 2791,
(20.02 °C). Dynamic viscosity: 15.82 mPa s−1 (20.00 °C), kinetic 1691, 1459, 1377, 1302, 1034, 850, 721. HRMS calcd for
viscosity: 18.73 mm2 s−1 (20.00 °C). 1H NMR (400 MHz, CDCl3, C33H69N: 479.5430 found: [M + 1] 480.5513.
δ/ppm): 0.84–0.92 (6 H, m), 1.16–1.35 (31 H, m), 1.36–1.49
(2 H, m), 1.56–1.67 (2 H, m), 2.22 (1 H, ttd, J 8.41, 5.53, 5.43, Di(2-heptyl-undecyl)methylamine
3.14), 9.55 (1 H, d, J 3.21). 13C NMR (100 MHz, CDCl3, δ/ppm):
205.8, 52.0, 31.9, 31.9, 29.7, 29.6, 29.6, 29.5, 29.4, 29.3, 29.2,
The synthesis was carried out following the ESI† General pro-
cedure 11† with 2-heptyl-1-undecanal (134.88 g, 0.5 mol),
29.0, 27.1, 22.7, 22.7, 14.1, 14.1. IR (Diamond-ATR, neat,
formic acid (69.1 g, 56.6 cm3, 1.5 mol), N-methyl-formamide
cm−1): 2955, 2922, 2853, 2691, 1727, 1690, 1465, 1377, 1051,
(133 g, 130 cm3, 2.25 mol), RANEY nickel (0.15 g). Isolated
721, 583, 506, 493, 473, 458, 453, 442, 434, 426, 414, 407.
yield 117.64 g, 87.4%. Colourless clear oil. Density: 0.82521 g
HRMS calcd for C20H40O: 296.3079 found: [M − 44] 252.2678.
cm−3 (20.02 °C). Dynamic viscosity: 42.956 mPa s−1 (20.00 °C),
kinetic viscosity: 52.055 mm2 s−1 (20.00 °C). 1H NMR
(400 MHz, CDCl3, δ/ppm): 0.84–0.93 (12 H, m), 1.18–1.32
(57 H, m), 1.40–1.48 (2 H, m), 2.04 (4 H, d, J 7.05), 2.09 (3 H,
s). 13C NMR (100 MHz, CDCl3, δ/ppm): 63.7, 43.4, 35.9, 14.1,
22.7, 32.3, 32.0, 32.0, 30.3, 30.3, 29.8, 29.7, 29.5, 29.4, 26.7. IR
(Diamond-ATR, neat, cm−1): 2955, 2921, 2852, 2791, 1731,
1690, 1460, 1377, 1301, 1034, 857, 720. HRMS calcd for
C37H77N: 535.6056 found: [M + 1] 536.6127.
Di(2-butyl-octyl)methylamine
The synthesis was carried out following the ESI General pro-
cedure 11† with 2-butyl-1-octanal (90.3 g, 0.5 mol), formic acid
(69.1 g, 56.6 cm3, 1.5 mol), N-methyl-formamide (133 g,
130 cm3, 2.25 mol), and RANEY nickel (0.15 g). Isolated yield
77.8 g, 86.4%. Colourless clear oil. Density: 0.81769 g cm−3
(20.02 °C). Dynamic viscosity: 16.568 mPa s−1 (20.00 °C),
kinetic viscosity: 20.262 mm2 s−1 (20.00 °C). 1H NMR
(400 MHz, CDCl3, δ/ppm): 0.80–0.96 (12 H, m), 1.17–1.33 (32
H, m), 1.40–1.48 (1 H, m), 2.04 (4 H, d, J 7.08), 2.09 (3 H, s).
13C NMR (100 MHz, CDCl3, δ/ppm): 63.7, 43.3, 35.8, 32.3, 32.0,
32.0, 29.9, 28.9, 26.6, 23.3, 22.7, 14.2, 14.1. IR (Diamond-ATR,
neat, cm−1): 2955, 2922, 2871, 2853, 2791, 2763, 1690, 1459,
1377, 1303, 1102, 1037, 864, 767, 722. HRMS calcd for
C25H53N: 367.4178 found: [M + 1] 368.4261.
Di(2-octyl-dodecyl)methylamine
The synthesis was carried out following the ESI General pro-
cedure 11† with 2-octyl-1-dodecanal (150 g, 0.5 mol), formic
acid (69.1 g, 56.6 cm3, 1.5 mol), N-methyl-formamide (133 g,
130 cm3, 2.25 mol), and RANEY nickel (0.15 g). Isolated yield
125.53 g, 83.8%. Colourless clear oil. Density: 0.82628 g cm−3
(20.02 °C). Dynamic viscosity: 48.272 mPa s−1 (20.00 °C),
kinetic viscosity: 58.421 mm2 s−1 (20.00 °C). 1H NMR
(400 MHz, CDCl3, δ/ppm): 0.88 (12 H, t, J 6.61), 1.19–1.34 (64
H, m), 1.39–1.49 (2 H, m), 2.04 (3 H, d, J 7.02), 2.09 (2 H, s).
13C NMR (100 MHz, CDCl3, δ/ppm): 63.7, 43.4, 35.8, 14.1, 22.7,
26.7, 32.3, 32.0, 30.3, 29.8, 29.7, 29.7, 29.4, 29.4. IR (Diamond-
ATR, neat, cm−1): 2955, 2921, 2852, 1730, 1689, 1462, 1377,
1301, 1179, 1036, 720. HRMS calcd for C41H85N: 591.6682
found: [M + 1] 592.6763.
Di(2-pentyl-nonyl)methylamine
The synthesis was carried out following the ESI† General pro-
cedure 11 with 2-pentyl-1-nonanal (85.29 g, 0.4 mol), formic
acid (69.1 g, 56.6 cm3, 1.5 mol), N-methyl-formamide (133 g,
130 cm3, 2.25 mol), RANEY nickel (0.15 g). Isolated yield
61.18 g, 71.9%. Colourless clear oil. Density: 0.81811 g cm−3
(20.02 °C). Dynamic viscosity: 25.06 mPa s−1 (20.00 °C), kinetic
viscosity: 30.631 mm2 s−1 (20.00 °C). 1H NMR (400 MHz,
CDCl3, δ/ppm): 2.09 (3 H, s), 2.04 (4 H, d, J 7.06), 1.40–1.49
(2 H, m), 1.17–1.35 (41 H, m), 0.82–0.95 (12 H, m). 13C NMR
Tri(2-butyl-octyl)amine
(100 MHz, CDCl3, δ/ppm): 63.7, 43.4, 35.9, 32.5, 32.3, 32.3, The synthesis was carried out following the ESI General pro-
30.3, 29.4, 26.7, 26.3, 22.8, 22.7, 32.0, 14.2, 14.1. IR (Diamond- cedure 12† with 2-butyl-1-octanal (14.59 g, 0.08 mol), formic
ATR, neat, cm−1): 2955, 2921, 2853, 2791, 2763, 1690, 1459, acid (45.0 g, 36.88 cm3, 0.75 mol), formamide (33.8 g,
1377, 1301, 1102, 1035, 863, 722. HRMS calcd for C29H61N: 29.89 cm3, 0.75 mol), and RANEY nickel (0.15 g). Isolated yield
423.4804 found: [M + 1] 424.4863.
6.58 g, 47.8%. Colourless clear oil. Density: 0.82859 g cm−3
(20.02 °C). Dynamic viscosity: 72.161 mPa s−1 (20.00 °C),
kinetic viscosity: 87.089 mm2 s−1 (20.00 °C). 1H NMR
Di(2-hexyl-decyl)methylamine
The synthesis was carried out following the ESI General pro- (400 MHz, CDCl3, δ/ppm): 2.02 (6 H, d, J 5.97), 1.12–1.38 (56
cedure 11† with 2-hexyl-1-decanal (120.13 g, 0.5 mol), formic H, m), 0.84–0.94 (19 H, m). 13C NMR (100 MHz, CDCl3,
acid (69.1 g, 56.6 cm3, 1.5 mol), N-methyl-formamide (133 g, δ/ppm): 61.1, 36.4, 32.4, 32.1, 32.0, 14.1, 14.2, 22.8, 23.4, 26.8,
130 cm3, 2.25 mol), and RANEY nickel (0.15 g). Isolated yield 29.1, 30.1, 30.1, 30.0. IR (Diamond-ATR, neat, cm−1): 2955,
94.96 g, 79.2%. Colourless clear oil. Density: 0.82189 g cm−3 2922, 2872, 2855, 2791, 2763, 1457, 1419, 1377, 1300, 1230,
(20.02 °C). Dynamic viscosity: 33.283 mPa s−1 (20.00 °C), 1192, 1165, 1141, 1113, 1099, 1041, 855, 779, 724, 560, 490,
kinetic viscosity: 40.496 mm2 s−1 (20.00 °C). 1H NMR 468, 452, 444, 437, 426, 419, 411, 403. HRMS calcd for
(400 MHz, CDCl3, δ/ppm): 0.82–0.96 (12 H, m), 1.16–1.35 C36H75N: 521.5899 found: [M + 1] 522.5970.
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9778–9791 | 9781

View Article Online
Paper
Organic & Biomolecular Chemistry
Tri(2-hexyl-decyl)amine
carbonate (2.80 g, 3 cm3, mmol), and methanol (0.792 g,
1 cm3, mmol). Isolated yield 0.902 g, 94.2%. Yellow clear oil.
1H NMR (400 MHz, CDCl3, δ/ppm): 1.83 (1 H, p, J 5.54, 5.46),
1.13–1.63 (32 H, m), 0.90 (12 H, dt, J 12.91, 6.61), 3.37–3.51 (11
H, m). 13C NMR (100 MHz, CDCl3, δ/ppm): 70.5, 33.8, 33.5,
33.2, 31.7, 29.4, 28.4, 26.3, 22.8, 22.6, 14.1, 14.1, 14.0, 50.7,
50.6, 50.8. IR (Diamond-ATR, neat, cm−1): 3372, 3010, 2954,
2925, 2856, 1626, 1486, 1466, 1377, 1042, 971, 892, 725, 582,
518, 510, 492, 467, 459, 426, 419, 414. HRMS calcd for
[C26H56N][Cl]: 417.4101 found: [M − 35] 382.4415.
The synthesis was carried out following the ESI General pro-
cedure 12† with 2-hexyl-1-decanal (25.96 g, 28.95 cm3,
0.1 mol), formic acid (45.0 g, 36.88 cm3, 0.75 mol), formamide
(33.8 g, 29.89 cm3, 0.75 mol), and RANEY nickel (0.15 g).
Isolated yield 17.54 g, 70.6%. Colourless clear oil. Density:
0.82979 g cm−3 (20.02 °C). Dynamic viscosity: 90.088 mPa s−1
(20.00 °C), kinetic viscosity: 108.57 mm2 s−1 (20.00 °C). 1H
NMR (400 MHz, CDCl3, δ/ppm): 0.83–0.94 (18 H, m), 1.11–1.37
(76 H, m), 2.01 (6 H, d, J 5.84). 13C NMR (100 MHz, CDCl3,
δ/ppm): 61.1, 36.4, 14.1, 22.7, 22.8, 26.8, 26.8, 29.5, 29.8, 30.1,
30.4, 32.0, 32.0, 32.4, 32.4. IR (Diamond-ATR, neat, cm−1):
2955, 2920, 2871, 2852, 2803, 1458, 1377, 1298, 1156, 1077,
890, 721, 543, 478, 458, 451, 438, 420, 410, 404. HRMS calcd
for C48H99N: 689.7777 found: [M + 1] 690.7862.
Di(2-pentyl-nonyl)dimethylammonium chloride
([C9,β5C9,β5C1C1N][Cl])
The synthesis was carried out following the ESI General pro-
cedure 13† with di(2-pentyl-nonyl)methylamine (0.877 g,
2.1 mmol), calcium chloride (95%, 0.6 g, 5.4 mmol), dimethyl
carbonate (2.80 g, 3 cm3, mmol), and methanol (0.792 g,
1 cm3, mmol). Isolated yield 0.915 g, 93.3%. Amber clear oil.
1H NMR (400 MHz, CDCl3, δ/ppm): 1.77–1.85 (1 H, m),
1.12–1.57 (41 H, m), 0.81–0.99 (12 H, m), 3.32–3.56 (11 H, m).
13C NMR (100 MHz, CDCl3, δ/ppm): 71.9, 70.5, 53.5, 50.7, 33.8,
33.7, 33.2, 31.9, 31.8, 29.7, 29.2, 26.3, 26.0, 22.6, 22.6, 14.1,
14.0. IR (Diamond-ATR, neat, cm−1): 3379, 2954, 2922, 2853,
1709, 1487, 1466, 1377, 971, 901, 722, 669, 558, 507, 486, 463,
456, 446, 439, 425, 417. HRMS calcd for [C30H64N][Cl]:
473.4727 found: [M − 35] 438.5037.
Tri(2-heptyl-undecyl)amine
The synthesis was carried out following the ESI General pro-
cedure 12† with 2-heptyl-1-undecanal (27.90 g, 0.1 mol),
formic acid (45.0 g, 36.88 cm3, 0.75 mol), formamide (33.8 g,
29.89 cm3, 0.75 mol), and RANEY nickel (0.15 g). Isolated yield
20.88 g, 77.8%. Colourless clear oil. Density: 0.83287 g cm−3
(20.02 °C). Dynamic viscosity: 103.19 mPa s−1 (20.00 °C),
kinetic viscosity: 123.90 mm2 s−1 (20.00 °C). 1H NMR
(400 MHz, CDCl3, δ/ppm): 1.15–1.35 (87 H, m), 2.01 (6 H, d,
J 5.87), 0.84–0.94 (18 H, m). 13C NMR (100 MHz, CDCl3,
δ/ppm): 61.1, 14.1, 22.7, 36.4, 32.4, 32.0, 32.0, 30.4, 30.4, 29.9,
29.8, 29.5, 29.4, 26.9, 26.9. IR (Diamond-ATR, neat, cm−1):
2955, 2920, 2871, 2851, 1465, 1377, 1297, 1104, 1077, 889, 721,
480, 462, 437, 421, 407. HRMS calcd for C54H111N: 773.8716
found: [M + 1] 773.8679.
Di(2-hexyl-decyl)dimethylammonium chloride
([C10,β6C10,β6C1C1N][Cl])
The synthesis was carried out following the ESI General pro-
cedure 13† with di(2-hexyl-decyl)methylamine (0.866 g,
1.8 mmol), calcium chloride (95%, 0.6 g, 5.4 mmol), dimethyl
carbonate (2.80 g, 3 cm3, mmol), and methanol (0.792 g,
1 cm3, mmol). Isolated yield 0.768 g, 80.29%. Amber clear oil.
1H NMR (400 MHz, CDCl3, δ/ppm): 1.57–1.65 (1 H, m),
0.98–1.27 (49 H, m), 0.63 (12 H, t, J 6.72), 3.02–3.35 (10 H, m).
13C NMR (100 MHz, CDCl3, δ/ppm): 71.2, 70.1, 53.3, 50.4, 48.7,
39.7, 33.5, 32.8, 31.5, 31.6, 29.5, 29.2, 29.2, 29.0, 26.0, 26.0,
22.4, 22.4. IR (Diamond-ATR, neat, cm−1): 3357, 2954, 2921,
2852, 1739, 1674, 1485, 1465, 1377, 1072, 889, 720, 579, 493,
481, 473, 451, 427, 420, 415. HRMS calcd for [C34H72N][Cl]:
529.5353 found: [M − 35] 494.5663.
Tri(2-octyl-dodecyl)amine
The synthesis was carried out following the ESI General pro-
cedure 12† with 2-octyl-1-dodecanal (28.54 g, 0.1 mol), formic
acid (45.0 g, 36.88 cm3, 0.75 mol), formamide (33.8 g,
29.89 cm3, 0.75 mol), and RANEY nickel (0.15 g). Isolated yield
22.59 g, 77.8%. Colourless clear oil. Melting point: −3.03 °C,
heat of fusion: 3.990
J ; crystallization temperature:
g−1
−18.50 °C, heat of crystallization: 3.450 J g−1. Density:
0.83355 g cm−3 (20.02 °C). Dynamic viscosity: 104.56 mPa s−1
(20.00 °C), kinetic viscosity: 125.44 mm2 s−1 (20.00 °C). 1H
NMR (400 MHz, CDCl3, δ/ppm): 0.82–0.92 (18 H, m), 1.13–1.36
(99 H, m), 2.01 (6 H, d, J 5.90). 13C NMR (100 MHz, CDCl3,
δ/ppm): 61.1, 36.4, 14.1, 22.7, 26.9, 32.4, 30.4, 29.4, 29.5, 29.8,
29.8, 29.9, 32.0, 32.0. IR (Diamond-ATR, neat, cm−1): 2955,
2920, 2851, 1465, 1377, 1298, 1151, 1073, 887, 720, 555, 474,
464, 436, 419, 412, 404. HRMS calcd for C60H123N: 857.9655
found: [M + 1] 859.8108.
Di(2-heptyl-undecyl)dimethylammonium chloride
([C11,β7C11,β7C1C1N][Cl])
The synthesis was carried out following the ESI General pro-
cedure 13† with di(2-heptyl-undecyl)methylamine (0.824 g,
1.5 mmol), calcium chloride (95%, 0.6 g, 5.4 mmol), dimethyl
carbonate (2.80 g, 3 cm3, mmol), and methanol (0.792 g,
1 cm3, mmol). Isolated yield 0.886 g, 98.4%. Amber clear oil.
1H NMR (400 MHz, CDCl3, δ/ppm): 1.77–1.83 (1 H, m),
1.09–1.57 (57 H, m), 0.88 (12 H, t, J 6.63), 3.28–3.60 (10 H, m).
13C NMR (100 MHz, CDCl3, δ/ppm): 72.0, 70.5, 53.5, 50.7, 33.8,
Di(2-butyl-octyl)dimethylammonium chloride
([C8,β4C8,β4C1C1N][Cl])
The synthesis was carried out following the ESI General pro- 33.2, 31.9, 31.8, 29.7, 29.7, 29.6, 29.6, 29.3, 29.2, 26.3, 22.7,
cedure 13† with di(2-butyl-octyl)methylamine (0.843 g, 22.6, 14.1, 14.1. IR (Diamond-ATR, neat, cm−1): 2954, 2921,
2.3 mmol), calcium chloride (95%, 0.6 g, 5.4 mmol), dimethyl 2852, 1673, 1485, 1465, 1377, 1076, 966, 917, 721, 587, 489,
9782 | Org. Biomol. Chem., 2019, 17, 9778–9791
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
466, 459, 434, 414. HRMS calcd for [C38H80N][Cl]: 585.5979
found: [M − 35] 550.6298.
L-Lysine is present in the class I aldolase active site as a
catalytic enamine forming moiety.46 Hence, it is foreseeable
that L-lysine itself could be a catalyst for aldol condensation. In
several protocols, L-lysine was reported to result in quantitative
conversion to the target product under mild conditions.38,39
Thus, we focused on L-lysine as the catalyst, also in relation to
its availability, low-cost, and safety profile.38 We performed
various tests to reproduce and improve the literature results
from a lab-scale to an industrially scalable green process.38
We performed a screening of solvents beneficial for the
process and the environment (Table 1, ESI General procedure
1†).47–51 The solvent choice affected both the reaction rate and
selectivity. We achieved selective conversion in weak and
mildly polar solvents (i.e. ethanol, isopropanol). Amongst the
tested solvents, most alcoholic solvents gave quantitative con-
version within 6 h at room temperature (20–22 °C). In alco-
holic solvents, the reaction rate showed an inverse proportion-
ality to the solvent chain length and branching. Overall, iso-
propanol and ethanol were the best performers. In all tested
solvents, the reaction achieved a selective conversion towards
the E-stereoisomer, confirmed via 1H NMR coupling
constants.52,53 This stereoselectivity appears therefore not to
be influenced by the solvent.
Di(2-octyl-dodecyl)dimethylammonium chloride
([C12,β8C12,β8C1C1N][Cl])
The synthesis was carried out following the ESI General pro-
cedure 13† with di(2-octyl-dodecyl)methylamine (0.826 g,
1.4 mmol), calcium chloride (95%, 0.6 g, 5.4 mmol), dimethyl
carbonate (2.80 g, 3 cm3, mmol), and methanol (0.792 g,
1 cm3, mmol). Isolated yield 0.757 g, 84.5%. Dark brown oil.
Density: 0.82628
g
cm−3 (20.02 °C). Dynamic viscosity:
48.27 mPa s−1 (20.00 °C), kinetic viscosity: 58.42 mm2 s−1
(20.00 °C). 1H NMR (400 MHz, CDCl3, δ/ppm): 3.38–3.51 (10 H,
m), 1.80 (1 H, d, J 5.67), 1.21–1.51 (65 H, m), 0.88 (12 H, dd, J
6.97, 2.08). 13C NMR (100 MHz, CDCl3, δ/ppm): 71.9, 70.5,
53.6, 50.7, 33.8, 33.2, 31.8, 31.8, 29.7, 29.5, 29.4, 29.3, 26.3,
26.3, 22.7, 22.6, 14.1, 14.1. IR (Diamond-ATR, neat, cm−1):
3353, 2954, 2922, 2853, 1713, 1668, 1652, 1485, 1465, 1377,
969, 894, 722, 588, 564, 531, 508, 488, 476, 468, 458, 440, 421,
414, 404. HRMS calcd for [C42H88N][Cl]: 641.6605 found:
[M − 35] 606.6912.
Results and discussion
Step I. Alcohol oxidation
In the literature, complete conversion and full selectivity is
reported for the oxidation of linear and branched alcohols to
aldehydes.28–30 The process is virtually optimal and sustain-
able, so, competing with these research studies was not an
objective of our research. Biermann et al. demonstrated the
feasibility of this step, in tandem to the following step.13,31,32
Moreover, aldehydes themselves are abundantly available
natural products (e.g. cinnamaldehyde, vanillin, citral,
piperonal).
Table 1 Solvent effect on aldol conversion
Solvent
Yield (%)
Cyclohexane
42 (24 h)f
0 (6 h)f
Diethyl ether
Methyl tert-butyl ether
2-Methyl-THF
Dimethyl sulphoxide (DMSO)
Water
Acetonitrile
Methanol
0 (6 h)f
0 (6 h)c,e, f
16 (24 h)
10 (3 h)e
0 (6 h)d, f
53 (24 h)e
≈99 (6 h)a,b,c, f
≈99 (6 h)a,b,c, f
≈99 (6 h)a,b,c, fg
≈99 (6 h)a,b,c, f
≈99 (6 h)a,b,c, f
41 (6 h)f
Ethanol
Step II. Aldol condensation
Propyl alcohol
Isopropyl alcohol
1-Butanol
2-Butanol
tert-Butanol
The aldol condensation is a two-step reaction directed to the
formation of a C–C double bond in β to the aldehyde group.
The homo aldol condensation is an important industrial reac-
tion giving access to α,β-unsaturated aldehydes, which are a
versatile synthetic class of compounds.33 α,β-Unsaturated alde-
hydes are very reactive, which turns them into useful inter-
mediates for organic synthesis, but makes them also sensitive
to both light and oxygen.19,33,34
Glycerol
7 (6 h)f
Methyl acetate
Propyl acetate
Isopropyl acetate
tert-Butyl acetate
Acetone
0 (6 h)f
0 (6 h)f
6–7% (6 h)f
0 (6 h)f
2 (24 h)
Dimethyl carbonate (DMC)
0 (6 h)f
In the literature, the aldol condensation was carried out
with various catalysts, amongst which were inorganic bases,35
Lewis acids,36 amines,37 and amino-acids.38,39 Amino-acid cat-
alysed condensations are appealing because of the low-cost,
high turnover and low-risk safety profile, which grants them
the status of green bio-catalysts. Amino acids are reported as
efficient catalysts in enamine-activated reactions.14,40–43 Amino
acids hold a privileged position within bio-catalysis, since they
are bearing both an amino and an acid function.44 This flexi-
bility, coupled to the robustness of the catalyst, allowed the
usage of amino acids in many multi-step one-pot reactions.45
a Only solvent and product signals observed. b Trace signals of inter-
mediate and/or starting material observed. c Terminal carbon area not
viable. Yield deduced not by signal ratio but by the absence of side-
product signals or starting materials. d Very broad signals, only quali-
tative result. e Byproduct observed. f Non-deuterated solvent.
g Confirmed via GC-MS. The reported yield is a rough estimate
obtained by the peak ratio between the ratio of the integration of the
terminal carbon area (i.e. 0.71–1.06 ppm) and the unsaturated alde-
hyde signals (i.e. 9.36, 6.36 ppm). In the evaluation it has to be con-
sidered that the aldehydic proton is labile and in a basic environment,
signal intensity comparison is best between proximal areas, and strong
signals near the ones in analysis may influence the overall integration
results. Signal integration has a minimum error above 5%.
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9778–9791 | 9783

View Article Online
Paper
Organic & Biomolecular Chemistry
Reaction by-products were observed in polar protic solvents to study the substrate scope of the reaction. Not having hydro-
(i.e. methanol, water, acetonitrile). The stabilisation of polar gen atoms in the position α to the aldehyde, the aldol reaction
intermediates and solvent coordination to the carbonyl oxygen product resulting from 2-ethyl-1-butanal could not undergo
may be some of the underlying factors responsible for the the dehydration step. The reaction achieved only partial con-
undesired condensation reactions (i.e. Mannich, cyclic aldols, version to the 2,2,4-triethyl-3-hydroxyhexanal derivative after
lactones and glycols). Solvents containing an ester or ketone several days. The low yield may be caused by steric hindrance.
group curtailed the L-lysine catalytic activity. This could be due Moreover, computational thermodynamics analysis showed
to a competitive inhibition caused by the solvent in the aldol condensation to be favoured by the exergonic dehydra-
L-lysine interaction with the aldehyde.17–19
tion step, which stabilises products over reagents, supporting
We tested catalyst loading at various concentrations report- and exploiting the reduced reactivity.14 All of these results
ing quantitative conversion within a reasonable time scale suggest that the dehydration step is rate-limiting. Under
(≈6 h) with a minimum of 0.05 eq. of L-lysine (ESI General pro- current conditions, the process is therefore limited to alde-
cedure 2†). Below this concentration, the reaction was signifi- hydes comprising two hydrogen atoms in the α position with
cantly slower. In order to reduce solvent consumption and to respect to the aldehyde functionality. Branched aldehydes
facilitate the upscaling of the reaction, we also optimised the such as 3,7-dimethyl-oct-6-enal(citronellal) and 3,7-dimethyl-
substrate concentration (ESI General procedure 3†). Reactions octanal(dihydrocitronellal), having their branching further in
at about and below a concentration of 5 mol L−1 resulted in the chain, were also tested; they showed quantitative conver-
quantitative conversion without substantial by-product sion in about 6 h at 40 °C. Citronellal therefore showed catalyst
formation.
compatibility thanks to its isolated double bonds and minor
A direct proportionality between the reaction temperature steric hindrance.
and the reaction rate was observed. This was deduced by the
We tested the reaction up to 2 mol scale without major vari-
initial higher conversion rate (Fig. 1) which occurred in conjunc- ations (ESI General procedure 5†). The exothermic reaction
tion with a noticeable heat release from the reaction pot. This may need temperature control at larger scales. We found
may be indicative of a self-catalytic reaction, in which the heat mixing not to be critical to the reaction outcomes (vide ESI
released by the exothermic condensation step accelerates the General procedure 1†). α,β-Unsaturated aldehydes are sensitive
imino-intermediate formation with L-lysine, favouring the overall to both light and oxygen. Because of this sensitivity, our
reaction conversion. Under the same conditions, substrates with attempts to isolate them caused partial product decompo-
longer chains required longer reaction times. Attempting the sition. Isolation was possible via flash chromatography or via
reaction at 40 °C allowed us to observe quantitative conversion batch solvent extraction.38 Thus, the avoidance of intermediate
and selectivity for all the substrates within 4.5–6 h.
purification would improve process sustainability. Hence, we
We tested the reaction scope under the optimised con- avoided isolation and used the crude adduct directly in the
ditions (ESI General procedure 4†). We used 2-ethyl-1-butanal next step.
Step III. Chemoselective alkene reduction
Selective reduction of α,β-unsaturated aldehydes was reported
using a wide variety of catalysts and reaction conditions.54–56
Interest on these compounds rises from their use as versatile
precursors. Some examples of possible derivatives are unsatu-
rated alcohols, saturated alcohols, and saturated aldehydes. In
our study, the latter product was synthesised via a catalytic
chemoselective alkene reduction.
Both palladium on carbon (Pd/C) and nickel sulphide
(Nix–Sy) were reported as selective catalysts for the reduction of
2-ethyl-2-hexenal under 0.9 MPa of hydrogen pressure.55 In a
benchmark study of 15 catalysts tested for the reduction of
2-ethyl-2-hexenal, palladium on carbon showed the highest
degree of selectivity towards the saturated aldehyde under
harsh conditions (240 °C, 7 h, 6 MPa).57 On the other hand,
under mild conditions, palladium on carbon displayed full
Fig. 1 Reaction screening of L-lysine catalysed hexanal condensation at
room temperature (22 °C) via 1H NMR in deuterochloroform. The signal
selectivity for the aldehyde derivative with various substrates
intensity was normalised against the integration range 1.08–0.73 ppm. (i.e. cinnamaldehyde, citral).58,59 Although nickel has a higher
The linear aldehyde (
orange full circle marker; 1H NMR, 1H(s)
elemental abundance, we preferred palladium on carbon
because of the high selectivity, high activity at near atmos-
pheric pressure, and reduced health risk profile. On these pre-
mises, we proceeded in further optimising the reaction
conditions.
9.71–9.58 ppm) reacts with L-lysine forming an enamino-adduct (
blue triangle marker; 1H NMR, 1H(s) 10.07–9.97 ppm), which condenses
with a second aldehyde equivalent to form the α,β-unsaturated aldehyde
(
red star marker, 1H NMR, 1H(s) 9.30–9.19 ppm;
green cross
marker, 1H NMR, 1H(t) 6.50–6.20 ppm).
9784 | Org. Biomol. Chem., 2019, 17, 9778–9791
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
We achieved a complete conversion within 48 hours at extended the reaction scope to primary and secondary
atmospheric pressure with a minimum of 3% w/w of the cata- amines.61 In 1930s, a renewed interest encouraged investi-
lyst (Pd/C with 10% Pd loading) (ESI General procedure 6†). In gations on the reaction pathway and the optimal reaction
contrast to what has been reported, the palladium catalyst conditions.62–69 These studies tested a wide variety of con-
showed gradual deactivation and we needed to reactivate the ditions and showed a remarkable variability of the reaction
catalyst after use (ESI General procedure 7†).55 The de- yields in relation to minor procedural variations (i.e. water
activation rate was limited and dependent on the reaction con- content, amine, temperature, catalysts, addition).63 Because of
ditions. After two days in contact with the α,β-unsaturated this variability, the reaction pathway was never fully identified.
aldehyde, we observed no further catalytic activity. Hence,
Following the reported results, we tailored the reaction to
facilitating the hydrogen transfer to the liquid phase was our substrates. In general, the Leuckart reaction forms primary
central to outrun catalyst deactivation. This could be achieved amines in good to excellent yields. However, we achieved ter-
via increasing the interfacial area between the heterogeneous tiary amines with traces of primary amines, secondary amines,
phases through vigorous mixing and via adjusting the topology and their formyl derivatives. A similar product distribution
of the reaction vessel. This was achieved by moving from stir- was previously reported by Horii et al. for the reaction of
ring the solution in a round bottom flask to a conical flask hydrophobic aldehydes (i.e. 1-hexanal), but using urea and
shaken on an orbital shaker, which increased the liquid–gas formic acid as in situ precursors of formamide.70
interfacial area, favouring hydrogen absorption by the
solution.
The substrates used in this study were immiscible in formic
acid and water, suggesting that hydrophobicity might be a key
After use, the catalyst was separated via vacuum filtration element of the divergence with the literature results.60,62 This
on Celite. The catalyst was removed from the Celite pad and allowed us to define a straightforward procedure towards ter-
reactivated via methanol washes (ESI General procedure 7†). tiary amines.
The catalyst could thus be recycled up to 5 times without a
The reaction optimisation was carried out with 2-butyl-
remarkable difference in activity (the test was carried out at octan-1-al as the substrate. Initial reaction screening was done
0.1 mol scale). We tested the reaction up to one mole of via the microwave procedure reported by Loupy et al.65 This
product. Furthermore, telescoping was applicable allowing us procedure was found to be effective with Guerbet aldehydes,
to directly couple aldol condensation and alkene reduction in achieving a mixture of amines and formyl amines with a sig-
a one-pot two step reaction (ESI General procedure 8†).
Purification of the crude sample was carried out via a batch
nificant component of the tertiary amine.
By cross-comparing the results achieved by Kost et al., Horii
extraction followed by a high vacuum distillation (ESI General et al. and Loupy et al., a set of favourable conditions was
procedure 10†). With the exception of 2-butyl-octan-1-al, direct devised to facilitate a batch Leuckart reaction.62,65,70 Two sets
distillation was not applicable for the synthesised compounds. of conditions were tested: in the former, all the reagents were
Under the conditions of distillation, thermal decomposition mixed at once, and in the latter, the reagents were dropped
processes impeded the achievement of the vacuum levels onto the hot amide over a period of 1.5 h.
needed for the distillation and the material decomposed
The one-pot procedure was ineffective, achieving an adduct
before getting distilled. Batch extraction was beneficial for unstable to the formyl acidic cleavage procedure (HCl, 100 °C,
1
avoiding the decomposition processes. Aqueous solutions of 1 h). The H NMR spectra of the adduct were compatible with
organic and inorganic acids were tested as work-up pro- a double imino urea intermediate (we achieved similar results
cedures, but they were not able to remove the impurities from with the procedure of Yang et al.).71 This hypothesis was also
the organic layer (i.e. 10% v/v of hydrochloric acid, sulphuric compatible with a reported equilibrium between urea, formic
acid, formic acid, phosphoric acid, and acetic acid). The crude acid and formamide (Scheme 4).70 However, full material
reaction mixture was miscible with all common organic sol- characterisation was not carried out because it was out of the
vents (i.e. alcohols, esters, ethers, alkanes, acetone, DMSO, scope of our research. Adding the reagents to the hot amide
DMF, dimethylcarbonate). In further miscibility tests, formic achieved instead a mixture of amines and formyl amines (ESI
acid was found to be poorly miscible with the aldehyde and General procedure 12†). To reach full conversion, the reaction
able to remove the impurities from the crude reaction mixture time was adjusted to 3.5 h.
via catalysing condensation processes. This allowed the defi-
The purification was realised via an anti-solvent separation
nition of an extractive reactive purification procedure for the of the tertiary amine. In methanol, the tertiary amine phase-
reaction mixture. Then, distillation was used to ensure com- separated, allowing for an easy purification of the final
plete purification of the β branched aldehyde and to avoid product. Because the crude mixture could easily form a sus-
possible degradative processes (e.g. aldol condensation).
Step IV. Modified Leuckart–Wallach reaction
By reacting a mixture of formamide and formic acid with an
aldehyde or a ketone, the Leuckart reaction yields a mixture of
amines and formyl amines.60 The reaction is also known as
Scheme 4 General reaction scheme of the interconversion between
the Leuckart–Wallach reaction after Otto Wallach, who urea and formamide in the presence of formic acid.
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9778–9791 | 9785

View Article Online
Paper
Organic & Biomolecular Chemistry
pension in methanol, we worked with small volumes to easily temperature (i.e. 170 °C), a fishy smell, lower yields, and a
break the emulsion via centrifugation of the sample. The con- darker colour developed, which suggested a partial decompo-
centrated methanol washings yielded a mixture of amines and sition of the amino starting material. Below the optimal reac-
formyl amines, which could partly substitute the tion temperature (i.e. 120 °C), the reaction did not occur.
Guerbet aldehyde in following batches. This variation did not
At 145 °C, the reaction did not reach completion,
significantly affect the final product yield. We then attempted suggesting that tuning the reaction conditions could help in
the substitution of formamide with N-methyl formamide. This forcing the reaction equilibrium (e.g. an increase of
modification was successful, allowing us to synthesise the pressure).14,79 To shift the reaction equilibrium to the direc-
monomethyl derivative (ESI General procedure 11†). Also, the tion of the products, we used calcium chloride as an in situ
substitution of formamide with N,N-dimethyl formamide was metathesis reagent. Under these conditions, we observed com-
tested but it did not yield the dimethyl amino derivative.
plete conversion after 48 hours, allowing for the isolation of
the ionic liquid in high yields. Then, the excess of calcium
chloride was removed via anti-solvent precipitation and fil-
tration (ESI General procedure 13†).
Step V. Dimethyl carbonate quaternisation
Methylation reactions commonly involve carcinogenous
chemicals. Over the latest years, the use of dimethyl carbonate
(DMC) as a green and effective methylation agent has been
described as a safer alternative to noxious chemicals such as
methyl halides, dimethyl sulphate or trimethyloxonium
tetrafluoroborate.72–74 Clear advantages are the harmlessness
and low cost of this material.75,76
Methylation with dimethyl carbonate occurs at above
120 °C, whereas acylation is the main reaction occurring at
about 90 °C (Scheme 5).74 β-Branched amines required harsh
reaction conditions for the methylation process. Due to the
immiscibility of the reagents, the reaction kinetics were slow,
but, at high pressure and temperature, the thermodynamics
pushed the reaction to completion.77 This approach is also
advantageous for the eventual metathesis process, which
happens because the driving force of metathesis is the methyl
carbonate anion decomposition. Under standard conditions,
the hydrolysis of methyl carbonate to carbon dioxide and
methanol is irreversible. This forces the metathesis and, in the
presence of an adequate excess of acid, it allows to attain a
quantitative conversion, leaving only trace impurities.
Therefore, an autoclave methylation was carried out with
dimethyl carbonate, which could be eventually adapted to an
industrial scale.
Physico-chemical characterisation
We carried out preliminary characterisation on some pro-
perties relevant to extractive metallurgy. Our analysis focused
on practical usability, chemical stability and both environ-
mental and economic viability of the process. Therefore, four
main features were addressed: rheology, thermal stability,
chemical stability, and water phase interactions (i.e. solubility,
saturation, phase interactions).
Rheology addressed the physical usability of the materials.
This allowed the assessment of the most appropriate con-
ditions of usage (e.g. viscous materials require either dilution
or high working temperatures). Moreover, β-branched
Guerbet alkyl chains generally hold a remarkably wide liquid
range, whose characterisation is valuable in the effort of
understanding these novel materials. Stability to process con-
ditions is a requirement for the development of a new process.
Hence, both chemical and thermal stabilities were evaluated.
Solubility in water was measured to assess the possible process
losses and the eventual recovery steps.
In the following paragraphs, we will analyse first the extrac-
tants (i.e. tertiary amines) and then the ionic liquids (i.e.
ammonium chlorides). If not stated otherwise, characteris-
ation followed the guidelines defined by OECD.80
We ran test reactions in a 15 cm3 polytetrafluoroethylene-
lined autoclave. The use of a lining excluded any possible
metal-catalysed acceleration of dimethyl carbonate decompo-
sition to dimethyl ether and methyl acetate.78 We tested
different reaction temperatures, achieving optimal reaction
conditions at 145 °C (ESI General procedure 13†). Above this
Amines
A viscosimetric analysis was carried out via a rolling-ball visco-
simeter (Fig. 2). By comparing compounds holding about the
same number of carbon atoms (i.e. tri(2-ethylhexyl)amine, di
(2-butyloctyl)amine and tris(octyl)amine), the viscosimetric
analysis shows a clear increase in viscosity with the increase in
chain branching. The difference in viscosity is significant at
room temperature, but on heating, the values lessen and con-
verge at 60 °C. The convergence is especially apparent for the
β-branched derivatives, whereas the linear homologue always
has a slightly lower viscosity. The reported viscosities are mod-
erate, suggesting a possible use of the materials also as
undiluted substances in metallurgical processes.
Densitometry was carried out via an oscillation densi-
tometer (Fig. 2). The density of the synthesised materials
ranges between 0.815–0.835 g cm−3 at 20 °C and 0.790–0.810 g
cm−3 at 60 °C. The reduced densities suggest that in a biphasic
Scheme 5 Nucleophilic substitution of dimethyl carbonate. Reaction
mechanism is outlined for BAC2 and BAL2, which takes place at about
90 °C and above 120 °C respectively. The methylation temperature has a
strong dependency on the nucleophile strength.
9786 | Org. Biomol. Chem., 2019, 17, 9778–9791
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
below the limit of detection of the titrant solution (below 1 mg
dm−3). To further verify our assumptions, an infrared spec-
trum of the aqueous layer was recorded. The total absence of
signals corresponding to the amino material confirmed a com-
plete insolubility in water.
To assess the chemical stability of the extractants to an
alkaline environment, a shaking test with aqueous and metha-
nolic solutions of sodium hydroxide was carried out (ESI
General procedure 15 and 16†). After shaking the samples for
one day at 55 °C, the methanol solutions changed from the
starting turbid white to a clear yellow whereas the aqueous
solution remained clear and colourless. The methanolic solu-
tion was diluted with water and then both the methanolic and
the aqueous solutions were extracted with deuterochloroform
and subjected to 1H NMR analysis.
For di(2-butyl-octyl)methylamine and tri(2-butyl-octyl)
amine, neither the methanolic system nor the aqueous solu-
tion showed significant decomposition, suggesting the virtual
absence of decomposition under the tested conditions and the
potential suitability of the two solvent systems for extractive
metallurgical processes under alkaline conditions.
To evaluate the material thermal stability, a set of dynamic
thermogravimetric analyses (TGA) was carried out (Fig. 3).
Thus, the amines showed an increased temperature stability
with an increase of the molecular mass. The amine onset
decomposition temperature ranges between 200 and 350 °C;
however, this is often an overestimation of the actual material
stability over a long term usage.81–85 Yet, decomposition temp-
eratures fell well above the standard working temperatures,
suggesting a stability well beyond the one needed for the
aimed application (Table 2).
Differential scanning calorimetry (DSC) was also carried out
in order to determine the phase behaviour of the synthesised
materials in the temperature range between −40 and 120 °C.
All of the synthesised amine had no phase transition in the
assayed range with the exception of tri(2-octyl-dodecyl)amine,
which melted at about −15 °C.
Fig. 2 Viscosimetric and densitometric analysis of the synthesised
amines. The measurements were carried out between 20 and 60 °C via
a scan of 2.5 degree step. In light blue are shown the di(alkyl)methyl-
amines and in green the tri(alkyl)amines. Each alkyl chain is depicted
with a different line-style (e.g. di(2-butyl-octyl)methylamine and tri(2-
butyl-octyl)amine are depicted with a full line).
Ionic liquids
system the material would generally be on top of a polar Water solubility tests were carried out on the lightest derivative
solvent, which generally holds a higher density (i.e. water and (i.e.
di(2-butyl-octyl)dimethylammonium chloride).
methanol).
Preliminary tests carried out via visual dissolution of a known
Preliminary solubility tests in water were carried out via a amount of sample showed a water solubility below 1 g dm−3
.
visual observation of the dissolution of a known amount of Further assessment was hampered by the formation of a room
sample substance. All the synthesised compounds showed temperature stable emulsion (Fig. 4[b]). Heating up the sample
solubilities below 1 g dm−3. Then, we tested the lightest deriva- caused the resolution of the emulsion, though, the saturated
tives in a more accurate assessment of their water solubility solutions appeared hazy even after several days (Fig. 4[c]). This
(i.e. di(2-butyl-octyl)methylamine and tri(2-butyl-octyl)amine).
effect was attributed to the molecules of water trapped within
The solubility in water was determined via a volumetric the ionic liquid, which would be also consistent with the high
titration of an aqueous solution saturated with the substance amount of water retained in the saturated samples (Table 2).
in analysis (OECD procedure, flask method). The amine con-
When tested by FTIR spectroscopy, the cloudy aqueous
centration in the water layer was characterised via an acid– layer showed the trace presence of the ionic liquid.
base titration with a solution of sulphuric acid (ESI General
procedure 14†).
After centrifugation (5000 rpm, 20 min), the water phase
appeared to be clear and virtually no signal of the ionic liquid
Direct titration of the amine saturated solution led to incon- was detected with FTIR spectroscopy. This confirmed the full
clusive results, which were indicative of a titrate concentration water insolubility of the ionic liquid. Although emulsion for-
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9778–9791 | 9787

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 2 Thermal stability of amines and quaternary ammonium ionic
liquids
Water saturation
(% w/w)
Tonset (°C)
Di(2-butyl-octyl)methylamine
Di(2-pentyl-nonyl)methylamine
Di(2-hexyl-decyl)methylamine
Di(2-heptyl-undecyl)methylamine
Di(2-octyl-dodecyl)methylamine
Tri(2-butyl-octyl)amine
Tri(2-pentyl-nonyl)amine
Tri(2-hexyl-decyl)amine
Tri(2-heptyl-undecyl)amine
Tri(2-octyl-dodecyl)amine
Di(2-butyl-octyl)dimethyl
ammonium chloride
—
—
—
—
—
—
—
—
214
238
271
288
308
255
286
299
316
332
168
—
—
29.26
Di(2-pentyl-nonyl)dimethyl
ammonium chloride
Di(2-hexyl-decyl)dimethyl
ammonium chloride
Di(2-heptyl-undecyl)dimethyl
ammonium chloride
Di(2-octyl-dodecyl)dimethyl
ammonium chloride
23.02
11.27
—
171
192
177
169
—
Degradation onset temperature determined via the step-tangent
method, measured under a nitrogen atmosphere with a heating rate of
10 °C min−1 between 20 and 500 °C. Water saturation limit determined
via Karl Fischer titration. Water saturation determination was discon-
tinued from di(2-heptyl-undecyl)dimethylammonium chloride because
of the formation of a stable emulsion.
Fig. 3 Dynamic TGA thermographs of the synthesised compounds
measured under a nitrogen flow of 60 cm3 min−1 with a heating rate of
10 °C min−1 between 20 and 500 °C. In the legend the ammonium ions
are reported with the following notation: di(2-butyl-octyl)dimethyl-
ammonium chloride ([C8,β4C8,β4C1C1N][Cl]), di(2-pentyl-nonyl)dimethyl-
ammonium chloride ([C9,β5C9,β5C1C1N][Cl]), di(2-hexyl-decyl)dimethyl-
ammonium chloride ([C10,β6C10,β6C1C1N][Cl]), di(2-heptyl-undecyl)di-
methylammonium chloride ([C11,β7C11,β7C1C1N][Cl]), and di(2-octyl-
dodecyl)dimethylammonium chloride ([C12,β8C12,β8C1C1N][Cl]).
Fig. 4 Phase behaviour of di(2-butyloctyl)dimethylammonium chloride
in contact with water. [a]. Before shaking. [b]. After shaking. [c]. After
shaking and heating at 70 °C.
mation was experienced with pure water, aqueous solutions of
inorganic salts did not form emulsions. This could be
explained by a salting-out effect.
To evaluate the water content of the saturated ionic liquid,
each ionic liquid (0.5 cm3) was presaturated via shaking with
Milli-Q water (1 cm3). To favour phase separation, the samples
were heated at 80 °C for 30 min. Using this treatment, we were
not able to break the emulsions of the two heaviest derivatives
(i.e. di(2-heptyl-undecyl)dimethylammonium chloride and di
(2-octyl-dodecyl)dimethylammonium chloride). Hence, we
attempted to remove the emulsion via centrifugation, but this
did not break it (5000 rpm, 20 min). Therefore, these samples
were not characterised for water content. The water content of
the three remaining ionic liquids was measured via Karl
Fischer titration. The amount of water in the ionic liquid
phase showed a reverse trend with respect to increasing chain
length (Table 2).
General procedure 18†). All the ammonium ions had a similar
stability with an onset decomposition temperature ranging
between 170 and 200 °C (Table 2). Different from the amines,
the increase in the molecular mass did not correlate with an
increased thermal stability. All the ionic liquids showed a
slope discontinuity in the thermograph (Fig. 3). A similar
trend was previously observed in some classes of tri(octyl)
methylammonium ionic liquids reported in the literature.86
This discontinuity suggests the decomposition of another
material with a higher stability with respect to the ammonium
ion. If the decomposition of the ammonium ion occurs via a
Hofmann elimination mechanism, the decomposition product
would be a tertiary amine (Scheme 2). In line with the pre-
viously acquired results, a tertiary amine would have a higher
onset decomposition temperature than the ammonium ion
To evaluate the thermal stability of the ionic liquids, a set
of dynamic thermogravimetric analyses was carried out (ESI
9788 | Org. Biomol. Chem., 2019, 17, 9778–9791
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
and this could explain the curve discontinuity. Moreover, the with the sodium hydroxide solution in analysis. All of the ionic
analyses showed in some samples the presence of a trace quan- liquids, but di(2-butyl-octyl)dimethylammonium chloride,
tity of compounds stable at 500 °C. These were supposedly in- were stable under the tested conditions to aqueous alkaline
organic salts, residues of the in situ metathesis. These results solutions.
suggest that the heavier derivatives may be more prone to with-
hold salts from precipitating in ethyl acetate. DSC was carried
out in order to determine the phase behaviour of the syn-
Conflicts of interest
thesised materials in the temperature range between 0 and
120 °C. All of the synthesised ionic liquids did not show a There are no conflicts to declare.
phase transition in the investigated temperature range.
Chemical stability towards strong bases was determined on
all the derivatives. The base stability was determined via
shaking the sample with two solutions of sodium hydroxide:
Acknowledgements
one in water and one in methanol. After shaking the samples The research leading to these results received funding from
for one day (2000 rpm, 55 °C), the methanol solutions the Horizon 2020 Programme ([H2020/2014-2019]) of the
changed from the starting turbid white to a clear yellow European Community under Grant Agreement no. 721385
whereas the aqueous solution remained clear and colourless. (MSCA-ETN SOCRATES). This publication reflects only the view
Then, the methanolic solution was diluted with water and of the authors, the European Community is exempted from
both the methanolic and the aqueous solutions were extracted any liability. Project website: http://etn-socrates.eu/. High
with deuterochloroform and subjected to 1H NMR analysis.
resolution mass spectrometry was made possible by the
The amount of decomposition decreased with increasing support of the Hercules Foundation of the Flemish
chain length. The methanolic system showed significant Government under Grant Agreement no. 20100225-7.
decomposition for all the derivatives (i.e. ∼10–50%), whereas
the aqueous system showed significant decomposition only for
di(2-butyl-octyl)dimethylammonium chloride (i.e. ∼20%). This
confirmed that base stability could be improved by an increase
References
in steric hindrance around the β carbon of the alkyl chain.
1 European Commission (EC), Next steps for a sustainable
European future: European action for sustainability, Brussels,
Belgium, 2016.
2 European Commission (EC), Sustainable Agriculture,
Forestry and Fisheries in the Bioeconomy: A Challenge for
Europe, Brussels, Belgium, 2015.
3 L. Wu, T. Moteki, A. A. Gokhale, D. W. Flaherty and
F. D. Toste, Chem, 2016, 1, 32–58.
4 P. Badger, in Trends in new crops and new uses, ed. J. Janick
and A. Whipkey, ASHS Press, Alexandria, VA, 2002, pp.
17–21.
5 B. Kamm and M. Kamm, Appl. Microbiol. Biotechnol., 2004,
64, 137–145.
6 M. Guerbet, C. R. Acad. Sci., Ser. IIc: Chim., 1909, 129–132.
7 A. J. O’Lenick Jr. and A. J. O’Lenick, J. Surfactants Deterg.,
2001, 4, 311–315.
8 F. L. Moore, Anal. Chem., 1957, 29, 1660–1662.
9 J. Pernak, M. Smiglak, S. T. Griffin, W. L. Hough,
T. B. Wilson, A. Pernak, J. Zabielska-Matejuk,
A. Fojutowski, K. Kita and R. D. Rogers, Green Chem., 2006,
8, 798–806.
Conclusions
A wide variety of extractants is commercially available but
many of them are generally petroleum-derived. Besides, many
of the retail available ammonium extractants cannot withstand
alkaline conditions; hence, their use is restrained to metal
recovery processes under either neutral or acidic conditions.
We developed an effective procedure to synthesise base
stable extractants from renewable feedstock. To the best of our
knowledge, this is the first study to synthesise and investigate
the physico-chemical behaviour of Guerbet amines and
ammonium ions. Guerbet aldehydes were targeted as func-
tional precursors of the amino derivatives. They were syn-
thesised in good yield and high purity via a green procedure
easily transferrable to a large scale with minor changes. Then,
Guerbet aldehydes were converted to tertiary amines via an
effective, simple, and moderately yielding scalable procedure.
The moderate selectivity of the reaction conditions was com-
plemented by the recyclability of side products, which avoided 10 D. M. Eike, J. F. Brennecke and E. J. Maginn, Green Chem.,
material losses. Then, the synthesised di(alkyl)methylamines 2003, 5, 323–328.
were quaternised via a near quantitative green dimethyl car- 11 R. Golden and M. Valentini, Regul. Toxicol. Pharmacol.,
bonate methylation procedure. 2014, 69, 178–186.
All the synthesised extractants were fully hydrophobic and 12 Sasol, Isofol C12–C32 Guerbet alcohols.
stable at elevated temperatures. The branched structure ham- 13 M. Biermann, H. Gruß, W. Hummel and H. Gröger,
pered ordered intermolecular interactions, extending the
ChemCatChem, 2016, 8, 895–899.
extractant liquid range. The stability to alkaline conditions of 14 D. Gabriëls, W. Y. Hernández, B. Sels, P. Van Der Voort and
the ionic liquids was dependent on their mutual solubility
A. Verberckmoes, Catal. Sci. Technol., 2015, 5, 3876–3902.
Org. Biomol. Chem., 2019, 17, 9778–9791 | 9789
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
Organic & Biomolecular Chemistry
15 J. H. Earley, R. A. Bourne, M. J. Watson and M. Poliakoff, 45 A. Bruggink, R. Schoevaart and T. Kieboom, Org. Process
Green Chem., 2015, 17, 3018–3025. Res. Dev., 2003, 7, 622–640.
16 N. M. Eagan, B. M. Moore, D. J. McClelland, A. M. Wittrig, 46 T. D. Beeson, M. Benohoud, J. W. Bode, S. Chen,
E. Canales, M. P. Lanci and G. W. Huber, Green Chem.,
2019, 21, 3300–3318.
M. Christmann, P.-C. Chiang, D. A. DiRocco, Y. C. Fan,
T. Furuta, P. García-García, S. Hatakeyama, Y. Hayashi,
J. Jia, T. Kawabata, N. J. Kerrigan, O. Kwon, Y. Liu,
D. W. C. MacMillan, N. Mase, P. Melchiorre, S. Mukherjee,
A. Piisola, P. M. Pihko, T. A. Ramirez, T. Rovis, E. C. Salo,
Y. Shi, A. D. Smith, K. Suzuki, H. Takikawa, X.-W. Wang,
Y. Wang, A. J. B. Watson, O. A. Wong, P. A. Woods and
S. M. Yliniemelä-Sipari, in Asymmetric Organocatalysis 1, ed.
B. List, Georg Thieme Verlag, Stuttgart, 2012.
17 A. T. Nielsen, J. Org. Chem., 1963, 28, 2115–2119.
18 A. T. Nielsen, J. Am. Chem. Soc., 1957, 79, 2518–2524.
19 A. T. Nielsen, J. Am. Chem. Soc., 1957, 79, 2524–2530.
20 C. M. Moore, O. Staples, R. W. Jenkins, T. J. Brooks,
T. A. Semelsberger and A. D. Sutton, Green Chem., 2017, 19,
169–174.
21 F. G. Mutti, T. Knaus, N. S. Scrutton, M. Breuer and
N. J. Turner, Science, 2015, 349, 1525–1529.
22 M. Biermann, D. Bakonyi, W. Hummel and H. Gröger,
Green Chem., 2017, 19, 405–410.
47 C. M. Alder, J. D. Hayler, R. Henderson, A. Redman,
L. Shukla, L. E. Shuster and H. F. Sneddon, Green Chem.,
2016, 4, 1166–1169.
23 A. W. von Hofmann, Liebigs Ann., 1851, 78, 253–286.
24 A. W. von Hofmann, Proc. R. Soc. London, 1851, 357–398.
25 A. W. von Hofmann, Proc. R. Soc. London, 1850, 93–131.
48 D. Prat, A. Wells, J. Hayler, H. Sneddon, C. R. McElroy,
S. Abou-Shehada and P. J. Dunn, Green Chem., 2016, 18,
288–296.
26 S. Raiguel, D. Depuydt, T. Vander Hoogerstraete, 49 I. T. Horváth, Green Chem., 2008, 10, 1024.
J. Thomas, W. Dehaen and K. Binnemans, Dalton Trans., 50 R. K. Henderson, C. Jiménez-González, D. J. C. Constable,
2017, 46, 5269–5278.
27 K. C. Lethesh, W. Dehaen and K. Binnemans, RSC Adv.,
2014, 4, 4472–4477.
S. R. Alston, G. G. a. Inglis, G. Fisher, J. Sherwood,
S. P. Binks and A. D. Curzons, Green Chem., 2011, 13, 854–
862.
28 Y. Yan, J. Yang, Z. Yu, M. Yu, Y.-T. Ma, L. Wang, C. Su, 51 D. Prat, J. Hayler and A. Wells, Green Chem., 2014, 16,
J. Luo, G. P. Horsman and S. Huang, Nat. Commun., 2016,
7, 13083.
29 M. Shibuya, M. Tomizawa, I. Suzuki and Y. Iwabuchi, J. Am.
Chem. Soc., 2006, 128, 8412–8413.
30 J. M. Hoover and S. S. Stahl, J. Am. Chem. Soc., 2011, 133,
16901–16910.
31 J. M. Hoover, B. L. Ryland and S. S. Stahl, ACS Catal., 2013,
3, 2599–2605.
4546–4551.
52 U. Vögeli, D. Herz and W. Von Philipsborn, Org. Magn.
Reson., 1980, 13, 200–209.
53 U. Vogeli and W. von Philipsborn, Org. Magn. Reson., 1975,
7, 617–627.
54 D. J. Collins, D. E. Grimes and B. H. Davis, Can. J. Chem.
Eng., 1983, 61, 36–39.
55 G. Smedler, Can. J. Chem. Eng., 1989, 67, 51–61.
32 M. A. Iron and A. M. Szpilman, Chem. – Eur. J., 2017, 23, 56 P. N. Rylander and D. R. Steele, US3655777, 1955.
1368–1378.
57 G. Liu, S. Liu, S. Liu, S. Yu, L. Li, F. Liu, C. Xie and X. Song,
Catal. Lett., 2017, 147, 987–995.
58 A. B. Crozon, M. Besson and P. Gallezot, New J. Chem.,
1998, 22, 269–273.
33 G. Desimoni, G. Faita and P. Quadrelli, Chem. Rev., 2018,
118, 2080–2248.
34 M. Häusermann, Helv. Chim. Acta, 1951, 34, 1482–1491.
35 O. Hertel, G. Boettger, W. Koernig, H. Wache, R. Schanz 59 X. Liu, Z. Zhang, Y. Yang, D. Yin, S. Su, D. Lei and J. Yang,
and W. Reiss, DE3231794A1, 1982. Chem. Eng. J., 2015, 263, 290–298.
36 R. Arias-Ugarte, F. S. Wekesa and M. Findlater, Tetrahedron 60 R. Leuckart, J. Prakt. Chem., 1889, 41, 330–340.
Lett., 2015, 56, 2406–2411.
61 H. W. Gibson, Chem. Rev., 1969, 69, 673–692.
62 A. N. Kost and I. I. Grandberg, Zh. Obshch. Khim., 1954,
1377–1381.
63 T. Lejon, I. Helland, W. Liufang, Z. Jinzhong, H. Xiaoying,
J. Møller, A. Senning, X.-K. Yao, H.-G. Wang,
J.-P. Tuchagues and M. Ögren, Acta Chem. Scand., 1999, 53,
76–78.
64 F. S. Crossley and M. L. Moore, J. Org. Chem., 1944, 09,
529–536.
65 A. Loupy, D. Monteux, A. Petit, J. M. Aizpurua,
E. Domínguez and C. Palomo, Tetrahedron Lett., 1996, 37,
8177–8180.
37 S. Hünig, Justus Liebigs Ann. Chem., 1950, 569, 198–226.
38 K. A. Ostrowski, D. Lichte, M. Stuck and A. J. Vorholt,
Tetrahedron, 2016, 72, 592–598.
39 B. Nozière and A. Córdova, J. Phys. Chem. A, 2008, 112,
2827–2837.
40 M. Lombardo, S. Easwar, F. Pasi, C. Trombini and
D. D. Dhavale, Tetrahedron, 2008, 64, 9203–9207.
41 D. B. Ramachary and Y. V. Reddy, Eur. J. Org. Chem., 2012,
865–887.
42 S. Bertelsen, M. Marigo, S. Brandes, P. Dinér and
K. A. Jørgensen, J. Am. Chem. Soc., 2006, 128, 12973–12980.
43 T. D. Beeson and D. W. C. MacMillan, in Asymmetric 66 J. F. Bunnett and J. L. Marks, J. Am. Chem. Soc., 1949, 71,
Organocatalysis 1, ed. B. List, Georg Thieme Verlag,
Stuttgart, 2012.
1587–1589.
67 C. B. Pollard and D. C. Young, J. Org. Chem., 1951, 16, 661–
44 L.-W. Xu and Y. Lu, Org. Biomol. Chem., 2008, 6, 2047–2053.
672.
9790 | Org. Biomol. Chem., 2019, 17, 9778–9791
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
68 W. F. Bruce and V. J. Webers, US2603661, 1952.
69 V. J. Webers and W. F. Bruce, J. Am. Chem. Soc., 1948, 70,
1422–1424.
70 Z. Horii, Y. Tamura and Y. Murakami, Yakugaku Zasshi,
1952, 72, 1206–1208.
71 L. Yang, J. Lin, L. Kang, W. Zhou and D. Y. Ma, Adv. Synth.
Catal., 2018, 360, 485–490.
72 F. Trotta, P. Tundo and G. Moraglio, J. Org. Chem., 1987,
52, 1300–1304.
78 M. Selva, M. Fabris and A. Perosa, Green Chem., 2011, 13,
863–872.
79 R. S. Kalb, M. Damm and S. P. Verevkin, React. Chem. Eng.,
2017, 2, 432–436.
80 OECD Guidelines for the Testing of Chemicals, Section 1:
Physical-Chemical properties, OECD Publishing, Paris, 1981.
81 K. J. Baranyai, G. B. Deacon, D. R. Macfarlane, J. M. Pringle
and J. L. Scott, Aust. J. Chem., 2004, 57, 145–147.
82 D. M. Fox, W. H. Awad, J. W. Gilman, P. H. Maupin, H. C. De
Long and P. C. Trulove, Green Chem., 2003, 5, 724–727.
83 M. T. Clough, K. Geyer, P. A. Hunt, S. Son, U. Vagt and
T. Welton, Green Chem., 2015, 17, 231–243.
73 P. Tundo and M. Selva, Acc. Chem. Res., 2002, 35,
706–716.
74 P. Tundo, Pure Appl. Chem., 2001, 73, 1117–1124.
75 M. Fabris, V. Lucchini, M. Noè, A. Perosa and M. Selva, 84 P. Zhang, Y. Gong, Y. Lv, Y. Guo, Y. Wang, C. Wang and
Chem. – Eur. J., 2009, 15, 12273–12282. H. Li, Chem. Commun., 2012, 48, 2334–2336.
76 G. Fiorani, A. Perosa and M. Selva, Green Chem., 2018, 20, 85 C. Maton, N. De Vos and C. V. Stevens, Chem. Soc. Rev.,
288–322. 2013, 42, 5963–5977.
77 D. Weisshaar, G. Earl, E. Villa, J. Zierke, C. Fry, K. Becvar, 86 J.-P. Mikkola, P. Virtanen and R. Sjöholm, Green Chem.,
S. Li and M. Schafer, Int. J. Chem. Kinet., 2010, 42, 221–225.
2006, 8, 250–255.
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9778–9791 | 9791