Molecules: Formylation of Amines
Molecules: Formylation of Amines
Molecules: Formylation of Amines
3390/molecules19067689
OPEN ACCESS
molecules
ISSN 1420-3049
www.mdpi.com/journal/molecules
Review
Formylation of Amines
Ciera J. Gerack and Lisa McElwee-White *
Received: 12 May 2014; in revised form: 29 May 2014 / Accepted: 5 June 2014 /
Published: 10 June 2014
Abstract: Methods to convert amines to formamides are of interest due to the many uses
of formamides as synthetic intermediates. These methods include stoichiometric reactions
of formylating reagents and catalytic reactions with CO as the carbonyl source. This review
discusses the reported stoichiometric and catalytic approaches for preparation of formamides.
1. Introduction
Formamides are an important class of compounds that appear as intermediates in fungicide [1,2]
and pharmaceutical syntheses [1,3–5], and isocyanate [6], formamidine [7], and nitrile formation [8].
Formamides also serve as reagents in functional group conversion [9], the Vilsmeier formylation
reaction [10], and the allylation [11] and hydrosilation [12] of carbonyl compounds. Due to their wide
range of applications, many approaches have been developed to synthesize formamides. The methods
discussed in this review include the use of stoichiometric formylating agents, acid catalysts, organic
catalysts, transition metal catalysts, and catalytic carbonylation.
The earliest methods for formylation of amines involved formylating agents such as chloral, formic
acid, formaldehyde, and formates. In 1952, Blicke reported the formylation of amines with chloral (1)
(Scheme 1) [13]. This method produced excellent yields at low temperature, producing only
Molecules 2014, 19 7690
chloroform as a byproduct. Successful substrates for this method include strongly basic primary
amines, diamines, cyclic secondary amines, and sterically hindered secondary amines.
Formic acid itself can be used to achieve formylation by dehydration (Scheme 2) [14]. The amine
and formic acid were dissolved in toluene and the solution was refluxed in the presence of a Dean-
Stark trap, which collected the water produced by the condensation reaction.
An example of N-formylation by formic acid under solvent-free conditions was reported by Hajra [15].
The amine and formic acid were heated to 80 °C until the reaction reached completion. Formamide
products were obtained in good to excellent yields from substituted aromatic amines as well as primary
and secondary alkyl amines. The yields produced by aliphatic amines yields were lower than those
from aromatic amines. When a mixture of primary and secondary amines was exposed to the reaction
conditions, the primary amines were formylated selectively. Hydroxyl substituents remained intact
after formylation of the amine and no isolable side products were observed.
Formic acid in polyethylene glycol has been shown to formylate anilines (Scheme 3) [16]. This
reaction can be carried out at room temperature under an inert atmosphere in relatively short reaction
times of 4–6 h. The conditions are tolerant of functional groups such as nitro, halogen, ester, ketone,
and alkyl groups. Attempts to formylate the oxygen of phenols with these conditions were
unsuccessful, presumably due to the lower nucleophilicity of the phenol.
Scheme 3. Formylation of aromatic amines using formic acid and polyethylene glycol.
There are many examples of N-formylation by acetic formic anhydride (AFA) [17–23]. One such
example is a one-pot procedure for N-monomethylation of primary amines that proceeds through
N-formylation followed by reduction [24]. Amines were allowed to react with AFA, which was
generated in situ from excess formic acid and acetic anhydride at −20 °C (Scheme 4). The reaction
Molecules 2014, 19 7691
reached completion for most amines in less than 15 min and the resulting formamides were isolated in
yields of 97%–100%. High yields were achieved for formylation of simple alkyl, aromatic,
multifunctional, and sterically hindered amines, such as 2.
Formylation is often used as a means of protecting amino groups in peptide synthesis. As early as
1932, du Vigneaud had reported a procedure for N-formylating amino acids with formic acid and
acetic anhydride [25]. This method was used as a way to protect the amines during resolution of
D,L-cystine by selective crystallization of the strychnine salt. This method of formylating amino acids
was applied by Yang as a means of protecting the amino group of many additional amino acids [26].
When the amino acids were exposed to the reaction conditions, the procedure yielded the N-formyl
aminoacids in yields of 78%–90%.
Problems with racemization of tert-butyl esters of amino acids during attempted formylation with
by AFA-were addressed using a modification that allows the formylation of tert-butyl amino acid
esters with minimal or no racemization [27]. This method combined formic acid with dicyclohexyl-
carbodiimide (DCC) to form the active formylating reagent, which was added to solutions of tert-butyl
amino acid esters. The protected amino acid esters were produced in high yields. A related method of
formylating amino acid esters was reported by Benoiton [28]. In this method N-ethyl-N′-(3-
dimethylaminopropyl)-carbodiimide is used to prepare the formic anhydride which is then allowed to
react with the ester salt in the presence of N-methylmorpholine. This method does not require any
specialized purification and produced high yields from methyl, benzyl and tert-butyl esters.
Formylation of amino acid esters can also be carried out employing cyanomethylformate as a
formylating agent [29]. This method was successfully applied to methyl, ethyl, benzyl, and tert-butyl
amino acid esters and afforded good to excellent yields without loss of optical purity.
Ammonium formate has been shown to formylate both anilines and secondary amines in good to
excellent yields upon reflux in acetonitrile (Scheme 5) [30]. With the exception of benzylamine,
primary amines produced alkyl formate salts instead of the expected formamide products.
These conditions are also applicable as protecting groups for chiral molecules. The benzyl ester of
L-proline (3) was successfully formylated to 4 in 75% yield without any observed racemization
(Scheme 6). When hydroxyl groups were present, ammonium formate selectively formylated the
nitrogen, leaving the hydroxyl group intact.
A recent method used the Reimer-Tiemann (R-T) reaction to produce formamides from secondary
amines (Scheme 7) [31]. Alkyl, cyclic, and N-methylaniline derivatives all produced formamides in
good to excellent yields, but the best yields were obtained with cyclic amines. A mechanistic pathway
consistent with the R-T reaction was proposed (Scheme 8). First, chloroform reacted with sodium
ethoxide to form the trichloromethyl carbanion 5, which readily loses chloride to generate
dichlorocarbene 6. Then carbene 6 reacted with the amine and produced the formamide product
through the R-T reaction.
Roberts initially reported that amines and triethyl orthoformate produced the corresponding N-ethyl
formamides in the presence of sulfuric acid at high temperature [19]. Swaringen performed a similar
transformation using p-toluenesulfonic acid and triethyl orthoformate [33]. Using a similar method,
Kaboudin reported simple N-formylation without an alkyl shift onto the nitrogen (Scheme 11) [34].
In this method, formylation of primary amines took place in water with triethyl orthoformate in the
absence of base, acid, or catalyst in moderate to good yields. No results from secondary amines were
reported. Water proved to be the optimal solvent after experiments with ethanol, ethyl acetate,
dichloromethane (DCM), chloroform, dimethylformamide (DMF), and dimethylsulfoxide (DMSO).
Product could obtained either by reflux in water for 24–48 h (method A) or by microwave irradiation
at 90 °C for 2–3 h (method B).
Formic acid and a catalytic amount of sodium formate have been reported to produce formamides
from amines at room temperature under solvent-free conditions [35]. Functionalized anilines, primary
amines, cyclic secondary amines, and sterically hindered secondary amines all produced good to
excellent yields of formamides in less than 8 h. The reaction was selective for N-formylation over
O-formylation in the presence of unprotected hydroxyl groups on the amine substrates. The sodium
formate used in the reaction could be isolated from the reaction mixtures and reused up to four times
without a loss of activity.
N-formylation of anilines and simple primary amines by formic acid can also be carried out in
solvent-free conditions with the reusable ion exchange resin Amberlite IR-120[H+] as an acid catalyst [36].
Mixtures of Amberlite IR-120, the amine and formic acid were exposed to microwave irradiation for
20 s intervals until all starting material was consumed. The reactions were complete between 60–120 s
regardless of amine or substituents and the formylated amines were obtained in excellent yields. At the
end of each reaction, the resin was easily isolated and was reusable up to five times without a loss in
activity. The proposed mechanistic pathway involved coordination to the resin through hydrogen
bonds (Scheme 12), followed by attack at the carbonyl by the amine. Subsequent rearrangement
produced the formamide, leaving water coordinated to the resin.
Scheme 12. Amberlite IR-120 catalyzed formylation. Scheme adapted from reference [36].
Jang reported N-formylation in solvent-free conditions with molecular iodine (I2) as a catalyst [37].
During optimization studies, it was determined that aniline, 5 mol% I2, and two equivalents of formic
acid produced formanilide in excellent yield after 2 h at 70 °C. After optimization was completed,
several aniline derivatives as well as primary and secondary amines were subjected to the reaction
conditions and produced formamides in good to excellent yields. As it is known that I2 reacts with
formic acid to produce HI [38], it was assumed that HI was the active catalytic species, generated
in situ. In the proposed mechanism, protonated formic acid is attacked by the amine, followed by
proton transfer to provide 8, and finally through elimination of water and a proton, the formamide is
formed (Scheme 13).
Molecules 2014, 19 7695
Hu reported the use of thiamine hydrochloride 9 as a catalyst to produce formamides from the
amine and formic acid in solvent-free conditions (Scheme 14) [39]. This method was successfully
applied to aromatic and aliphatic amines, with yields ranging from 88%–96%. When other carboxylic
acids were used in place of formic acid, the corresponding amides were produced. While the
mechanistic pathway is not known, it was suggested that the catalyst activates formic acid through
hydrogen bonding. After nucleophilic attack of the amine at the carbonyl of formic acid, the
formamides were produced through the elimination of water (Scheme 15).
Other solvent-free, acid catalyzed methods have been reported. Hajela reported silica supported
perchloric acid (HClO4-SiO2) catalyzed N-formylation of aromatic and cyclic secondary amines
Molecules 2014, 19 7696
(Scheme 16) [40]. When substrates with hydroxyl groups were exposed to this reaction, formylation
occurred selectively at the amino position. Other silica supported acids including sulfuric (H2SO4),
fluoroboric (HBF4), and trifluoroacetic (TFA) acids were examined but all produced lower yields of
products. The catalyst was easily removed at the completion of the reaction and after washing and
drying, could be used up to three times without a loss in activity.
Scheme 17. Formylation of amines by formic acid with a supported acid catalyst.
Another transition metal catalyst that acts as an acid was reported by Akamanchi [42]. Sulfated
tungstate 10 catalyzed the reaction of amines with formic acid to produce formamides in solvent-free
conditions (Scheme 18).
Scheme 18. N-formylation of amines with formic acid and sulfated tungstate catalyst.
Molecules 2014, 19 7697
The optimized conditions are 10 mol% 10, 70 °C and 1.2 equivalents of formic acid for 10–45 min.
The catalyst was easily isolated after the reaction and could be reused up to four times without
experiencing any loss of activity. Yields ranged from 85%–99% for formylation of primary,
secondary, aromatic, heteroaromatic, and alkyl amines as well as α-amino acids. Studies of the
interaction between the catalyst and the reagents indicated that formic acid was adsorbed onto the
catalyst, but amine was not adsorbed. This suggests a mechanism in which the catalyst activates the
formic acid followed by nucleophilic attack, similar to other acid catalyzed formylations.
Formylation with formic acid and 2-chloro-4,6-dimethoxy[1.3.5]triazine (CDMT, 11) was reported
by Giacomelli [43]. Amines and amino acid esters were formylated in nearly quantitative yields either
in DCM at reflux (method A) or under microwave irradiation (method B) (Scheme 19) to yield the
formamide products in a one pot process. In method A, dry formic acid and the amine were treated
with 11 and 4-(dimethylamino)pyridine (DMAP) as a catalyst. N-methylmorpholine (NMM) and DCM
were added, and the solution was refluxed. This reaction required 5–20 h to reach completion.
However, with the use of microwave irradiation in method B, the reaction produced the formamides
after only 3–6 min. When chiral amino acid esters were used, optical purity was maintained but steric
hindrance in the amine resulted in slightly lowered yields. The proposed mechanistic pathway involved
the formation of a formate ester intermediate composed of formic acid and 11, which then is attacked
by the amine to form the formamide (Scheme 20).
Scheme 19. One step preparation of formamides by method A at reflux or method B via
microwave irradiation.
Scheme 20. Pathway of formamide formation from formic acid and 11.
Molecules 2014, 19 7698
Deutsch reported formylation of amines by methyl formate and catalytic base (Scheme 21) [44].
Amidine and guanidine catalysts were examined in the formylation of morpholine and
tert-butylamine at room temperature. The best catalyst for this reaction was 1,5,7-triazabicyclo[4.4.0]dec-
5-ene (TBD, 12).
Ionic liquids (IL) have recently been reported to catalyze formylation. ILs are attractive because of
their stability, ease of removal, and easy synthesis. Baghbanian reported that amine, formic acid, and
TBD based ionic liquids produced formamides from aromatic, alkyl, and heteroaromatic amines as
well as amino alcohols in good to excellent yields (Figure 1) [45]. Three related ionic liquids were
examined (13, 14, and 15) with 12 being preferred (Scheme 22). Addition of solvent to the reaction
mixtures resulted in decreased yields. The IL was easily separated from the product and reused up to
six times without a loss in activity. A proposed mechanism indicated that the reaction was mediated
through hydrogen bonding with the catalyst (Scheme 23).
Figure 1. TBD based ILs examined for catalytic activity towards N-formylation.
Metals have also been used as catalysts for formylation of amines using stoichiometric amounts of
formic acid or another formylating agent. Jang reported solvent-free conditions in which amine, formic
acid, and 10 mol% indium at 70 °C produced formamide in moderate to excellent yields (Scheme 24) [46].
Without indium, the yields were considerably lower. Aniline derivatives, primary amines, secondary
amines, and amino alcohols were all successfully formylated under these conditions, with the reaction
time varying between 1–24 h depending on the electronics and sterics of the amine. Amino groups of
methyl and benzyl α–amino acid esters could be protected in good yields under these conditions with
no racemization.
A variety of transition metal Lewis acids have also been used in N-formylation of amines. ZnO was
reported by Hosseini-Sarvari as a Lewis acid catalyst for the solvent-free formylation of aromatic,
primary and secondary amines with formic acid in good to excellent yield [47]. Optimal conditions for
this reaction are 3 equivalents of formic acid, 50 mol% catalyst and 70 °C for 10–720 min (Scheme 25).
The reaction could be scaled up from 1 mmol amine to 100 mmol amine without any loss of yield. The
ZnO catalyst was filtered out of the reaction mixture at the completion of the reaction and washed with
DCM, after which it could be successfully recycled up to three times. Longer reaction times were
Molecules 2014, 19 7700
necessary for aromatic amines containing electron withdrawing groups as well as for secondary
amines. When a mixture of primary and secondary amines was subjected to the reaction conditions,
primary amines were preferentially formylated. Amines containing hydroxyl groups were selectively
formylated at the amino group and competitive O-formylation was not observed.
Similar Lewis acid (LA)-catalyzed, solvent-free conditions for formylation were reported by Rao [48].
A Lewis acid catalyst and formic acid were used to produce high yields of the desired formamide
products (Scheme 26). Although Lewis acids such as FeCl3, AlCl3, and NiCl2 could be used, the
inexpensive, environmentally friendly catalyst ZnCl2 produced the best results. The optimum
conditions are 10 mol% catalyst, 3 equivalents of formic acid and 70 °C for 10–900 min. The reaction
required longer reaction times for electron poor aromatic amines and secondary amines but tolerated a
variety of functional groups such as nitro, halogen, ester, ketone, and alkyl. The proposed reaction
mechanism is similar to other acid catalyzed reactions of formic acid and amines.
Formylation of amines with formic acid and the nanoparticle photocatalyst TiO2-P25 or sulfated
titania was reported by Swaminathan as an extension of research on semiconductor photocatalysts [49].
In this work, either TiO2-P25 or TiO2-SO42− catalyzed formylation of amines with formic acid in short
reaction times at room temperature (Scheme 27). This method was applied to substituted aromatic
amines as well as primary and secondary aliphatic amines. In all cases studied, TiO2-SO42− produced
better yields, ranging from moderate to excellent. In recyclability tests, TiO2-SO42− could be reused up
to five times without a loss in activity, while TiO2-P25 suffered a 50% drop in activity during the
second trial. A proposed mechanism invokes acid catalyzed formylation (Scheme 28). Lewis acidic
sites on the catalyst coordinate formic acid and facilitate nucleophilic attack of the amine on the
carbonyl. The formamide product is then produced by loss of water.
Scheme 27. Amine formylation from formic acid catalyzed by TiO2-P25 or TiO2-SO42−.
Molecules 2014, 19 7701
More recently, Hong reported that the fluorous silica gel-supported hafnium
(IV)bis(perfluorooctanesulfonyl)imide complex (FSG-Hf[N(SO2C8F17)2]4)-catalyzed formylation of
amines in aqueous formic acid (Scheme 29) [50]. Optimum conditions were 1 mol% catalyst, 70 °C,
and 3 equivalents of formic acid. The catalyst could be reused for up to three cycles without loss of
activity. Aromatic amines produced the desired formamides in high yields regardless of substituent.
However, when electron withdrawing groups were present, longer reaction times were necessary.
Aliphatic N-butylamine and secondary diphenylamine produced good yields. It was proposed that the
high loading of formic acid on the catalyst resulted in hydrogen bonding that rendered the carbonyl
more electrophilic and facilitated attack by the amine (Scheme 30).
Scheme 30. Electronic assistance for nucleophilic attack of amine on formic acid.
Molecules 2014, 19 7702
Formylation of dimethylamine with formaldehyde on silver and gold surfaces was studied
collaboratively by Madix and Friend [52,53]. Oxygen assisted formylation of dimethylamine on
metallic silver surfaces was reported by Madix [52]. The proposed mechanistic pathway involves
dissociative adsorption of O2 on the silver surface, along with coordination and deprotonation of the
amine. When formaldehyde is introduced, it inserts into the surface-amide bond and β-hydride
elimination occurs to produce the formamide (Scheme 32). A similar mechanism was proposed by
Friend for gold surfaces (Scheme 32) [53]. One important distinction between these two mechanisms is
that gold requires ozone (O3) to introduce adsorbed oxygen, while silver can employ oxygen (O2).
These are examples in which the catalyst interacts with the amine directly during reaction with a
formylating agent.
Gold nanoparticles were reported by Ishida to formylate amines with methanol [54]. Formamide
was produced as the primary product when Au/Al2O3 or Au/NiO was used. Gold nanoparticles were
later reported by Sakurai to catalyze the formylation of amines with methanol or formaldehyde [55].
Gold nanoclusters stabilized by poly(N-vinyl-2-pyrrolidione) (Au:PVP) acted as the catalyst under
aerobic oxidation conditions. Optimum conditions for the reaction with methanol were 10 atom%
catalyst, 200 mol% LiOH as base and 1:2 methanol:water solvent at 80 °C (reflux) for 8 h (Scheme 33).
When these conditions were applied to N-methylaniline, two products were formed: 94% yield of
N-methylanilide 16 and 5% yield of anilide 17. Because methanol oxidation leads to formaldehyde,
formic acid, methyl formate, and carbon dioxide, control experiments with these possible formylating
agents were used to ascertain which intermediate was reacting with the amine. Without methanol or a
formylating agent, no reaction occurred. When formaldehyde was used in place of methanol as a 37%
solution, 16 was formed in 81% yield. When either methyl formate or formic acid was used in place of
methanol, no reaction occurred. Optimum conditions for this reaction when formaldehyde was used as
the formyl source were 1.5 equivalents of formaldehyde, 1 atom% catalyst, 100 mol% NaOH as base
and 1:2 ethanol‒water solvent at 27 °C for 9 h. Under these new conditions, the yields were best for
electron-rich aromatic amines such as aniline, 4-methylaniline, N,4-dimethylaniline, and indoline.
Sterically hindered and electron poor aromatic species produced little to no product. Primary and
secondary alkyl amines also produced high yields of formamides.
Scheme 34. Ru-NHC catalyzed methanol activation and formylation of amines. (A)
Catalyst 18. (B) Optimum conditions.
Molecules 2014, 19 7704
When these conditions (1 mol% 18, 1.5 equivalents of alcohol, refluxing toluene, inert atmosphere,
24 h) produced only trace amount of the formamide, the amount of methanol was increased to 3.3
equivalents. Attempted optimizations of reactant stoichiometry, concentration, solvent, and
temperature did not increase the yield. When the reaction was run in a sealed container, the yield was
lower and a build-up of hydrogen gas was observed. The introduction of styrene as a sacrificial
hydrogen acceptor increased the conversion of starting material to formamide to 96%.
The scope was examined using primary, secondary, tertiary, and benzyl amines. Overall yields
ranged from 27%–99%, with the lowest yields obtained from bulkier substrates and electron poor
benzyl amines. Aromatic amines did not react. Optically pure phenylethylamine produced a 77% yield
of the formamide with no loss of enantiomeric purity. During the examination of the reaction scope the
catalyst 18 was formed in situ from the pre-catalyst Ru(cod)(2-methylallyl)2, the HCl salt of the NHC
ligand, and base (Scheme 34). Control experiments demonstrated that the reaction was not base
catalyzed. The reaction did not occur in the absence of the NHC and use of NHCs other than ICy
resulted in lower conversion of starting material. Through examination of the reaction with NMR
studies, a mechanism was proposed (Scheme 35). Coordination of methoxide to Ru is followed by
β-hydride elimination to produce coordinated formaldehyde. The formaldehyde undergoes
nucleophilic attack by the amine, then hydrogen (H2) is lost as the amine is deprotonated. A second
β-hydride elimination occurs to form the formamide product, which is displaced by methanol. A
second H2 is liberated as the original methoxide complex is formed, closing the catalytic cycle.
Scheme 35. Mechanism for Ru-catalyzed formylation using methanol and 18.
Reddy reported formylation of primary and secondary amines by the catalytic oxidation of
methanol by copper salts with hydrogen peroxide as the terminal oxidant (Scheme 36) [57]. Optimized
conditions for the formylation of amines were amine, 30 mol% CuCl2•H2O and 3.4 equivalents of
Molecules 2014, 19 7705
6.0% w/w H2O2 in room temperature methanol for 45–90 min. Primary and secondary amines were
formylated in 63%–80% yields. Slow addition of H2O2 to the reaction mixture was important for rapid
formation of formylated product. When the same amount of H2O2 was added as two equal portions,
decomposition of the peroxide resulted in longer reaction times and a need for additional H2O2.
Scheme 36. Oxidative N-formylation of amines with formaldehyde in the presence of copper salts.
An example of IL-catalyzed formylation using CO as the carbonyl source was reported by Lee [58].
Mixtures of the amine and IL under 40 atm CO produced formamides from primary and secondary
amines in moderate to excellent yields. No urea products were observed in the reaction mixtures.
Selections of IL and counterions were examined, with 1-butyl-3-methylimidazolium carbonate (19)
exhibiting the best performance for N-formylation. Other optimized conditions include 40 atm CO,
methanol solvent, 140 °C, and 1 mol% 19 (Scheme 37). The catalyst could be used for five trials with
no loss in selectivity and only a 20% reduction in activity.
Transition metal catalysts have also been used to formylate amines using CO as the carbonyl
source [59–67]. However, when transition metal catalysts are used, catalytic carbonylation more
commonly yields ureas instead of formamides [68–71]. Formylation of amines by metal complexes
and carbon monoxide was reported by Saegusa [72]. This reaction selectively formed formamides with
only trace amounts of urea being observed. Various catalysts were examined, with the highest activity
being obtained from CuCl. Acceleration of this reaction in water was attributed to favorable formation
of the CuCl-CO complex in water. Secondary aliphatic amines proved to be better substrates for this
reaction than primary aliphatic amines. Aromatic amines did not yield formamides with CuCl catalyst
but they could be formylated when chloroauric acid (HAuCl4•H2O) was used as catalyst.
Remple reported that the ruthenium catalyst [Ru(CO)2(OCOMe)]n. produced formamides from
cyclic secondary amines using only 1 atm of carbon monoxide gas at 75 °C (Scheme 38) [61]. The
mild conditions in this reaction required long reaction times (20–200 h) to ensure completion. Neither
primary amines nor acyclic secondary amines could be formylated under these conditions.
Molecules 2014, 19 7706
Triruthenium dodecacarbonyl (Ru3(CO)12) has also been found to catalyze formylation of amines
using CO as the carbonyl source [73]. Different ruthenium and rhodium catalysts were examined in
this work, with the rhodium complexes producing significant amounts of urea. The best catalytic
activity toward formamide production was shown by Ru3(CO)12. Optimum conditions were 0.17 mol%
catalyst and 40 atm CO in benzene at 120–180 °C for 6 h. Primary aliphatic amines were successfully
converted to the formamide products under these conditions. However, carbonylation of piperidine
produced the formamide in only moderate yield due to competitive formation of urea.
Jenner reported that ruthenium compounds catalyzed primary and secondary amines with CO as the
carbonyl source [59]. Ruthenium trichloride trihydrate (RuCl3•3H2O) showed the highest activity and
selectivity towards the formation of formamides from primary amines. The cobalt catalyst
Co(OAc)2•4H2O showed low conversion, and the rhodium catalyst RhCl3•3H2O showed high activity
but low selectivity towards the formamide product. Using the ruthenium catalyst, primary amines were
carbonylated, with the exception of the sterically crowded tert-butylamine. Aniline also failed to
produce formamide product. Cyclic secondary amines were carbonylated to the corresponding formamides.
When acyclic secondary amines were examined, a competing pathway occurred in which
transalkylation formed a tertiary amine and a primary amine from two equivalents of secondary amine.
The newly formed primary amine was then carbonylated to the formamide product. Increased
temperature led to lowered transalkylation but higher production of ureas. Increased pressure
suppressed the reaction but did favor formamide over transalkylation products. Cobalt-ruthenium
co-catalysts improved the selectivity for formylation of dialkylformamides from acyclic secondary
amines over the transalkylation pathway. Using RuCl3•3H2O and Co(OAc)2•4H2O, the optimum ratio
of Ru:Co for formamide production was 1.3:1.
In related work, Jenner reported the effect of solvent on carbonylation of amines with RuCl3•3H2O [60].
No correlation between the yield of formamide and dielectric constant was found. Methanol was the
best solvent for ruthenium catalyzed carbonylation. Results in water were poorer, with greater yields of
urea or lower conversion of amine observed. The previously examined cobalt-ruthenium catalyst,
RuCl3•3H2O and Co(OAc)2•4H2O used at a ratio of 1.3:1, produced selectivity for formamide in
methanol similar to that of the ruthenium catalyst. Raising the pressure to 750 atm and raising the
temperature to 180 °C increased formamide selectivity with respect to urea. The best conversion and
selectivity were observed when increased methanol was present. The turnover value was the highest
with equal volumes of amine and methanol.
The proposed reaction pathway for this reaction was initial formation of methyl formate from
methanol and CO, followed by attack from the dialkylamine forming the formamide and regenerating
methanol. This reaction was successfully applied to dialkylamines and aromatic amines. However,
aromatic amines still show lower conversion and selectivity than alkylamines. Sterically hindered
Molecules 2014, 19 7707
amines such as tert-butylamine, which was unreactive without methanol, underwent selective
formylation in the presence of ruthenium catalyst and methanol solvent.
McElwee-White reported tungsten complex 20 that produced formamides and ureas from secondary
and primary amines respectively (Scheme 39) [63]. Using this method, secondary amines selectively
form formamides in yields ranging from 8%–61% and primary amines selectively form ureas in yields
ranging from 56%–105% (yields calculated per equivalent of tungsten).
R
H N O
R R
CH2Cl2 N H
Ph
I [ox] R
N
OC I CO
W W
OC I CO
I N H
Ph N R O
20 H H H
N N
CH2Cl2
[ox] R R
McElwee-White later reported that W(CO)6 with an oxidant would carbonylate amines with CO to
form ureas [74–82] and hydantoins [83]. During control experiments, it was discovered that a
NaIO4/NaI oxidant/promoter system with CO in CH2Cl2 would carbonylate amines without the metal
catalyst [84]. Reaction of 4-methoxybenzylamine afforded both urea and formamide products (Scheme 40).
The NaIO4 mediated reaction has been optimized for formation of the formamide in methanol
solvent [85]. Isotopic labeling experiments using 13C-labeled methanol showed that CO is the carbonyl
source, not oxidized methanol. Deuterium labeling experiments revealed that the formyl hydrogen
came from the acidic proton of methanol. During further investigation into this reaction, it was
determined that the oxidant was unnecessary for the formylation, although it is necessary in the urea
synthesis [86]. The necessary components of this reaction are amine, CO, and base (Scheme 41). The
functional group tolerance of this reaction is broad. Substrates include primary, benzyl, cyclic
secondary and acyclic secondary amines.
Molecules 2014, 19 7708
7. Conclusions
Methods of formylating amines include stoichiometric reagents such as chloral, formic acid, acetic
formic anhydride and ammonium formate, as well as catalytic processes. Both acid catalysts and
organic catalysts can be used with formic acid and formates to produce formamides. Metal catalysts
can formylate amines using formic acid, paraformaldehyde, formaldehyde, and methanol as the source
of the formyl moiety. Catalytic carbonylation routes produce formamides from CO in the presence of
ionic liquids, transition metals, or oxidants. Base-mediated carbonylation with CO is also reported.
These methods can be applied to a large scope of amines such as primary, cyclic and acyclic
secondary, sterically hindered, aromatic, amino acids, and amino acid esters. Methods are available for
formylation of amines with preservation of enantiomeric purity. Although there is a long history of
formylation chemistry, room remains for development of new catalytic methods.
Acknowledgments
We thank the donors of the American Chemical Society Petroleum Research Fund for support of
our carbonylation work through the Green Chemistry Institute. We are also grateful to our excellent
coworkers whose names appear in the references.
Author Contributions
Ciera J. Gerack wrote the first draft under the supervision of Lisa McElwee-White. Both authors
edited and revised the manuscript.
Conflicts of Interest
References
1. Chen, B.C.; Bednarz, M.S.; Zhao, R.; Sundeen, J.E.; Chen, P.; Shen, Z.; Skoumbourdis, A.P.;
Barrish, J.C. A new facile method for the synthesis of 1-arylimidazole-5-carboxylates.
Tetrahedron Lett. 2000, 41, 5453–5456.
2. Grant, H.G.; Summers, L.A. Synthesis of N-methyl-N-(2,2,2-trichloro-1-arylaminoethyl)tormamides
and related-compounds as potential fungicides. Aust. J. Chem. 1980, 33, 613–617.
3. Kobayashi, K.; Nagato, S.; Kawakita, M.; Morikawa, O.; Konishi, H. Synthesis of 1-formyl-1,2-
dihydroquinoline derivatives by a lewis acid-catalyzed cyclization of o-(1-hydroxy-2-
alkenyl)phenyl isocyanides. Chem. Lett. 1995, 24, 575–576.
Molecules 2014, 19 7709
4. Jackson, A.; Meth-Cohn, O. A new short and efficient strategy for the synthesis of quinolone
antibiotics. J. Chem. Soc. Chem. Commun. 1995, 1319–1319.
5. Pettit, G.; Kalnins, M.; Liu, T.; Thomas, E.; Parent, K. Notes- potential cancerocidal agents. III.
Formanilides. J. Org. Chem. 1961, 26, 2563–2566.
6. Faraj, M.K. Synthesis of Isocyanate Precursors from Primary Formamides. U.S. Patent 5,686,645,
1997.
7. Han, Y.; Cai, L. An efficient and convenient synthesis of formamidines. Tetrahedron Lett. 1997,
38, 5423–5426.
8. Arlt, D.; Klein, G. Preparation of Nitriles from Formamides. U.S. Patent 4,419,297, 1983.
9. Ding, S.; Jiao, N. N,N-dimethylformamide: A multipurpose building block. Angew. Chem. Int. Ed.
2012, 51, 9226–9237.
10. Downie, I.M.; Earle, M.J.; Heaney, H.; Shuhaibar, K.F. Vilsmeier formylation and glyoxylation
reactions of nucleophilic aromatic compounds using pyrophosphoryl chloride. Tetrahedron 1993,
49, 4015–4034.
11. Kobayashi, S.; Nishio, K. Facile and highly stereoselective synthesis of homoallylic alcohols
using organosilicon intermediates. J. Org. Chem. 1994, 59, 6620–6628.
12. Kobayashi, S.; Yasuda, M.; Hachiya, I. Trichlorosilane-dimethylformamide (Cl3SiH-DMF) as an
efficient reducing agent. Reduction of aldehydes and imines and reductive amination of aldehydes
under mild conditions using hypervalent hydridosilicates. Chem. Lett. 1996, 25, 407–408.
13. Blicke, F.F.; Lu, C.-J. Formylation of amines with chloral and reduction of the N-formyl
derivatives with lithium aluminum hydride. J. Am. Chem. Soc. 1952, 74, 3933–3934.
14. Jung, S.H.; Ahn, J.H.; Park, S.K.; Choi, J.-K. A practical and convenient procedure for the
N-formylation of amines using formic acid. Bull. Korean Chem. Soc. 2002, 23, 149–150.
15. Rahman, M.; Kundu, D.; Hajra, A.; Majee, A. Formylation without catalyst and solvent at
80 °C. Tetrahedron Lett. 2010, 51, 2896–2899.
16. Das, B.; Krishnaiah, M.; Balasubramanyam, P.; Veeranjaneyulu, B.; Nandan Kumar, D.
A remarkably simple N-formylation of anilines using polyethylene glycol. Tetrahedron Lett.
2008, 49, 2225–2227.
17. Strazzolini, P.; Giumanini, A.G.; Cauci, S. Acetic formic anhydride a review. Tetrahedron 1990,
46, 1081–1118.
18. Barton, D.H.R.; Bringmann, G.; Lamotte, G.; Motherwell, W.B.; Motherwell, R.S.H.;
Porter, A.E.A. Reactions of relevance to the chemistry of aminoglycoside antibiotics. Part 14. A
useful radical-deamination reaction. J. Chem. Soc. Perkin 1 1980, 2657–2664.
19. Evans, D.; Eastwood, F. Synthesis of an arylhydroxytetronimide and of 3-hydroxy-4(1H)-
quinolone derivatives. Aust. J. Chem. 1974, 27, 537–542.
20. Terashima, S.; Takashima, K.; Sato, T.; Yamada, S.I. Stereochemical studies. XXII. Thermal
rearrangement of s(−)-1-phenylethyl isocyanide. Chem. Pharm. Bull. 1973, 21, 1135–1139.
21. Johannsen, F.; Jorgensen, A.; Pedersen, E.B. Reactions of heterocyclic o-aminonitriles with acetic
formic anhydride. Chem. Scr. 1986, 26, 347–351.
22. Van Dort, M.; Neubig, R.; Counsell, R.E. Radioiodinated p-iodoclonidine. A high-affinity probe
for the α2-adrenergic receptor. J. Med. Chem. 1987, 30, 1241–1244.
Molecules 2014, 19 7710
23. Nolte, R.J.M.; van Zomeren, J.A.J.; Zwikker, J.W. Poly(iminomethylenes). 6. Synthesis and
polymerization of α- and β-D-glucopyranosyl isocyanide. J. Org. Chem. 1978, 43, 1972–1975.
24. Krishnamurthy, S. A highly efficient and general N-monomethylation of functionalized primary
amines via formylation--borane:methyl sulfide reduction. Tetrahedron Lett. 1982, 23, 3315–3318.
25. Du Vigneaud, V.; Dorfmann, R.; Loring, H.S. A comparison of the growth-promoting properties
of D- and L-cystine. J. Biol. Chem. 1932, 98, 577–589.
26. Sheehan, J.C.; Yang, D.D.H. The use of N-formylamino acids in peptide synthesis. J. Am. Chem. Soc.
1958, 80, 1154–1158.
27. Waki, M.; Meienhofer, J. Efficient preparation of nα-formylamino acid tert-butyl esters. J. Org.
Chem. 1977, 42, 2019–2020.
28. Chen, F.M.F.; Benoiton, N.L. A general method for formylating sensitive amino acid esters.
Synthesis 1979, 1979, 709–710.
29. Duczek, W.; Deutsch, J.; Vieth, S.; Niclas, H.J. A simple and convenient synthesis of N-formyl
amino acid esters under mild conditions. Synthesis 1996, 1996, 37–38.
30. Ganapati Reddy, P.; Kishore Kumar, G.D.; Baskaran, S. A convenient method for the
N-formylation of secondary amines and anilines using ammonium formate. Tetrahedron Lett.
2000, 41, 9149–9151.
31. Shastri, L.A.; Shastri, S.L.; Bathula, C.D.; Basanagouda, M.; Kulkarni, M.V. Mild, simple, and
efficient method for N-formylation of secondary amines via Reimer-Tiemann reaction. Synth.
Commun. 2009, 41, 476–484.
32. Yang, X.; Zhang, Y. Melamine trisulfonic acid-catalyzed N-formylation of amines under
solvent-free conditions. Res. Chem. Intermed. 2012, 1–6.
33. Swaringen, R.A.; Eaddy, J.F.; Henderson, T.R. Reaction of ortho esters with secondary amines.
J. Org. Chem. 1980, 45, 3986–3989.
34. Kaboudin, B.; Khodamorady, M. Organic reactions in water: A practical and convenient method
for the N-formylation of amines in water. Synlett 2010, 2010, 2905–2907.
35. Brahmachari, G.; Laskar, S. A very simple and highly efficient procedure for N-formylation of
primary and secondary amines at room temperature under solvent-free conditions. Tetrahedron Lett.
2010, 51, 2319–2322.
36. Muthukur Bhojegowd, M.R.; Nizam, A.; Pasha, M.A. Amberlite IR-120: A reusable catalyst for
N-formylation of amines with formic acid using microwaves. Chin. J. Catal. 2010, 31, 518–520.
37. Kim, J.G.; Jang, D.O. Facile and highly efficient N-formylation of amines using a catalytic
amount of iodine under solvent-free conditions. Synlett 2010, 2010, 2093–2096.
38. Hammick, D.L.; Zvegintzov, M. CXLIV.-The rate of reaction between formic acid and iodine in
aqueous solution. J. Chem. Soc. 1926, 129, 1105–1108.
39. Lei, M.; Ma, L.; Hu, L. A convenient one-pot synthesis of formamide derivatives using thiamine
hydrochloride as a novel catalyst. Tetrahedron Lett. 2010, 51, 4186–4188.
40. Ansari, M.I.; Hussain, M.K.; Yadav, N.; Gupta, P.K.; Hajela, K. Silica supported perchloric acid
catalyzed rapid N-formylation under solvent-free conditions. Tetrahedron Lett. 2012, 53,
2063–2065.
Molecules 2014, 19 7711
41. Ma'mani, L.; Sheykhan, M.; Heydari, A.; Faraji, M.; Yamini, Y. Sulfonic acid supported on
hydroxyapatite-encapsulated-γ-Fe2O3 nanocrystallites as a magnetically Brønsted acid for
N-formylation of amines. Appl. Catal. A 2010, 377, 64–69.
42. Pathare, S.P.; Sawant, R.V.; Akamanchi, K.G. Sulfated tungstate catalyzed highly accelerated
N-formylation. Tetrahedron Lett. 2012, 53, 3259–3263.
43. De Luca, L.; Giacomelli, G.; Porcheddu, A.; Salaris, M. A new, simple procedure for the
synthesis of formyl amides. Synlett 2004, 2004, 2570–2572.
44. Deutsch, J.; Eckelt, R.; Köckritz, A.; Martin, A. Catalytic reaction of methyl formate with amines
to formamides. Tetrahedron 2009, 65, 10365–10369.
45. Baghbanian, S.M.; Farhang, M. Protic [TBD][TFA] ionic liquid as a reusable and highly efficient
catalyst for N-formylation of amines using formic acid under solvent-free condition. J. Mol. Liq.
2013, 183, 45–49.
46. Kim, J.-G.; Jang, D.O. Indium-catalyzed N-formylation of amines under solvent-free conditions.
Synlett 2010, 2010, 1231–1234.
47. Hosseini-Sarvari, M.; Sharghi, H. ZnO as a new catalyst for N-formylation of amines under
solvent-free conditions. J. Org. Chem. 2006, 71, 6652–6654.
48. Shekhar, A.C.; Kumar, A.R.; Sathaiah, G.; Paul, V.L.; Sridhar, M.; Rao, P.S. Facile
N-formylation of amines using lLewis acids as novel catalysts. Tetrahedron Lett. 2009, 50,
7099–7101.
49. Krishnakumar, B.; Swaminathan, M. A convenient method for the N-formylation of amines at
room temperature using TiO2-P25 or sulfated titania. J. Mol. Catal. A Chem. 2011, 334, 98–102.
50. Hong, M.; Xiao, G. Hafnium(IV) bis(perfluorooctanesulfonyl)imide complex supported on
fluorous silica gel catalyzed N-formylation of amines using aqueous formic acid. J. Fluorine
Chem. 2013, 146, 11–14.
51. Saidi, O.; Bamford, M.J.; Blacker, A.J.; Lynch, J.; Marsden, S.P.; Plucinski, P.; Watson, R.J.;
Williams, J.M.J. Iridium-catalyzed formylation of amines with paraformaldehyde. Tetrahedron Lett.
2010, 51, 5804–5806.
52. Zhou, L.; Freyschlag, C.G.; Xu, B.; Friend, C.M.; Madix, R.J. Direct selective oxygen-assisted
acylation of amines driven by metallic silver surfaces: Dimethylamine with formaldehyde.
Chem. Commun. 2010, 46, 704–706.
53. Xu, B.; Zhou, L.; Madix, R.J.; Friend, C.M. Highly selective acylation of dimethylamine
mediated by oxygen atoms on metallic gold surfaces. Angew. Chem. Int. Ed. 2010, 49, 394–398.
54. Ishida, T.; Haruta, M. N-formylation of amines via the aerobic oxidation of methanol over
supported gold nanoparticles. ChemSusChem 2009, 2, 538–541.
55. Preedasuriyachai, P.; Kitahara, H.; Chavasiri, W.; Sakurai, H. N-formylation of amines catalyzed
by nanogold under aerobic oxidation conditions with MeOH or formalin. Chem. Lett. 2010, 39,
1174–1176.
56. Ortega, N.; Richter, C.; Glorius, F. N-formylation of amines by methanol activation. Org. Lett.
2013, 15, 1776–1779.
57. Tumma, H.; Nagaraju, N.; Reddy, K.V. A facile method for the N-formylation of primary and
secondary amines by liquid phase oxidation of methanol in the presence of hydrogen peroxide
over basic copper hydroxyl salts. J. Mol. Catal. A Chem. 2009, 310, 121–129.
Molecules 2014, 19 7712
58. Choi, Y.S.; Shim, Y.N.; Lee, J.; Yoon, J.H.; Hong, C.S.; Cheong, M.; Kim, H.S.; Jang, H.G.;
Lee, J.S. Ionic liquids as benign catalysts for the carbonylation of amines to formamides.
Appl. Catal. A 2011, 404, 87–92.
59. Jenner, G.; Bitsi, G. Ruthenium-cobalt and ruthenium-cobalt-promoted carbonylation of primary
and secondary-amines. Appl. Catal. 1987, 32, 293–304.
60. Bitsi, G.; Jenner, G. Solvent effect in the ruthenium catalyzed carbonylation of amines. Selective
synthesis of dialkylformamides. J. Organomet. Chem. 1987, 330, 429–435.
61. Byerley, J.J.; Rempel, G.L.; Takebe, N.; James, B.R. Catalytic carbonylation of amines using
ruthenium complexes under mild conditions. J. Chem. Soc. D 1971, 1482–1483.
62. Ovchinnikov, M.V.; Guzei, I.A.; Angelici, R.J. Amine attack on the carbonyl ligands of the
protonated dicyclopentadienyl-bridged diruthenium complex [{(η5-C5H3)2(SiMe2)2}Ru2(Co)4(μ-H)]+.
Organometallics 2001, 20, 691–696.
63. McCusker, J.E.; Abboud, K.A.; McElwee-White, L. Carbonylation of amines with a tungsten(IV)
carbonyl complex. Organometallics 1997, 16, 3863–3866.
64. Süss-Fink, G.; Langenbahn, M.; Jenke, T. Rutheniumcluster als Katalysatoren für die
Carbonylierung von cyclischen Aminen. J. Organomet. Chem. 1989, 368, 103–109.
65. Jenner, G.; Bitsi, G.; Schleiffer, E. Ruthenium-catalyzed carbonylation of cyclic amines. J. Mol.
Catal. 1987, 39, 233–236.
66. Benedini, F.; Nali, M.; Rindone, B.; Tollari, S.; Cenini, S.; Lamonica, G.; Porta, F. The
bis(salicylaldehyde)ethylenediimine cobalt(II)-catalyzed oxidative carbonylation of primary and
secondary-amines. J. Mol. Catal. 1986, 34, 155–161.
67. Dombek, B.D.; Angelici, R.J. Pentacarbonyliron-catalyzed carbonylation of amines to
formamides. J. Catal. 1977, 48, 433–435.
68. Gabriele, B.; Salerno, G.; Costa, M. Oxidative carbonylations. In Catalytic Carbonylation
Reactions; Beller, M., Ed.; Springer: Heidelberg, Germany, 2006; pp. 239–272.
69. Díaz, D.J.; Darko, A.K.; McElwee-White, L. Transition metal-catalyzed oxidative carbonylation
of amines to ureas. Eur. J. Org. Chem. 2007, 4453–4465.
70. Ragaini, F. Away from phosgene: Reductive carbonylation of nitroarenes and oxidative
carbonylation of amines, understanding the mechanism to improve performance. Dalton Trans.
2009, 2009, 6251–6266.
71. Gabriele, B.; Mancuso, R.; Salerno, G. Oxidative carbonylation as a powerful tool for the direct
synthesis of carbonylated heterocycles. Eur. J. Org. Chem. 2012, 6825–6839.
72. Saegusa, T.; Kobayashi, S.; Hirota, K.; Ito, Y. Synthetic reactions by complex catalysts. XIII.
Carbonylation of amines by group IB and IIB metal compound catalysts. Bull. Chem. Soc. Jap.
1969, 42, 2610–2614.
73. Tsuji, Y.; Ohsumi, T.; Kondo, T.; Watanabe, Y. Dodecacarbonyltriruthenium catalyzed
carbonylation of amines and hydroamidation of olefins. J. Organomet. Chem. 1986, 309, 333–344.
74. McCusker, J.E.; Logan, J.; McElwee-White, L. Oxidative carbonylation of primary amines to
ureas using tungsten carbonyl catalysts. Organometallics 1998, 17, 4037–4041.
Molecules 2014, 19 7713
75. McCusker, J.E.; Main, A.D.; Johnson, K.S.; Grasso, C.A.; McElwee-White, L. W(CO)6-catalyzed
oxidative carbonylation of primary amines to N,N'-disubstituted ureas in single or biphasic solvent
systems. Optimization and functional group compatibility studies. J. Org. Chem. 2000, 65,
5216–5222.
76. McCusker, J.E.; Qian, F.; McElwee-White, L. Catalytic oxidative carbonylation of aliphatic
secondary amines to tetrasubstituted ureas. J. Mol. Catal. A Chem. 2000, 159, 11–17.
77. McCusker, J.E.; Grasso, C.A.; Main, A.D.; McElwee-White, L. Catalytic oxidative carbonylation
of primary and secondary α,ω-diamines to cyclic ureas. Org. Lett. 1999, 1, 961–964.
78. Qian, F.; McCusker, J.E.; Zhang, Y.; Main, A.D.; Chlebowski, M.; Kokka, M.; McElwee-White, L.
Catalytic oxidative carbonylation of primary and secondary diamines to cyclic ureas.
Optimization and substituent studies. J. Org. Chem. 2002, 67, 4086–4092.
79. Hylton, K.G.; Main, A.D.; McElwee-White, L. Catalytic carbonylation of functionalized
diamines: Application to the core structure of DMP 323 and DMP 450. J. Org. Chem. 2003, 68,
1615–1617.
80. Darko, A.K.; Curran, F.C.; Copin, C.; McElwee-White, L. Carbonylation of functionalized
diamine diols to cyclic ureas: Application to derivatives of DMP 450. Tetrahedron 2011, 67,
3976–3983.
81. Zhang, L.; Darko, A.K.; Johns, J.I.; McElwee-White, L. Catalytic oxidative carbonylation of
arylamines to ureas with W(CO)6/I2 as catalyst. Eur. J. Org. Chem. 2011, 6261–6268.
82. Díaz, D.J.; Hylton, K.G.; McElwee-White, L. Selective catalytic oxidative carbonylation of amino
alcohols to ureas. J. Org. Chem. 2006, 71, 734–738.
83. Dumbris, S.M.; Díaz, D.J.; McElwee-White, L. Preparation of hydantoins by catalytic oxidative
carbonylation of α-amino amides. J. Org. Chem. 2009, 74, 8862–8865.
84. Shelton, P.A.; Zhang, Y.; Nguyen, T.H.H.; McElwee-White, L. NaIO4-oxidized carbonylation of
amines to ureas. Chem. Commun. 2009, 2009, 947–949.
85. Gerack, C.J.; McElwee-White, L. Oxidative carbonylation of amines to formamides using NaIO4.
Chem. Commun. 2012, 48, 11310–11312.
86. Gerack, C.J.; Johns, J.I.; McElwee-White, L. Base-mediated carbonylation of amines to
formamides. Department of Chemistry, University of Florida, Gainesville, FL 32611-7200, USA,
2014; Unpublished work.
© 2014 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/3.0/).