Research ArticleChemistry

Direct N-alkylation of unprotected amino acids with alcohols

See allHide authors and affiliations

Science Advances  08 Dec 2017:
Vol. 3, no. 12, eaao6494
DOI: 10.1126/sciadv.aao6494


N-alkyl amino acids find widespread application as highly valuable, renewable building blocks. However, traditional synthesis methodologies to obtain these suffer from serious limitations, providing a major challenge to develop sustainable alternatives. We report the first powerful catalytic strategy for the direct N-alkylation of unprotected α-amino acids with alcohols. This method is highly selective, produces water as the only side product leading to a simple purification procedure, and a variety of α-amino acids are mono- or di-N-alkylated, in most cases with excellent retention of optical purity. The hydrophobicity of the products is tunable, and even simple peptides are selectively alkylated. An iron-catalyzed route to mono-N-alkyl amino acids using renewable fatty alcohols is also described that represents an ideal green transformation for obtaining fully bio-based surfactants.


α-Amino acids are a prominent class of naturally occurring chiral compounds (1) that may serve as renewable alternatives to amines derived from fossil resources (Fig. 1A) (2, 3). In particular, N-alkyl amino acids are highly valuable as chiral building blocks for the synthesis of pharmaceutically active compounds (4), biodegradable polymers (5), ligands for asymmetric catalysis (6), or in other specialized applications (Fig. 1B) (79). N-alkyl amino acids have shown promise as amphoteric surfactants (10). Being commodities, surfactants are produced on a large scale, and there is a pressing need to develop direct sustainable catalytic methods to obtain environmentally friendly (11) and fully bio-based alternatives (12), especially in the context of the bio-based economy (13). Despite the obvious potential of N-alkyl amino acids, there is a complete lack of selective catalytic methods to obtain these compounds via direct functionalization of unprotected α-amino acids. In general, N-alkylation of α-amino acids is performed using stoichiometric methods, such as reductive alkylation with aldehydes, which use inorganic reductants or nucleophilic substitution with alkyl halides (14). These conventional strategies usually suffer from limited availability, versatility, or stability of the starting materials and, in particular, the formation of stoichiometric amounts of by-products and tedious purification procedures (Fig. 2A). Herein, we demonstrate the catalytic direct N-alkylation of amino acids with alcohols as inexpensive and renewable reaction partners in a fully atom-economic process avoiding racemization and producing only water as by-product (Fig. 2B). Setting our design criteria, the ultimate method should be applicable to the exclusive use of biorenewable feedstock amino acids and alcohols, be based on abundant Fe-based catalysts, produce only innocuous water as side product, and directly lead to a common product, that is, a surfactant.

Fig. 1 Sustainable catalytic methods for the synthesis of N-alkyl amino acids and areas of application.

(A) Fossil versus renewable pathways to value-added amines. The nonrenewable path proceeds via the Haber-Bosch process and uses petrochemicals as reaction partners to obtain simple amines, which are further functionalized. The alternative pathway uses natural amino acids, which are directly coupled with alcohols derived from renewable resources to obtain fully bio-based N-alkyl amino acids. The fundamental challenge is the development of efficient and sustainable catalytic methods to enable this step. ee, enantiomeric excess. (B) Various areas of application of N-alkyl amino acids.

Fig. 2 Classical strategies versus sustainable pathways for N-alkylation of unprotected amino acids.

(A) Classical methods for the N-alkylation of α-amino acids. (B) Sustainable, waste-free direct coupling of α-amino acids with alcohols. (C) Proposed mechanism of the direct N-alkylation of α-amino acids with alcohols via the “borrowing hydrogen” strategy.

Alcohols can be derived from renewable resources through the fermentation of carbohydrates (13), catalytic depolymerization of lignocellulose (13, 15), as well as the reduction of fatty acids contained in plant triglycerides (Fig. 1A) (16). Thus, alcohols are widely abundant, ideal starting materials to accomplish the direct N-alkylation of α-amino acids.

A desired catalytic approach to carry out such a privileged N-alkylation protocol is based on the borrowing hydrogen strategy (17, 18) that proceeds via a sequence of reaction steps (Fig. 2C) involving dehydrogenation of the alcohol substrate to yield the corresponding carbonyl compound followed by imine formation and subsequent imine hydrogenation, using an appropriate catalyst. Although a number of transition metal complexes (17) are known to catalyze the coupling of a variety of amines and alcohols using this method, the direct alkylation of amino acids as substrates using homogeneous hydrogen borrowing catalysis is unprecedented.

There are several reasons why the selective N-alkylation of unprotected amino acids, compared to simple amines, is expected to be highly challenging. First, most amino acids have limited solubility in nonpolar organic solvents, and their zwitterionic character renders these substrates sensitive to changes of pH and basic or acidic reagents. Furthermore, competing esterification of the amino acid instead of N-alkylation could take place. Moreover, racemization of the substrate or the formed products under the reaction conditions may be a serious problem, because many hydrogen borrowing methods require the addition of a strong base (17, 18). To the best of our knowledge, only very few examples of N-alkylation or N-allylation of free amino acids are known to mainly use heterogeneous catalysts (1921).


To accomplish the novel N-alkylation of amino acids with alcohols according to the reaction sequence shown in Fig. 2C, we have initially selected a well-defined, homogeneous ruthenium complex, the Shvo catalyst (Cat 1) (22), which has already demonstrated potential in transfer hydrogenation of polar double bonds (23) and trialkylation of ammonia salts with alcohols (24). Because of its unique structure, this complex is capable of bifunctional activation of the alcohol substrate without the need for any additional base, an important requirement for preserving enantiopurity of the amino acid substrate and derived products.

In preliminary experiments, to establish proof of principle for the desired hydrogen borrowing sequence involving Cat 1 (Fig. 2C), we chose the cyclic unprotected amino acid proline (1a) and the simple N-alkylation reagent ethanol (2a) as substrates, whereby ethanol served both as reagent and solvent. The reaction temperature was kept at 90°C to minimize possible competing esterification reactions. Gratifyingly, full conversion of 1a was observed using a catalyst loading as low as 1 mole percent (mol %) Cat 1, and upon simply removing the excess of ethanol, N-ethyl-proline 3aa was obtained in quantitative yield and excellent purity [as evidenced by 1H nuclear magnetic resonance (NMR) measurement; see p. S36 in the Supplementary Materials], not requiring any purification steps. To demonstrate the power of this highly selective N-alkylation methodology, we fully converted a variety of amino acids such as glycine (1b), alanine (1c), valine (1d), leucine (1e), and phenylalanine (1f) to the corresponding N,N-di-alkylated analogs with perfect selectivity (Fig. 3). Even serine (1g), bearing a free hydroxyl functionality at the side chain, provided the desired N,N-diethyl-serine (3ga) in quantitative yield. No product was observed with lysine (1h); however, N6-Ac-lysine (1i) provided the desired product in 74% isolated yield. A crucial requirement for a method for selective modification of amino acids is that the valuable chiral information contained in the starting material should be retained upon functionalization. Excellent ee values (93 to 99%) were measured for products 3aa, 3da, 3ea, and 3fa. Racemization occurred to a small extent in the case of N,N-diethyl-serine (86% ee) and N,N-diethyl-alanine (84% ee), which is very likely due to the dehydrogenation of the amine functionality in alanine and serine or their monoalkylated analogs. The activity of the Shvo catalyst in amine dehydrogenation was previously described by Casey and Johnson (25) and Beller and co-workers (26).

Fig. 3 N-alkylation of amino acids and N-terminal modification of peptides with various alcohols.

See the General procedure in Materials and Methods for the description of the experimental procedure and tables S2 to S5 for further details on these experiments. General reaction conditions: 0.2 mmol of 1, 1 to 5 ml or 0.6 to 2 mmol of 2, 1 mol % Cat 1, neat when using ethanol (2a), 1 ml of CF3CH2OH (2d) when using iPrOH (2b), 18 to 47 hours, and 90°C unless otherwise specified. Isolated yields are shown, and the ee of the product was measured by high-performance liquid chromatography (HPLC) upon established derivatization procedures (see Determination of the optical purity of products in the Supplementary Materials). #Neat; ^CF3CH2OH was used as solvent; a2 mol % Cat 1 was used; b100°C.

Next, a simple secondary alcohol, isopropanol (2b), was applied in the N-alkylation of α-amino acids. Whereas N-isopropyl-proline (3ab) was readily obtained upon functionalization of proline (1a) in neat isopropanol, other amino acids had limited solubility. This prompted us to investigate methanol (2c) and CF3CH2OH (2d) as solvents. CF3CH2OH was found to be an excellent solvent for the isopropyl functionalization of amino acids 1c to 1g to provide products 3cb to 3gb, respectively, in quantitative yield. In all cases, selective monoalkylation was observed, likely due to the steric hindrance created after the insertion of the first isopropyl moiety. This general method allows easy access to mono-N-alkylated amino acids for the synthesis of modified proteins with higher lipophilicity.

Excellent scope was achieved in the functionalization of proline (1a) with diverse aliphatic and aromatic alcohols 2e to 2i (Fig. 3). Notably, the chloro-substituted amino acid derivatives 3ag and 3ai would allow for further functionalization of these building blocks. Subsequently, the functionalization of phenylalanine (1f) with 1-butanol (2e), 1-nonanol (2j), and 1,5-pentane-diol (2k) was readily achieved, yielding the corresponding di-N-alkylated products 3fe, 3fj, and 3fk, respectively. This represents a convenient strategy for modulating the hydrophilicity/lipophilicity of the obtained products (3fk versus 3fj). The modular functionalization of the simplest amino acid, glycine 1b, was further explored. With 2-butanol (2l), selective monoalkylation while using benzylalcohol (3h) dialkylation was observed. Interestingly, with 1-pentanol (2m), selectivity toward the corresponding mono-N-alkylated (3bm′) or di-N-alkylated (3bm) product could be achieved simply by adjusting the glycine–to–1-pentanol ratio. Long-chain N-alkylated amino acids have surfactant properties; however, their synthesis and purification have thus far been proven challenging. Using our methodology, long-chain N-alkylated amino acids are easily synthesized by direct coupling of glycine with 1-nonanol (2j) and 1-dodecanol (2n) to yield 3bj and 3bn, respectively, in excellent yields (>90%).

Encouraged by the generality of the method for the functionalization of amino acids, we envisioned the selective N-terminal modification of simple peptides. On the basis of this new protocol, the hydrophobicity or hydrophilicity of peptides could be easily tuned by introducing either longer-chain alkyl groups or polar moieties such as hydroxyl groups through simple variation of the alcohol reaction partner. First, it was shown that dipeptide glycylalanine (4a) can be quantitatively diethylated (5aa) in neat ethanol using Cat 1, and even the tripeptide leucylglycylglycine (4b) underwent dialkylation on the N terminus, yielding product 5ba. Furthermore, when dipeptide 4a was reacted with 1-dodecanol (2n), the corresponding dialkylated product 5an, a lipophilic dipeptide, was obtained, which can be used for transporting metal ions across cell membranes (8). On the other hand, the reaction of 4a with diol 2k afforded a dipeptide 5ak with increased hydrophilicity. These reactions represent the first examples of the selective and high-yielding dialkylation of peptide substrates on their N terminus with simple alcohols, allowing for extremely easy purification procedures (27). This methodology also has potential for simple N-terminal modification of proteins to affect protein activation and degradation, thereby further diversifying biological functions (28).

The waste-free synthesis of environmentally benign and renewable surfactants is a high-impact and, from an industrial perspective, a particularly relevant area of application for our novel methodology (11, 12, 16). Amino acids serve as versatile building blocks for the preparation of nontoxic and biodegradable surface-active materials, either through functionalization of the carboxylic acid moiety or via N-acylation or N-alkylation of the amino group using acyl or alkyl halogenides (Fig. 4A) (29). The latter strategy involves the release of stoichiometric amounts of waste (Fig. 2A) and challenging purification procedures. To overcome these bottlenecks, we set to develop an alternative, highly modular catalytic strategy that provides convenient access to a broad variety of existing and novel surfactant structures solely derived from renewable resources. Such a fully sustainable approach directly couples α-amino acids with long-chain aliphatic alcohols, which can be produced from natural fats and oils (16) to obtain N-alkyl amino acids, preferably using an Earth-abundant metal catalyst. Recently, our group reported the direct and selective N-alkylation of simpler amines with alcohols using well-defined iron complexes (30). On the basis of the results presented here, we attempted the direct N-alkylation of glycine (1b) with 1-dodecanol (2n) using an Fe-based catalyst that is a structural analog of the Ru complex (Cat 1) used above. Gratifyingly, ideal reaction conditions were found at 110°C with 5 mol % Cat 2, and mono-N-dodecylglycine (3bn′) was isolated in 54% yield, besides a smaller amount 8% of N,N-didodecylglycine (3bn). The monoalkylation product (3bn′) has already been identified as a surfactant (10). After adding KOH and H2O to 3bn′, a rich foam formation was observed (Fig. 4B). Next, a variety of fatty alcohols including 1-nonanol (2j), 1-decanol (2o), 1-tetradecanol (2p), 1-hexadecanol (2q), and 1-octadecanol (2n) were reacted with 1b, and the corresponding mono-N-alkyl glycine derivatives were isolated (Fig. 4B). Furthermore, alanine (1c) and proline (1a) were successfully N-alkylated with 1-dodecanol (2n) and 1-nonanol (2j), respectively. These examples demonstrate a novel, straightforward and fully sustainable route to completely bio-based surfactants because both unprotected amino acids and alcohols can be derived from renewable resources.

Fig. 4 Synthesis of bio-derived surfactant.

(A) Conventional stoichiometric pathways for synthesizing amino acid–based surfactants. (B) Novel, catalytic method for the direct N-alkylation of amino acids with fatty alcohols with a well-defined iron catalyst, using alcohols of various lengths and cyclic and acyclic amino acids, to provide amino acid–based surfactants after the addition of aqueous KOH. See General procedure in Materials and Methods for the description of the experimental procedure and table S6 for further details on these experiments. General reaction conditions: 0.5 mmol of 1, 1 ml or 2 mmol of 2, neat or CF3CH2OH (2d) as solvent, 5 mol % Cat 2, 18 to 42 hours, and 110°C. Isolated yields are shown. aThe corresponding methyl esters were isolated. TMS, tetramethylsilane.

This study shows the direct N-alkylation of unprotected α-amino acids and simple peptides with a variety of alcohols using 1 mol % homogeneous Ru catalyst representing a high-yield and atom-economic transformation with excellent selectivities and only water as by-product. The method allows the use of a range of alcohols of various lengths as reaction partners, leading to functionalized amino acids with modular properties such as low or high hydrophilicity. Finally, the demonstration that long-chain alcohols and amino acids can be used as only reaction partners to provide directly mono-N-alkyl amino acid surfactants, applying a nonprecious iron catalyst with water as the only waste, represents a nearly ideal transformation, illustrating the great potential for future fully sustainable production of completely bio-based products.


General methods

Merck silica gel type 9385 (230 to 400) mesh or Merck Al2O3, 90 active neutral, was used for chromatography, whereas Merck silica gel 60 (0.25 mm) was used for thin-layer chromatography (TLC). Components were visualized by ultraviolet, ninhydrin, or I2 staining. Progress of the reactions was determined by 1H NMR spectroscopy. Mass spectra were recorded on an AEI MS-902 mass spectrometer [EI+ (positive electron ionization)] or an LTQ Orbitrap XL [ESI+ (positive electrospray ionization)]. 1H and 13C NMR spectra were recorded on a Varian AMX400 (400 and 100 MHz, respectively) using CDCl3, CD3OD, D2O, or dimethyl sulfoxide-d6 (DMSO-d6) as solvents. Chemical shift values are reported in parts per million with the solvent resonance as the internal standard (CDCl3, 7.26 for 1H and 77.00 for 13C; CD3OD, 3.31 for 1H and 49.00 for 13C; D2O, 4.79 for 1H; and DMSO-d6, 2.50 for 1H and 39.52 for 13C). Data are reported as follows: chemical shifts, multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; br., broad; m, multiplet), coupling constants (in hertz), and integration.

All reactions were carried out under an argon atmosphere using oven-dried (110°C) glassware and using standard Schlenk techniques. Toluene was collected from a MBRAUN solvent purification system (MB SPS-800). CF3CH2OH (>99.0%) was purchased from TCI without further purification. Complex Cat 2a (see table S5) was synthesized according to a previously reported procedure (31). Cat 1 was purchased from Strem. All other reagents were purchased from Sigma, TCI, or Acros in reagent or higher grade and were used without further purification.

Representative procedures

General procedure for the N-alkylation of amino acids with alcohols. An oven-dried 20-ml Schlenk tube equipped with stirring bar was charged with an amino acid (or peptide; given amount), Cat 1, Cat 2a + 2 equiv. of Me3NO or Cat 2 (given amount), and alcohol (given amount); solvent was added when indicated, otherwise the reactions were performed under neat conditions. Amino acid (or peptide) and catalyst were added into the Schlenk tube under air, the Schlenk tube was subsequently connected to an argon line, and a vacuum-backfill cycle was performed three times. Alcohol and solvent were charged under an argon stream. The Schlenk tube was sealed with a screw cap, and the mixture was rapidly stirred at room temperature for 1 min and then was placed into a preheated oil bath at the appropriate temperature and stirred for a given time. The reaction mixture was cooled down to room temperature and concentrated in vacuo. The crude product was characterized by 1H NMR to determine conversion. In several cases, no additional purification was required. As specified, where needed, further purification was conducted through flash column chromatography or crystallization to provide the pure N-alkyl amino acid (or peptide) product.

Esterification procedure (for the preparation of methyl ester of alky amino acids 3bp′, 3bq′, and 3br′). Continuing the General procedure, until the reaction mixture was cooled down to room temperature, 3 ml of benzene was added, and under stirring, TMSCHN2 (2 M in toluene) was added. The reaction was monitored by TLC [SiO2, mono-N-alkyl amino acid, retention factor (Rf) = 0.3 in ethylacetate/MeOH (1:1); methyl mono-N-alkyl amino acid ester, Rf = 0.3 in Et2O]. Finally, the corresponding methyl ester was purified by flash column chromatography [toluene/Et2O (50:50 to 0:100)].

Preparation of Cat 2. An oven-dried 250-ml Schlenk tube was charged with 100 ml of dry acetone and 2 ml of dry CH3CN, and the solution was degassed with N2 for 20 min. Then, 1 g of Cat 2a (2.38 mmol) was added under N2 and stirred for 1 min until it fully solubilized, after which 216 mg of Me3NO (1.2 equiv.) was added under N2. A direct color change from yellow to orange was observed within 5 s. The conversion of Cat 2a can be monitored by TLC [pentane/ethyl acetate (1:1), on silica gel, RfCat2a = 0.95, RfCat2 = 0.35). After 1 hour, the solvent was removed by vacuum; Cat 2 was purified by flash column chromatography and obtained as a brown solid (0.91 g, 88% yield). 1H NMR (400 MHz, CDCl3): δ 2.05 to 2.48 (m, 4H), 2.21 (s, 3H), 1.38 to 1.73 (m, 4H), and 0.22 (s, 18 H). 13C NMR (100 MHz, CDCl3): δ 212.80, 180.12, 126.00, 106.58, 69.91, 24.83, 22.31, 4.43, and −0.12. The physical data were identical in all respects to those previously reported (32). Cat 2a and Cat 2 are slightly light-sensitive but air-stable.


Supplementary material for this article is available at

table S1. Optimization of reaction conditions for the direct N-ethylation of unprotected amino acids (1a to 1i) with ethanol (2a).

table S2. Direct mono-N-isopropylation of an amino acid (1) with isopropanol (2b).

table S3. Direct N-alkylation of a free amino acid (1) with various alcohols (2).

table S4. Direct N-alkylation of di- and tripeptides (4a to 4b) with alcohols (2).

table S5. Iron-catalyzed N-ethylation of amino acids with ethanol (2a).

table S6. Iron-catalyzed N-alkylation of unprotected amino acids (1a to 1c) with long-chain aliphatic alcohols (2) to produce surfactants.

Spectral data of isolated compounds

Determination of the optical purity of products

HPLC traces

1H and 13C NMR spectra

References (3345)

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial license, which permits use, distribution, and reproduction in any medium, so long as the resultant use is not for commercial advantage and provided the original work is properly cited.


Acknowledgments: Funding: This study was funded by the European Research Council (ERC) [advanced grant 694345 to B.L.F. and ERC Starting Grant 2015 (CatASus) 638076 to K.B.] and the Dutch Ministry of Education, Culture and Science (Gravitation program 024.001.035) as well as the Netherlands Organisation for Scientific Research (NWO) [Vidi grant for K.B.]. Author contributions: T.Y. conceived the idea and performed the experiments. T.Y., B.L.F., and K.B. designed the experiments, analyzed the results, and wrote the manuscript. B.L.F. and K.B. guided the research. Competing interests: The authors of this work are all authors on a patent related to this work (European patent application no. 17163878.6). The authors declare that they have no other competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Additional data related to this paper may be requested from the authors.
View Abstract

Navigate This Article