Evaluation and optimization of synthetic routes from dihydroartemisinin to the alkylamino-artemisinins artemiside and artemisone: A test of N-glycosylation methodologies on a lipophilic peroxide

Article abstract of DOI:10.1016/j.tet.2018.04.027

10-Alkylamino-artemisinins including artemiside and artemisone display enhanced activities against malaria. Earlier, dihydroartemisinin (DHA) TMS ether was converted by trimethylsilyl bromide into the 10-β-bromide that with amine nucleophiles provided the amino-artemisinins. In an attempt to develop more economic approaches, direct N-glycosylation of DHA was examined but 2-deoxyartemisinin was invariably obtained. However, hydroxyl group activation by conversion into the 10β-halide in non-polar solvents with anhydrous HCl and Group I and II metal halides, oxalyl chloride or thionyl chloride with catalytic DMSO, and oxalyl bromide did succeed. The β-halides were converted in situ by thiomorpholine into artemiside, and by thiomorpholine-1,1-dioxide into artemisone respectively in scalable reactions. Hydrogen peroxide-acetonitrile or the urea-hydrogen peroxide complex efficiently oxidized the sulfide artemiside to the sulfone artemisone. Overall, a generalized approach to 10-alkylamino-artemisinins is now available.

Full text of DOI:10.1016/j.tet.2018.04.027

Accepted Manuscript
Evaluation and optimization of synthetic routes from dihydroartemisinin to the
alkylamino-artemisinins artemiside and artemisone: A test of N-glycosylation
methodologies on a lipophilic peroxide
Wing Chi Chan, Dennis Ho Wai Chan, Kin Wo Lee, Wing Shan Tin, Ho Ning Wong,
Richard K. Haynes
PII:
S0040-4020(18)30406-X
10.1016/j.tet.2018.04.027
TET 29442
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 5 March 2018
Revised Date: 5 April 2018
Accepted Date: 10 April 2018
Please cite this article as: Chan WC, Wai Chan DH, Lee KW, Tin WS, Wong HN, Haynes RK, Evaluation
and optimization of synthetic routes from dihydroartemisinin to the alkylamino-artemisinins artemiside
and artemisone: A test of N-glycosylation methodologies on a lipophilic peroxide, Tetrahedron (2018),
doi: 10.1016/j.tet.2018.04.027.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Evaluation and Optimization of Synthetic Routes from Dihydroartemisinin to the Alkylamino-artemisinins
Artemiside and Artemisone: A Test of N-Glycosylation Methodologies on a Lipophilic Peroxide.†
Wing Chi Chana, Dennis Ho Wai Chana, Kin Wo Leea, Wing Shan Tina, Ho Ning Wongb, Richard K. Haynesa,b,*
a Department of Chemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon,
Hong Kong, PR China.
b Centre of Excellence for Pharmaceutical Sciences, Faculty of Health Sciences, North-West University,
Potchefstroom, South Africa.<<please note Dr H. N. Wong affiliation is NWU>>
Abstract
10-Alkylamino-artemisinins including artemiside and artemisone display enhanced activities against
malaria. Earlier, dihydroartemisinin (DHA) TMS ether was converted by trimethylsilyl bromide into the 10-β-
bromide that with amine nucleophiles provided the amino-artemisinins. In an attempt to develop more economic
approaches, direct N-glycosylation of DHA was examined but 2-deoxyartemisinin was invariably obtained.
However, hydroxyl group activation by conversion into the 10β-halide in non-polar solvents with anhydrous HCl
and Group I and II metal halides, oxalyl chloride or thionyl chloride with catalytic DMSO, and oxalyl bromide did
succeed. The β-halides were converted in situ by thiomorpholine into artemiside, and by thiomorpholine-1,1-
dioxide into artemisone respectively in scalable reactions. Hydrogen peroxide-acetonitrile or the urea-hydrogen
peroxide complex efficiently oxidized the sulfide artemiside to the sulfone artemisone. Overall, a generalized
approach to 10-alkylamino-artemisinins is now available.
Graphical Abstract
Keywords: Malaria, Antimalarial Drugs, Artemisinins, N-Glycosylation, Amino-artemisinins, Artemisone
†
Dedicated to the memory of Professor Sir Derek Barton, one of the greatest of organic chemists of the 20th
Century and a master of all trades within the profession of organic chemistry. He was an inspiring mentor who
graciously and enthusiastically shared his knowledge, and he effectively imparted the sense of foresight and
introspection required for the conception and successful execution of the varied and potent research projects in
which the members of his research groups were engaged. The art and form thereof represented the ideal models
upon which to base our activities once we had moved on to become engaged in the complicated tasks of
conceiving and then supervising our own research projects. The time at Imperial College was indeed the high point
in my early career path.

2
ACCEPTED MANUSCRIPT
1. Introduction
The frequency and severity of malaria infections experienced by Vietnamese and Chinese personnel
during the US-Vietnam war compounded with restricted access to antimalarial medicines in the 1960s impelled the
Chinese leadership to initiate a nation-wide effort to solve the malaria problem. Thereupon, with the involvement
of the Chinese National Commission of Science and Technology, the Ministry of Health, the Academy of Science,
regional Institutes Materia Medica, and government pharmaceutical agencies, Project 523 was launched on the 23
May 1967 – just over 50 years ago – with the aim of both developing synthetic compounds and surveying herbs
used in traditional Chinese medicine for new antimalarial drugs. The Chinese medicinal herb qīng hāo 青蒿
(Artemisia annua) that featured in early Chinese pharmacopoeia was selected as a potential source, and through
bioassay guided fractionation of extracts in the early 1970s, an antimalarial-active principle was identified. It was
at the Shandong Institute Materia Medica and the Yunnan Institute Materia Medica that the active constituent was
isolated in sufficient purity so as to actually cure patients with vivax and falciparum malaria respectively. This was
the crystalline compound qīng hāo sū 青蒿素, the active principle of qīng hāo. There followed a coordinated effort
under the Project 523 umbrella focusing on determining its structure, toxicity, interaction with other drugs, , and
efficacy in clinical trials against malaria. With determination of its X-ray crystal structure in 1975 and absolute
configuration in 1978, coupled with chemical degradation methods, the compound was shown to be the
sesquiterpene lactone peroxide 1 (Fig. 1). Additionally, with a third oxygen atom in the peroxide-containing ring, it
was the first example of a 1,2,4-trioxane isolated from a natural source. The English name of artemisinin sub-
sequently appeared, and was later used in the Chinese pharmacopoeia. These notable historical aspects have been
carefully reviewed.1,2,3
Fig. 1. Artemisinin 1 and its current clinical derivatives: the hemiacetal dihydroartemisinin (DHA) 2, the lactol
ether artemether 3, and the hemiester artesunate 4. The latter two are rapidly converted into DHA via facile
metabolism or hydrolysis respectively. DHA rearranges irreversibly under physiological conditions into the
peroxyhemiacetal 5 that in turn rearranges to the inert deoxyartemisinin 6 (ref. 23).
Because of solubility and formulation problems with artemisinin 1 itself, a synthetic chemistry effort
involving the Institutes Materia Medica (IMM) in Beijing and Shanghai and the Shanghai Institute of Organic
Chemistry and others aimed at preparing more tractable derivatives.1,2 Thus, artemisinin was reduced to the lactol
dihydroartemisinin (DHA) 2 by borohydride-based reagents in methanol, a notable reaction given the general
inertness of lactones to reduction with such reagents.4 At the Shanghai IMM in 1976, DHA was converted into the
oil-soluble lactol ether artemether 3 and its α-epimer by reaction of DHA with methanol and an acid catalyst;
fractional crystallization was required to isolate the pure β-epimer 3. At the Guilin Pharmaceutical Factory in 1977
DHA was converted into the hemiester artesunate 4, produced as the pure α-epimer, by reaction of DHA and
succinic anhydride, as part of a drive to generate water soluble derivatives.5,6 Thereby, the artemisinins were
introduced into the armamentarium of drugs for combatting malaria. Indeed, their effectiveness in rapidly clearing
parasitaemia was recognized early in Chinese clinical trials, and they rapidly assumed front-line status as

3
ACCEPTED MANUSCRIPT
antimalarial drugs on a worldwide basis. Overall, the discovery of the antimalarial activity of artemisinin and its
derivatives represents one of the notable developments in medicine in the latter third of the 20th century.7
Irrespective of the clinical effectiveness of these compounds for the treatment of malaria, their advent has raised
fascinating questions on the nature of their intracellular mechanisms,3,8,9 and raised awareness of the importance of
peroxide xenobiotics in modulation of cell physiology. In addition, these compounds and newer derivatives have
evinced considerable potential for the therapy of parasitic diseases other than malaria,10 for non-parasitic infectious
diseases11,12,13 and cancer.14
The artemisinins 1-4 are used in artemisinin combination therapies (ACTs) with longer half-life drugs for
treatment of malaria.3,15 Their metabolism is facile and subject to marked auto-induction. Thus, time-dependent
enhancement of metabolism as observed in malaria patients may result in drug plasma levels falling below the
effective therapeutic dosage.16 Artesunate 4 is hydrolysed in vitro at physiological pH (pH 2 or 7.4) and in vivo,17
and artemether 3 is metabolized by oxidative dealkylation,18 to DHA 2. The half-life of artesunate is several
minutes, whereas artemether 3 is similar to that of DHA at ~1 h.19 Intriguingly, DHA undergoes a facile and
irreversible rearrangement under physiological conditions into the peroxyhemiacetal 5 that also possesses
antimalarial activity.20,21 Importantly, this compound has been detected in the plasma of patients treated with
artesunate,22 and is unstable by virtue of a further facile rearrangement into the inert deoxyartemisinin 6.23 The
neurotoxicity of the artemisinins, in particular of DHA, is well demonstrated.24 Comparative assays with neuro-
toxins such as paraquat suggest that neurotoxicity is due to inhibition of the respiratory chain and rapid induction
of oxidative stress in neuronal cells at concentrations >0.001 µg/mL. Additionally, degradation of cytoskeleton
elements continues beyond the period of drug administration, a classical indicator of neurotoxicity.25,26 In vivo
screens elicit a characteristic neurotoxic behaviour pattern and histopathology reveals a neurodegeneration in the
brain stem with impairment of auditory, vestibular, cerebellar, motor and reticular activating systems.27 Whilst by
and large neurotoxicity generally has not been expressed in humans treated with these drugs,28 there are evident
exceptions,29 and the issue has been debated.30 However, the perception of a neurotoxic burden is enhanced by the
report of a death due to an artesunate overdose.31 Yet remarkably, in recent times the problem of neurotoxicity has
been subsumed by the need to suppress artemisinin resistance by applying more protracted dose regimens of
ACTs, where clearly enhanced pharmacovigilance is warranted, particularly in a pediatric situation.32
A large number of artemisinin derivatives has been made, primarily with the intent of exploring structure-
activity relationships and establishing activities against disease targets in addition to malaria, although relatively
few have been carried forward for further development.2,33,34 In our own program, we planned to replace the
exocyclic oxygen atom at C-10 in the current derivatives by nitrogen, based on ADME precepts.35 Thereby, with
the aims of preventing metabolism to DHA and countering toxicity, and improving the antimalarial efficacy, we
prepared 10-alkylamino- and 10-arylamino-artemisinin and other derivatives. 36 Of the aminoartemisinins,
artemisone 8 (BAY 44-9585, also named artemifone), prepared from the thiomorpholine precursor 7 (artemiside,
BAY 44-9584) (Fig. 2), emerged as an early clinical development candidate from a collaborative venture involving
the Hong Kong University of Science and Technology, Bayer AG and Bayer HealthCare AG, and in part the
Medicines for Malaria Venture, Geneva.37 Like other aminoartemisinins, artemisone is obtained exclusively as the
α-epimer (cf. artemether 3, Fig. 1), is highly crystalline with an aqueous solubility at pH 7.2 of 89 mg/L and a Log
P (partition coefficient n-octanol-water) at pH 7.4 of 2.49 that characterizes it as a relatively non-lipophilic drug. In
in vitro screens, it elicits no neurotoxicity towards brain stem cell cultures and neurofilaments up to 10 µg/mL, and
has no effect on the respiratory chain or on oxidative stress in neuronal cells.26 Lack of neurotoxicity was also
verified in in vivo screens and in all other animal screens involved in the preclinical development of artemisone. In
particular in pilot tolerability studies involving oral administration in sesame oil to female rats at 50 mg kg-1 d-1 for
14 days, increase in body weight was as for the controls with no detectable adverse events.26,38

4
ACCEPTED MANUSCRIPT
Fig. 2. The amino-artemisinins artemiside and artemisone in which the exocyclic oxygen atom at C10 of the
current clinical artemisinins (Fig. 1) is replaced by a nitrogen atom. Comparative in vitro and in vivo assays
conducted at the same time under the same conditions with artemisone, other amino artemisinins and one or more
of the clinical artemisinins indicate artemisone is the most active, least toxic compound (ref. 37).
Artemisone is up to nine-fold more active than artesunate against the malaria parasite Plasmodium
falciparum in vitro37,39,40 and at least three-fold in in vivo. It appears to be the most potently active of artemisinins
and of synthetic peroxides in general against P. vivax in vitro.41 The enhanced efficacy is reflected in the outcome
of a Phase IIa trial on patients with uncomplicated falciparum malaria in Thailand, wherein cure was achieved with
an oral administration of 2 mg/kg per day for two days followed by a final dose of 25 mg/kg of mefloquine. This
corresponds to a three-fold lower curative dose compared with that of artesunate (4 mg/kg per day for 3 days).37,42
The feasibility of employing lower dose regimens over a shorter time period is also indicated by the results of
primate studies.37 Artemisone is not metabolized to DHA 2, but rather to metabolites bearing unsaturation in the
thiomorpholine-S,S-dioxide ring, and/or hydroxyl groups at C-5 or C-7 that are also active antimalarials.37,43 DHA
is not formed upon incubation of artemisone with liver microsomes, hepatocytes and recombinant CYP isoforms,
and is not present in the plasma of malaria patients treated with artemisone.42 Importantly, it displays no clinically-
relevant autoinduction of metabolizing enzymes.37,43 The benefit is that plasma concentrations of artemisone and
its active metabolites do not vary significantly over the administration period; this was demonstrated by daily
assessment of plasma concentrations in subjects in the Phase I and II studies.43,44,45 In comparison to DHA,46
artemisone possesses a longer elimination half-life (~3 h vs ~1 h), lower plasma clearance (~3.5 L/h/kg vs ~5.4
L/h/kg) and a larger volume of distribution (14.5 L/kg vs 7.7 L/kg).44,45 Overall, artemisone is the only artemisinin
derivative developed outside China that has progressed into a clinical trial against malaria. 47 Artemisone has now
been registered with the US FDA as an orphan drug planned for use in treatment of severe malaria via intravenous
administration.48
Artemiside and artemisone are active against the apicomplexan parasite Toxoplasma gondii, in particular
in a murine model wherein the compounds are significantly more active than artemisinin.49 Artemisone is active
against the veterinary parasite Neospora caninum in an animal model50 and against N. caninum tachyzoites in vitro
with an IC50 of 3 nM.51 For non-parasitic pathogens, artemisone is cytotoxic towards the ulcerogenic bacterium
Helicobacter pylori, in particular in combination with known antibiotics,52 and is a potent inhibitor of human
cytomegalovirus with an efficacy comparable to that of ganciclovir.53 Artemisone is among the most potent of all
antimalarial peroxides when screened against tumour cell lines,54 an effect which is enhanced by addition of other
cancer drugs.55 In relation to the mechanism of action, artemisone like other amino-artemisinins36 does not interact
with haemoglobin-Fe(II), and reacts poorly with iron(II) or with heme iron(II) under biomimetic conditions,
although partial hydrolysis to DHA is observed.56 Biological activity is claimed to be associated with alkylation of
heme iron(II) based on the outcome of non-biomimetic experiments when heme iron(III) is reduced with sodium
dithionite in anhydrous DMSO in an inert gas atmosphere, and then treated with the artemisinin.57 However,
artemisinins retain antimalarial activities in vitro against malaria parasites cultured under an atmosphere containing
carbon monoxide. CO reacts with heme-Fe(II) to form the stable carboxyheme-Fe(II) complex that does not react

5
ACCEPTED MANUSCRIPT
with artemisinins.56 Significantly, those artemisinins that do react with heme-Fe(II) under biomimetic conditions
show improved activities against the malaria parasite under the CO atmosphere, whereas those that do not react
with heme-Fe(II), such as artemisone, retain the same activities. Overall, the suppositions on the biological
significance of the notably abiotic heme-alkylation experiments have been dealt with, namely the reaction with
heme in a biological medium represents a competing degradation which results in attrition of efficacy, and is
unrelated to mechanism of action.9,36,56,58 The ability of the peroxide group to oxidize reduced flavin cofactors
important in modulating oxidative stress in the malaria parasite has been demonstrated.20, 59 That amino-
artemisinins are optimum substrates for eliciting parasiticidal action may be rationalized in terms of this oxidant
model.60
In general, as compared to the relatively facile preparations of the artemisinins of Fig. 1, amino-arte-
misinins are less accessible - N-glycosylation methodologies have to be used on DHA 2 containing a peroxide that
confers chemical and thermal instability. However, the considerable promise of clinical use of artemisone 8 against
malaria and other parasitic and non-parasitic pathogens impelled us to improve the preparation of amino-
artemisinins, and in particular, of artemisone 8 such that this becomes economically accessible on a large scale.
Use of reagents and solvents and distinct chemical operations must be minimized, and the product must be isolated
without the need for chromatographic purification. We describe here the efforts that eventually led to new scalable
processes for preparation of the amino-artemisinins from DHA 2.
2. Discussion
2.1. Attempted Direct Preparation from Artemisinin or DHA
As artemisinin 1 is extracted from Artemisia annua, its direct conversion into artemisone will confer a
cost advantage. Although a lactone, artemisinin reacts like a ketone with hydride and other nucleophiles s to give
stable hemiacetal adducts.4,6,61,62 In addition, artemisinin reacts cleanly with ammonia and primary alkyl amines
under protic conditions to give adducts that in a subsequent step are transformed into 11-aza-artemisinins.63
However, attempted reduction of the putative adducts formed in situ from artemisinin and 1° and 2° amines
including thiomorpholine with sodium triacetoxyborohydride64 or borane-amine based reagents65 was unsuccessful,
and mixtures of products arising via reduction of the peroxide were generally obtained.
DHA 2 in solution is an equilibrating mixture of the α- and β-epimers (Scheme 1) whose relative amounts
depend upon the solvent and the presence of organic bases.66, 67 The pure β-epimer 2β with axial hydroxyl
crystallizes from solution or may be sublimed from solid DHA under high vacuum.66 Epimerization to the α-
epimer in solution likely occurs via the aldehyde 14. The Kotchetkov aminoglycosylation reaction involves
condensation of amines with aldoses in aqueous alcohol containing inorganic salts,68 and works well for 2° amines
such as indoles with glucose69 and substituted piperazines with glucose or galactose.70 The reaction proceeds via
ring opening of the aldose to the aldehyde, that condenses with the amine to generate the iminium intermediate;
this undergoes ring closure to the N-glycosylamine. However, under the conditions of the Kotchetkov reaction it
was not possible to intercept the aldehyde 14 from DHA with thiomorpholine or other 2° amines, and 2-deoxy-
artemisinin 6 (Fig. 1) was obtained as the only identifiable product. Its formation likely proceeds via ring opening
of DHA under the protic conditions to an open hydroperoxy keto-aldehyde, reclosure to the peroxyhemiacetal 5
and collapse of the latter by a Kornblum-de la Mare reaction.23,34 Its appearance under these conditions underscores
the labile nature of DHA, and its general unsuitability as a drug for malaria.23

6
ACCEPTED MANUSCRIPT
Scheme 1. Chair pyranose forms of equatorial 2α and axial epimers 2β of DHA, and their proposed equilibration
via aldehyde 14.
2.2. Preparation of artemiside 7
Activation of the anomeric hydroxyl group in pyranose sugars followed by displacement with nucleo-
philes leads to glycosides.71 However, the reactivity of the peroxide in DHA restricts the range of N-glycosylation
reactions that can be used.In addition, the C-10 (anomeric) hydroxyl in DHA is attached to a chair pyranose ring
buttressed against a cis-fused cyclohexano system (Scheme 2). Thus, for an amine nucleophile, an S 2 reaction
N
with the α-equatorial activated DHA leading to the β-axial product is inhibited by the steric hindrance due to the
axial C8-C8a bond; similarly an S 1 process involving Re-face attack on the oxonium ion 15 is also disfavoured
N
(Scheme 2). The β-axial-activated DHA will be prone to E2 elimination because of the trans-diaxial arrangement
of H-9 and leaving group at C-10. However, the predominant pathways for both activated epimers are likely to be
E1 and S 1 reactions involving the oxonium ion 15, with steric or stereoelectronic (anomeric) effects determining
N
the outcome of the latter reactions. In this respect, the α-equatorial trichloroacetimidate obtained via the Schmidt
procedure on DHA provides β-axial esters according to the anomeric effect.67 As amines are not subject to such an
effect, α-aminated products are expected to form.
Scheme 2. Axial and equatorial activated intermediates from DHA 2 that via S 2 and S 1 reactions involving
N
N
oxonium ion 15 may provide α- and β-aminated products with amine nucleophiles. The axial 8-8a bond in the α-
equatorial activated DHA will inhibit S 2 reactions. Si-face addition of the nucleophile leading to the α-aminated
N
product is also favoured for S 1 reactions because of the axial 8-8a bond in 15. E2 and especially E1 reactions
N
leading to anhydrodihydroartemisinin (glycal) 16 will compete when basic amine nucleophiles are used. The cases
below involve the β-axial activated DHA where LG = Br or Cl to provide the α-aminated products.
Thus, conversion of the hydroxyl into halide in aprotic solvents under the mildest possible conditions was
required. We originally converted DHA by treatment with diethylaminosulfur fluoride into the 10β-fluoride,72,73
but the expense of the reagent and the low yield and instability of the product excluded this activation method.
According to the conversion of fucopyranose trimethylsilyl ether into the axial iodide by treatment with trimethyl-
silyl iodide,74 DHA was converted into the crystalline α-trimethylsilyl ether 17 that with trimethylsilyl bromide
(TMSBr) in dichloromethane cleanly provided in situ the axial β-bromide 18 (Scheme 3). Exclusive formation of
1
the β-product as favoured by the anomeric effect was revealed by H NMR analyses of reaction mixtures.73 The
bromide was converted by treatment in situ with 1° and 2° alkylamines and 1° arylamines into 10-α-amino-arte-
misinins accompanied by the glycal 16.36 The amino derivatives were formed exclusively as the α-epimers. 10-

7
ACCEPTED MANUSCRIPT
Arylamino-artemisinins have been made by other methods,75 but our method is the first for making 10-alkylamino-
artemisinins. In turning to artemisone 8, we now describe optimization of the preparation of artemiside 7 from
DHA and thiomorpholine that is converted in a separate step into artemisone, and the direct preparation of
artemisone from DHA and thiomorpholine-1,1-dioxide.
2.2.1. DHA trimethylsilyl ether and trimethylsilyl bromide: We originally reported that treatment of β-bromide 18
generated in situ from DHA TMS ether 17 and TMSBr with thiomorpholine (2.5 equiv.) provided artemiside 7 in
58% yield.37 In seeking to augment the amount of thiomorpholine by using amine bases triethylamine and hexa-
methyldisilazane or inorganic bases, yields were not significantly improved. Larger scale preparations (to 10
mmole DHA) concluding with a direct recrystallization of the crude product mixture gave lower isolated yields
(30-35%) of artemiside. The non-polar glycal 16 invariably formed as a by-product interfered with the crystal-
lization of artemiside, and normally had to be separated by chromatography. The preparation of artemiside
according to this process, in which discrete isolation of a product is labelled as a 'step', is summarized in Scheme 3.
Disadvantages include the use of TMSBr and excess thiomorpholine, and eventual chromatography of mother
liquors to separate residual artemiside 7 from the glycal 16.
Scheme 3. First process for the preparation of artemiside.
2.2.2. Trimethylsilyl bromide in situ: Although generation of TMSBr in situ requires polar solvents,76 we found that
the β-bromide 18 was readily generated by the treatment of DHA and sodium bromide with trimethylsilyl chloride
(TMSCl) in toluene.37 Next, addition of thiomorpholine to the mixture gave 7 in 58% isolated yield for a reaction
conducted on a 1 mmole scale (Scheme 4). Again, use of amine or inorganic bases did not significantly affect
yields, and the glycal 16 was always formed. At a 2-3 mole scale, for which a direct crystallization was employed,
the isolated yields decreased to ~41%. Thus, the process is scalable, but although it required fewer steps and
cheaper reagents, the use of an excess of thiomorpholine, and lower yields at large scale detract from its utility.
Scheme 4: Second process for the preparation of artemiside
As noted above (Scheme 2), S 1/S 2 and E1/E2 processes convert the β-activated DHA derivative,
N
N
namely bromide 18, into artemiside 7 and glycal 16 respectively. Thus, effect of solvents on the artemiside:glycal
ratios required evaluation. An NMR method was used to assess relative amounts of unreacted DHA, artemiside and
glycal in the product mixtures by comparing signal integrals with those of the aromatic protons of the internal
standard 1,3,5-trimethoxybenzene. The chemical shifts of the signals due to DHA and the products in the
downfield region are well separated (Table 1), and assessment of ratios was straightforward.

8
ACCEPTED MANUSCRIPT
Table 1: 1H-NMR chemical shifts of characteristic signals.a
Compounds
Chemical Shifts δH (ppm) and Coupling Constants J (Hz)
DHA 2
α-epimer: 4.73, d, J10,9 = 8.7 Hz, H-10
β-epimer: 5.31, d, J10,9 = 3.6 Hz, H-10 β-epimer: 5.61, s, H12
α-epimer: 5.38, s, H12
Glycal 16
Artemiside 7
Artemisone 8
6.13, q, J9-Me,10 = 1.33 Hz, H-10
3.93, d, J9,10 = 10.2 Hz, H-10
4.21, d, J9,10 = 10.3 Hz, H-10
5.53, s, H-12
5.23, s, H-12
5.28, s, H-12
6.08, s
1,3,5-Trimethoxybenzeneb 3.75, s
a300 MHz, CDCl3; binternal standard.
Lewis basic solvents (Table 2, entries 1-3) generally enhanced formation of the glycal 16 whereas non-
polar solvents enhanced formation of artemiside 7. The bromide 18 decomposes to the glycal via E1 or E2
processes where the basic solvent or thiomorpholine abstracts α-axial H-9 (Scheme 2). Thus, substitution is
favoured in non-polar solvents. The best artemiside:glycal ratios were obtained with toluene, cyclohexane and
dichloromethane, and best product yields with toluene. The lower solubility of DHA in cyclohexane resulted in a
poorer conversion. Use of longer reaction times did not improve the yield. In contrast, the crude product mixture
from reactions in toluene contained very little DHA. Thus, focus turned to use of halogenated solvents and toluene.
Table 2: Results of Reactions in Various Solvents a
Artemiside 8:Glycal 16
% Artemiside 7 formed
Entry
Solvent
1
2
3
4b
5
6
7
8
(CH3)2C=O
CH3CN
29:71
33:67
38:62
65:35
61:39
73:27
63:37
74:26
22
11
13
53
54
58
27
36
CH3CO2CH2CH3
(CH3CH2)2O
CH2Cl2
PhCH3
n-hexane
cyclohexane
a.All reactions performed with 1 mmol of DHA 2, 1.1 mmol of sodium bromide and 2.2 mmol of trimethylsilyl chloride
b
in 1 mL of solvent stirred for 15 min and then treated with 2.5 mmol of thiomorpholine, except for entry 4; the
mixture in diethyl ether was stirred overnight before addition of thiomorpholine.
2.2.3. Hydrogen chloride: In accord with the chemistry of anhydro sugars,77 glycal 16 is rapidly converted into the
β-chloride 19 by gaseous HCl in dichloromethane. The reaction was used to prepare isotopically labelled artemi-
sone from labelled glycal.37 However, attempts to convert DHA in dichloromethane into 19 with gaseous HCl, or
by HCl generated in situ, e.g. from acetyl chloride,78 were unsuccessful. In contrast, when anhydrous lithium, mag-
nesium or calcium chlorides (1.2-1.6 equiv.) were added, DHA was now rapidly converted into chloride 19. This
1
was assessed for reactions in CD2Cl2 or CDCl3 where after filtration to remove the salts, the H NMR spectrum
cleanly revealed the presence of the β-chloride 19 (Fig. 3).

9
ACCEPTED MANUSCRIPT
1
Figure 3. H NMR spectrum of the β-chloride 19 (300 MHz, CD2Cl2). The signal due to H-12 is at δ 5.62 ppm,
and H-10 is at δ 6.19 ppm; with J10,9 = 3.3 Hz, H-10 is axial with a cis axial-equatorial coupling to H-9. Signals at
δ 5.53 and 6.19 ppm are due the glycal 16.
Treatment of the β-chloride 19 in situ with thiomorpholine-triethylamine gave artemiside in 54-57%
isolated yields (Scheme 5). The salts may assist in activating the hydroxyl group to displacement by chloride,
and/or by scavenging water, suppress hydrolysis of the β-chloride to DHA. Use of anhydrous sodium chloride or
Lewis acids such as zinc or tin(II) chlorides resulted in lower yields of artemiside, usually at the expense of the
glycal. Reaction mixtures obtained from preparative scale experiments with 18 mmole (ca. 5 g) or more of DHA
were quenched with brine, followed by separation of the organic layer, direct evaporation, and recrystallization of
the residue. No drying was used. However, a simple filtration of the reaction mixture, especially at larger scale, to
remove the insoluble metal halide and amine hydrochloride salts formed in the reaction, followed by direct
evaporation of the filtrate sufficed to provide crude product that was crystallized directly. Successive crops of the
product were obtained from the mother liquors. However, dilute solutions of DHA had to be used for optimum
yields – at an 18 mmole scale, 0.2 M solutions of DHA were required. Overall, the HCl process is readily scalable,
but from an industrial aspect dichloromethane is not regarded as a suitable solvent. However, with toluene
containing anhydrous magnesium chloride, the isolated yield of artemiside decreased to 17%. With magnesium
bromide that should generate the more reactive HBr in situ from the gaseous HCl in toluene, DHA provided
artemiside in 50% isolated yield, but the solution was difficult to handle once the thiomorpholine had been added,
and in order to isolate artemiside, the mixture had to be diluted with dichloromethane.
Scheme 5. Third process for the preparation of artemiside.
2.2.4. Trimethylsilyl chloride-dimethyl sulfoxide: TMSCl and catalytic DMSO in dichloromethane convert alcohols
into alkyl chlorides.79 The reaction proceeds via a siloxysulfonium chloride formed from the DMSO and TMSCl
followed by activation of the alcohol via a sulfurane intermediate and displacement by chloride. Because of the
cheapness of the reagents, the reaction was examined here. DHA 2 in CD2Cl2 treated with TMSCl (2 equiv.) and

10
ACCEPTED MANUSCRIPT
1
d6-DMSO (0.25 equiv.) gave cleanly the 10β-chloride 19 as verified by H NMR spectroscopy. At a 1 mmole
scale, artemiside was obtained in 60% yield, but more of the glycal 16 was now formed. Use of preparative scale
experiments to 18 millimole resulted in lower isolated yields (46%) (Scheme 6); this is ascribed to difficulty of
inducing crystallization of artemiside in the presence of large amounts of the glycal, as noted above. The reaction
did not succeed in toluene.
Scheme 6. Fourth process for the preparation of artemiside
2.2.5. Oxalyl chloride: Oxalyl chloride has long been used to convert alcohols into alkyl halides.80 It reacts by
various pathways depending upon the amount of the reagent and the conditions.81 In the presence of pyridine, an
oxalate hemiester is formed from an alcohol that on heating decomposes to the alkyl chloride.82 In the absence of a
base, the reaction with a 3° alcohol directly provides the alkyl halide, assisted by the generation of HCl during the
course of the reaction.80 Treatment of DHA and triethylamine with oxalyl chloride (0.5 equiv.) in CD2Cl2 gave a
product presumed to be the bis-oxalate ester, whose formation was indicated by the appearance of a signal in the
1H NMR spectrum at δ 5.83 ppm (H-10, J10,9 9.9 Hz) characteristic of α-esters of DHA in which H-10 displays a
trans-diaxial coupling with H-9.67 Attempts to isolate the ester by crystallization from the reaction mixture resulted
in its decomposition to the glycal 16. Treatment of the bis-oxalate ester with thiomorpholine in situ did not give
significant amounts of artemiside. The reaction was therefore repeated in the absence of triethylamine. Although
clean formation of the β-chloride 19 from DHA and oxalyl chloride in CD2Cl2 was confirmed by 1H NMR spectro-
scopy, the reaction was surprisingly slower than the DHA-HCl-salt reactions (cf. Scheme 5) in this solvent. At a
1.76 mmole scale at 0 °C, 10 minutes were required for the latter reaction, whereas the DHA-oxalyl chloride
reaction required 4 hours. Treatment of the reaction mixtures with thiomorpholine-triethylamine gave artemiside 7
in 45-50% isolated yields. Concentrated solutions of DHA ranging from 0.45-0.9M DHA could now be used
without diminution in yields. In contrast, for the HCl reactions at an 18 millimole scale, limiting concentrations of
approximately 0.3M DHA had to be used for optimum yields. However, the best isolated yields of artemiside
obtained for the oxalyl chloride reactions, irrespective of the scale of the reaction, did not exceed 50%. In contrast
to the HCl reactions, toluene could now be used as solvent, although isolated yields of artemiside were inferior to
those from reactions in dichloromethane.
2.2.6. Oxalyl chloride-dimethyl sulfoxide and other promoters: DMSO is paired with oxalyl chloride in the Swern
oxidation that proceeds via the chlorosulfonium intermediate 20. Formation of the chlorosulfurane adduct from the
alcohol is followed by its conversion into the carbonyl product and dimethyl sulfide in the presence of a 3° amine
base. In the absence of base, the sulfurane of an S 1-reactive alcohol is anticipated to decompose to the alkyl
N
chloride. The idea is illustrated in Scheme 7 for DHA wherein the DHA-sulfurane adduct 21 may equilibrate with
the oxonium-stabilized cation 15, chloride ion and DMSO, the last which will thus cycle in a catalytic fashion.

11
ACCEPTED MANUSCRIPT
Scheme 7. Proposal for conversion of DHA 2 into the β-chloride 19 with oxalyl chloride and DMSO; the reaction
is catalytic with respect to DMSO. The sulfurane 21 will be generated as the α-epimer as it is formed by nucleo-
philic attack by the α-epimer of DHA with equatorial hydroxyl on the sulfonium salt, as is discussed elsewhere for
reaction of DHA with other electrophiles (ref. 67).
1
The reaction was examined by H NMR spectroscopy with oxalyl chloride, d6-DMSO and DHA in d8-
toluene. Whilst DHA is partially soluble in toluene, a colourless solution corresponding to formation of the β-
chloride 19 formed as soon as the oxalyl chloride was added. At a 1.76 mmole scale, formation of 19 in toluene
was complete in 30 minutes. Treatment with thiomorpholine-triethylamine gave artemiside 7 in 55% isolated yield.
At larger scale, the best conditions involved treating DHA in toluene containing DMSO (0.1 equiv.) with oxalyl
chloride (1 equiv.), then adding thiomorpholine and triethylamine and allowing the reactions to run for 2 h.
Composition of the crude product mixtures isolated from several experiments was assessed by 1H NMR
spectroscopy with 1,3,5-trimethoxybenzene as standard (cf. Table 1); yields ranged from 76-79%. Concentrated
solutions could now be used: DHA (17.7 mmole, 5 g) in 15 mL of toluene, corresponding to a 1.2 M solution, gave
artemiside in 66% isolated yield (Scheme 8). This method is facile, but the industrial scalability needs to be
assessed.
Scheme 8: Fifth process for the preparation of artemiside
Although Swern oxidation of DHA to artemisinin 1 did not take place, there was a slight odour of di-
methyl sulfide associated with the reaction work-up. However, use of the non-volatile p-chlorophenyl sulfoxide as
the activator resulted in a slower reaction, and yields of product were not significantly improved. Dimethyl
formamide may be activated by oxalyl chloride as the Vilsmeier reagent to chlorinate alcohols, but its use here
instead of DMSO did not give any product.
2.2.7. Oxalyl bromide: Although oxalyl bromide is more expensive than oxalyl chloride, it is more reactive; it also
will provide a more economic route to artemiside than the use of TMSBr (cf. Scheme 3). The enhanced reactivity
of oxalyl bromide also means that DMSO may not be required as a promoter in toluene solvent. In the event, at a
1.78 mmole scale, DHA was converted by oxalyl bromide (0.6 equiv.) and thiomorpholine-triethylamine into

12
ACCEPTED MANUSCRIPT
artemiside in 69% isolated yield. Clearly, HBr formed in the reaction of DHA with oxalyl bromide is also able to
convert DHA into the β-bromide. This was verified by treating DHA in d8-toluene with HBr gas, and examining
1
the reaction mixture by H NMR spectroscopy. Thus HBr in contrast to HCl works without any metal halide
promoters in toluene. The reaction succeeded well with a concentration of DHA of approximately 0.4 M (Scheme
9).
Scheme 9: Sixth process for the preparation of artemiside
2.2.8. Thionyl chloride-dimethyl sulfoxide: Thionyl chloride efficiently converts DMSO into the chlorosulfonium
intermediate 20 wherein it is a more effective activator of DMSO than is oxalyl chloride in the Swern oxidation
(Scheme 10).83 Although the caveat is the production of sulfur dioxide, the reaction is more atom-economical than
with oxalyl chloride.
H
O
O
ꢀ
Clꢀ
O
S
O
H
SO2
O
S
O
+
Cl
Scheme 7
Cl
S
Cl
Cl
O
CH3
CH3
O
21
+
15
+
19
O
S
+
Cl
+
S
Oꢀ
+
S
DHA 2
CH3
CH3
Cl
CH3
CH3
CH3
Cl
+
CH3
O
Oꢀ
20
S
S
H
HCl
+
CH3
Clꢀ
Clꢀ
S
CH3
Clꢀ
CH3
+
CH3
Scheme 10. Proposal for thionyl chloride activation by DMSO (cf. Scheme 7). As in the case of oxalyl chloride,
the reaction is catalytic with respect to DMSO, but here SO2 rather than carbon oxides are the gaseous by-products.
DHA 2 at a 1.76 mmol scale and DMSO (0.2 equiv.) in toluene was treated with thionyl chloride (1
equiv.) and after 60 min, with thiomorpholine-triethylamine according to the foregoing conditions. Examination of
the product mixture by 1H NMR spectroscopy indicated the presence of artemiside (54-75%) and variable amounts
of glycal (15-27%). For larger scale experiments (to 14.1 mmole or 4 g of DHA), the best conditions involved
treating DHA in toluene containing DMSO (0.2 equiv.) with thionyl chloride (1 equiv.), then adding thiomorphol-
ine and triethylamine and allowing the reactions to run for 3 h. Direct recrystallization of the product mixture from
methanol provided artemiside (46%), although additional material was able to be recovered by fractional
crystallization of mother liquors. As in the TMSCl-DMSO reaction above (Scheme 6), the glycal 16 interfered
with isolation of the artemiside. It is of note that a control reaction using thionyl chloride in toluene in the absence
of DMSO over 2 h returned unchanged DHA (92%). Although product yields of the thionyl chloride-DMSO
1
reactions as determined by H NMR spectroscopy and by direct isolation are currently lower than those from the
oxalyl chloride-DMSO reactions, the reactions are easily carried out.

13
ACCEPTED MANUSCRIPT
Scheme 11: Seventh process for the preparation of artemiside
The process as encapsulated in Scheme 11 potentially represents the most economical approach to
artemiside. It is clear here and in the above cases that there is a need to suppress formation of the glycal in order to
enhance yields and facilitate isolation of the artemiside.
2.3. Conversion of artemiside to artemisone
We previously used tetrapropylammonium perruthenate-N-methylmorpholine-N-oxide (TPAP) for con-
version of artemiside 7 into artemisone 8.37 However, metal-free oxidation is preferable from a drug development
viewpoint. At a 1.36 mmol scale (500 mg artemiside), use of the Payne system,84 namely 30% aqueous H2O2-
acetonitrile in a mixed methanol-ether or methanol-dichloromethane solvent system containing potassium
carbonate at 0 °C was highly effective in providing artemisone in 89% yield (Scheme 12). Use of other alcohols as
co-solvents – ethanol, isopropanol – resulted in lower yields. The H2O2-urea complex, a convenient solid source of
H2O2 that efficiently oxidizes sulfides to sulfones,85 under the foregoing conditions also provided comparable
yields of artemisone from artemiside.
Scheme 12: Oxidation of artemiside to artemisone.
The latter reagent was used to treat crude product mixtures obtained from the DHA-oxalyl bromide-
thiomorpholine reactions described above (Scheme 9), whence at a 35 mmole scale (10 g DHA), crystalline
artemisone was obtained in 48% yield from DHA.
2.4. Preparation of Artemisone from DHA
Reaction of activated DHA 2 with thiomorpholine-1,1-dioxide was now examined. The latter, in contrast
to thiomorpholine, is a solid whose poor solubility in non-polar solvents requires adjustment of the above
procedures. Two solvents were used to dissolve the dioxide – dichloromethane and acetone. The preparative
methods focussed on oxalyl chloride and oxalyl bromide, as each reagent can be used in relatively concentrated
reaction mixtures containing DHA. At a 17.7 mmole scale (5 g), DHA on treatment with oxalyl chloride-DMSO
(0.1 equiv.) in toluene followed by treatment with thiomorpholine-1,1-dioxide and triethylamine in dichlorometh-
1
ane gave artemisone (42% isolated yield) (Scheme 13). As above, product yields were also estimated by the H
NMR method with 1,3,5-trimethoxybenzene as standard; at a 1.77 mmole scale, yields of artemisone ranged from
64-70%. Reactions using acetone to dissolve the 1,1-dioxide provided artemisone in similar yields; however, the
formation of a minor amount of an unidentified by-product derived from the DHA was evident. Surprisingly,
when oxalyl bromide (0.6 equiv.) in dichloromethane or toluene was used to activate DHA followed by treatment
with the 1,1-dioxide and triethylamine in dichloromethane, the yields from reactions conducted at a 1.77 mmole
scale were lower (to 50%).

14
ACCEPTED MANUSCRIPT
Scheme 13: Direct preparation of artemisone from DHA with oxalyl chloride-DMSO.
2.5. Preparation of Artemiside from Anhydrodihydroartemisinin (glycal)
The crystalline glycal 16 that is readily obtained from DHA 2 by treating with BF3 etherate or other
dehydrating agent467 inevitably appears among the reaction products here; thus by converting this by-product into
artemiside or artemisone on an industrial scale, the effective overall consumption of DHA will be enhanced. Acid-
catalysed methods can be used to add nucleophiles to the C10 position via protonation of the double bond at C9 to
produce the oxonium ion intermediate 15.86 Si-face addition of the proton at C9 is favoured because of the steric
hindrance of the adjacent pseudoaxial C8-C8a bond on the Re-face (Schemes 2, 14). We have already converted
isotopically-labelled glycal via addition of HCl and treatment with thiomorpholine into labelled artemiside, and
1
thence, the corresponding labelled artemisone wherein the β-chloride is detected as the intermediate by H NMR
spectroscopy.37 However, we re-evaluated the procedure here by using the HCl conditions described above. In the
event, treatment of glycal 16 with gaseous HCl followed by thiomorpholine-triethylamine provided artemiside
(59% isolated yield). Notably, use of thiomorpholine-1,1-dioxide under the foregoing conditions gave a
1
comparatively much lower yield (21%) of artemisone as estimated by H NMR spectroscopy, indicating that
thiomorpholine S,S-dioxide is a poorer nucleophile than the thiomorpholine.
Scheme 14: Addition of HCl to glycal 16 to provide β-chloride 19 and conversion into artemiside 7.
3. Conclusion
We have evaluated various processes for N-glycosylation of DHA 2 involving conversion into artemiside 7
and thence into the lead compound artemisone 8. The use of gaseous HCl coupled with anhydrous Group I or II
metal halide salts in toluene represent a particularly economical approach for converting DHA via the β-chloride
into artemiside 7, although relatively dilute solutions of DHA (to 0.3M) are required for optimum yields. The
oxalyl chloride process involving DMSO as promoter in toluene appears to be the best at this stage. Although DHA
is a lactol, it could in principle be oxidized to artemisinin via the Swern reaction under these conditions; however
such a process does not compete. The benefit also is that more concentrated solutions (to 0.9M DHA) can now be
used without diminution in yields. Whilst oxalyl bromide works well for preparation of artemiside 7 from DHA in
toluene in the absence of promoter, use of the reagent in conjunction with thiomorpholine-1,1-dioxide for direct
preparation of artemisone 8 is not as successful. Although the dioxide is insoluble in toluene, it could be added to
the toluene reaction mixtures as a solution in dichloromethane or acetone. However, in a direct comparison, it is
noted to react less efficiently with the β-chloride 19 in dichloromethane than does thiomorpholine itself. The
process for obtaining artemisone using oxalyl bromide-thiomorpholine, and then oxidation of the crude product

15
ACCEPTED MANUSCRIPT
mixture with urea-hydrogen peroxide directly provides artemisone in 48% yield at a multigram scale from DHA.
The reagents used, oxalyl bromide and urea-hydrogen peroxide, are easy to handle and no halogenated solvent is
used through the whole process. Extractive work up and chromatography are not required, and the product is easily
crystallized from isopropanol.
4. Experimental Section
4.1. General
DHA was obtained from the Kunming Pharmaceutical Corporation, Kunming, Yunnan Province, China,
or from the Dang Quang Trading Company, Hanoi, Vietnam. Dichloromethane was distilled from and stored over
calcium hydride, triethylamine and thiomorpholine were distilled from calcium hydride and stored over potassium
hydroxide, toluene was dried over sodium wire and distilled from sodium benzophenone ketyl immediately prior to
use. Dichloromethane was distilled from calcium hydride immediately prior to use. Thiomorpholine-1,1-dioxide
was provided by Bayer AG Zentralforschung, and used as such. Acetone, AR grade (Lab-Scan), oxalyl chloride
(98%) and oxalyl bromide (97%), and urea-hydrogen peroxide (98%) (Sigma-Aldrich) were used as received.
Dimethyl sulfoxide (BDH) was dried and fractionally distilled from calcium hydride, and stored over 4 Å
molecular sieves under nitrogen. TLC was performed with Merck Kieselgel 60 F254 plates and visualised with
ultra violet light (254 nm) and/or heating after treatment with 5% ammonium molybdate in 10% concentrated
sulfuric acid. Column chromatography was performed with Merck silica gel 60 (0.04-0.063 mm).
1H spectra were recorded on a Varian Mercury-300 spectrometer. Chemical shifts are in ppm relative to
tetramethylsilane (0.03% v/v) as 0.0 ppm for proton and CHCl3 in CDCl3 as 7.26 ppm. Melting points were carried
out on a Leica Hot Stage DME E microscope and are corrected. Mass spectral data were obtained on Finnigan
TSQ 7000 mass spectrometer (CI) and on an API QSTAR high performance triple quadrupole time-of-flight mass
spectrometer with electrospray ionization. Infrared spectra were recorded either on a Perkin Elmer PC 16 or a
1
Perkin Elmer Spectrum One spectrometer. Yield readouts from H NMR spectra were determined by adding the
internal standard crystalline 1,3,5-trimethoxybenzene (0.33 equiv. with respect to DHA) to the crude product
mixture. This has a singlet signal from the three aromatic protons at 6.08 ppm. The yields of product were
obtained by comparing the integral of this signal with the integral of the signal due to H-12 in the products (cf.
Table 1).
Hydrogen chloride gas was generated by dropwise addition of concentrated H2SO4 onto NH4Cl powder at a
rate of 1 drop per 5 s with a flow rate of ca. 5.5 mmol per min. The cannulas used for transfer of reaction mixtures
were 30.5 cm (12") long with ID 0.8 mm, OD 1.0 mm for reactions with 500 mg DHA, and 43.2 cm (17") long
with ID 1.2 mm, OD 1.5 mm for 5 g or larger scale reactions. Magnetic stirrer bars were 2 x 1 cm for 500 mg
reactions and 2.5 x 1.3 cm for 5 g reactions of DHA. The stirring velocity was from 650 to 750 min-1. Nitrogen
flushing required in the first step to assist in removal of excess of hydrogen chloride was effected by passing dry
nitrogen gas over the reaction mixture.
4.2. Preparation of Artemiside 7
Data for artemiside and procedures for its preparation via the first process with DHA TMS ether and TMSBr
(Scheme 4), and the second process involving generation of TMSBr in situ (Scheme 5) are given in the Supporting
Information 1 associated with ref. 37.
Attempted preparation from DHA by the Kotchetkov Reaction: A solution of thiomorpholine (0.12 mL, 1.2 mmol)
and dihydroartemisinin (DHA, 285.6 mg, 1.01 mmol) in ethanol (20 mL) under nitrogen was heated under reflux
for 20 h. Half of the ethanol was then evaporated, and saturated aqueous ammonium chloride (5 mL) was added

16
ACCEPTED MANUSCRIPT
followed by evaporation of remaining solvent under reduced pressure. The residue was triturated with
dichloromethane, filtered and the filtrate was washed with brine (5 mL). The organic layer was separated and
dried (MgSO4), and after filtration, the solvent was evaporated under reduced pressure to leave a residue that was
purified by column chromatography (25:75 ethyl acetate-hexane) to give deoxyartemisinin 6 (179 mg, 67%). This
was recrystallized from ethyl acetate–hexane to give 6 as colourless rectangular plates, m.p. 114–115.8 °C, δH
0.95-0.97 (3H, d, J = 5.0 Hz, 6 Me), 1.20-1.23 (3H, d, J = 2.4 Hz, 9 Me), 1.27-1.29 (2H, m), 1.54-1.65 (7H, m),
1.76-2.06 (5H, m), 3.18-3.24 (1H, m), 5.70 (1H, s, H-12); m/z (%) 151.2 (13) 164.2 (30), 193.2 (16), 203.2 (34),
222.3 (21), 267.2 (100); HRMS (ESI) calcd. 267.1596 [M++1], found 267.1605. This sample was identical with
that of deoxyartemisinin obtained by catalytic hydrogenation of artemisinin.4 Repetition of the procedure with each
of methanol, ammonium chloride (0.16 g 3 mmol) in 4:1 ethanol-water, and MgCl2 (0.1 equiv) in ethylene glycol
at 40 °C provided 2-deoxyartemisinin 6 as the only identifiable product in varying yields.
4.2.1. From DHA, HCl and Group I and II metal halides
Dichloromethane: HCl gas was passed over a stirred mixture of DHA 2 (5.0 g, 17.6 mmol) and anhydrous LiCl
(1.2 g, 28.3 mmol) in dichloromethane (80 mL) at 0 °C. The colour of the reaction mixture changed from
colourless to pale purple. The system was gently flushed by passage of nitrogen gas over the reaction mixture for 5
min. After gentle degassing further under reduced pressure with alternate nitrogen flushing, dichloromethane (20
mL) was added. Then, the mixture was treated dropwise with a neat mixture of thiomorpholine (2.0 mL, 2.1 g,
19.9 mmol) and triethylamine (2.5 mL, 1.8 g, 18.0 mmol). The colour of the reaction mixture changed to pale
yellow. After 45 min., the reaction mixture was quenched with brine (50 mL). The organic layer was separated,
and the aqueous layer was extracted with dichloromethane (2 x 20 mL). The combined organic extracts were dried
(MgSO4). Filtration and evaporation of the filtrate gave a pale yellow solid, which was recrystallized directly from
methanol to give artemiside 7 as a fine white needles (3.58 g, 55%), mp 147.0-147.6 °C (lit.37 152.5-153 °C).
Further material was able to be recovered by fractional crystallization from the mother liquors.
The reaction was repeated according to the conditions above with DHA 2 (5.0 g, 17.6 mmol) and each of
anhydrous MgCl2 (1.90 g, 20.0 mmol) and CaCl2 (2.2 g, 19.8 mmol) followed by treatment with the neat mixture
of thiomorpholine and triethylamine as above; quenching with brine followed by workup and recrystallization gave
artemiside 7 (3.60 g, 55% and 3.49 g, 54%, respectively).
The following procedure omitted the quenching step. HCl gas was passed over a stirred mixture of DHA 2
(5.0 g, 17.6 mmol) and anhydrous MgCl2 (0.6 g, 6.30 mmol) in dichloromethane (80 mL) at 0 °C. The colour of
the reaction mixture became pale purple. After 10 min., ingress of HCl was halted, and the reaction vessel was
flushed by gentle passage of nitrogen gas over the reaction mixture at 0 °C for 5 min. After gentle degassing
further under reduced pressure with alternate nitrogen flushing, dichloromethane (20 mL) was added to dilute the
reaction mixture. The mixture was treated with a neat mixture of thiomorpholine (2.0 mL, 2.1 g, 19.9 mmol) and
triethylamine (2.5 mL, 1.8 g, 18.0 mmol). The colour of the reaction mixture changed to pale yellow. After 45
min, the reaction mixture was filtered directly through a sintered glass funnel. The residue in the funnel was
washed with dichloromethane (2 x10 mL), and the washings and filtrate were combined. The solution was
evaporated under reduced pressure to leave a pale yellow solid, which was recrystallized from methanol to give
artemiside (3.50 g, 54%). The glycal 16 and additional artemiside 7 were recovered from the mother liquors after
chromatography over silica gel, and eluting with ethyl acetate-hexane (8:92).
Toluene: HCl gas was passed over a stirred mixture of DHA 2 (5.0 g, 17.6 mmol) and MgBr2 (1.65 g, 8.96 mmol)
in toluene (80 mL) at 0 °C. The colour of the reaction mixture became dark blue. After 10 min., ingress of HCl
was halted, and the system was flushed by passage of nitrogen gas over the reaction mixture for 15 min. After

17
ACCEPTED MANUSCRIPT
gently outgassing under reduced pressure with alternate nitrogen flushing, the reaction mixture was treated with
slow dropwise addition of a neat mixture of thiomorpholine (2.0 mL, 2.1 g, 19.9 mmol) and triethylamine (2.5 mL,
1.8 g, 18.0 mmol). The colour of the reaction mixture changed to pale yellow with concomitant formation of a
white bulky precipitate. Dichloromethane (20 mL) had to be added to the resulting very thick mixture so as to
facilitate stirring. After 45 min., the reaction mixture was quenched with brine (50 mL). The organic layer was
separated, and the aqueous layer was extracted with dichloromethane (2 x 20 mL). The combined organic extracts
were dried (Na2SO4). Filtration and evaporation of the filtrate gave a pale yellow solid, which was recrystallized
directly from methanol to give artemiside (3.26 g, 50%). More artemiside was recovered from the mother liquors
by evaporation and chromatography over silica gel with chromatography (8:92 ethyl acetate-hexane).
4.2.2. From DHA, trimethylsilyl chloride and DMSO
Trimethylsilyl chloride (4.50 mL, 3.85 g, 35.46 mmol) was added to a stirred solution of DHA 2 (5.0 g, 17.6
mmol) in dichloromethane (80 mL) under nitrogen at 0 ºC. After 15 min., DMSO (0.315 mL, 0.345 g, 4.42 mmol)
was added. The colour of the reaction mixture changed to pink and then to pale purple after 1.5 h. The mixture
was treated by slow dropwise addition of a neat mixture of thiomorpholine (2.0 mL, 2.1 g, 19.9 mmol) and
triethylamine (2.5 mL, 1.8 g, 18.0 mmol), whereupon the colour changed to pale yellow. After 45 min., the
reaction mixture was quenched with brine (50 mL). The organic layer was separated, and the aqueous layer was
extracted with dichloromethane (2 x 20 mL). The combined organic extracts were dried (Na2SO4). Filtration and
evaporation of the filtrate under reduced pressure left a pale yellow solid, which was recrystallized from methanol
to give artemiside 7 (3.0 g, 46%). More artemiside was recovered from the mother liquors by evaporation and
chromatography over silica gel (8:92 ethyl acetate-hexane).
4.2.3. From DHA and oxalyl chloride
Dichloromethane: A solution of DHA 2 (504.3 mg, 1.78 mmol) and oxalyl chloride (0.16 mL, 1.76 mmol) in
dichloromethane (8 mL) was stirred under nitrogen at 0 °C for 4 h. Thereafter, a solution of thiomorpholine (0.2
mL, 1.99 mmol) and triethylamine (0.25 mL, 1.8 mmol) in dichloromethane (1 mL) was transferred by cannula
into the reaction mixture, and the resulting mixture was stirred at room temperature for another 2 h. This was then
quenched with saturated aqueous NaHCO3 (5 mL), the organic layer was separated and the aqueous layer was
extracted with dichloromethane (3 x 5 mL). The combined organic layer was washed with water (2 x 5 mL) and
brine (2 x 5 mL), and then dried (MgSO4) and filtered, and the filtrate evaporated. The yellow residue was purified
by column chromatography (8:92 ethyl acetate-hexane) to give artemiside 7 as a white crystalline solid (310.6 mg,
48%).
Toluene: A solution of DHA 2 (497 mg, 1.75 mmol) and oxalyl chloride (0.16 mL, 1.76 mmol) in toluene (8 mL)
was stirred under a gentle stream of nitrogen at room temperature for 4 h. Thereafter, a solution of thiomorpholine
(0.2 mL, 1.99 mmol) and triethylamine (0.25 mL, 1.8 mmol) in toluene (1 mL) was added and the reaction mixture
was stirred for another 1.75 h. The mixture was quenched with saturated aqueous NaHCO3 (5 mL) and processed
as described above, to give after column chromatography (8:92 ethyl acetate-hexane) artemiside 7 as a white
crystalline solid (247.5 mg, 39%).
DHA, oxalyl chloride and DMSO in toluene: Oxalyl chloride (1.6 mL, 18.3 mmol, 1.03 equiv.) was added to a
stirred mixture of DHA (5.027 g, 17.7 mmol) and dimethyl sulfoxide (DMSO, 0.13 mL, 1.8 mmol) in toluene (15
mL) in a 100 mL two-necked round bottom flask under gentle nitrogen flush over the reaction mixture at room
temperature. After 30 min, triethylamine (2.5 mL, 18.0 mmol) followed by thiomorpholine (2.0 mL, 21.1 mmol)

18
ACCEPTED MANUSCRIPT
was added to the reaction mixture that was stirred for another 2 h; the nitrogen flush was continued for the first 15
minutes. The mixture was quenched with saturated aqueous NaHCO3 (50 mL). The aqueous layer was extracted
with ethyl acetate (3 x 50 mL). The combined organic layer was washed with water (2 x 50 mL) and then brine (2
x 50 mL) and dried (MgSO4). After filtration and concentration of the filtrate under reduced pressure, direct
crystallization from methanol gave artemiside (4.31 g, 66%), with further amounts being recovered from the
mother liquors. In general practice, however, it was usually necessary to submit the residue obtained by
evaporation of the mother liquors to column chromatography (8:92 ethyl acetate-hexane) to separate the non-polar
glycal 16 from artemiside which could then be recovered. A further two experiments were conducted and 1H NMR
spectroscopy with 1,3,5-trimethoxybenzene as internal standard to measure amount of 7 in the residue obtained
directly after workup indicated the formation of artemiside (72 and 79% respectively, based on reacting DHA),
with small amounts of unreacted DHA 2 (3 and 4%) and glycal 16 (22% for each experiment, based on reacting
DHA) also present.
4.2.4. From DHA and oxalyl bromide
A solution of oxalyl bromide (0.1 mL, 1.033 mmol, 0.6 equiv.) in toluene (1 mL) was transferred
dropwise via cannula under nitrogen into a suspension of DHA 17 (504.2 mg, 1.78 mmol) in toluene (3 mL) at 0
°C under nitrogen. The resulting mixture was stirred at 0 °C with gentle nitrogen flush over the reaction mixture
for 30 minutes. A solution of thiomorpholine (0.2 mL, 1.99 mmol) and triethylamine (0.25 mL, 1.8 mmol) in
toluene (1 mL) was added to the reaction mixture which was allowed to warm to room temperature and stirred for
another 1 h. The mixture was quenched with saturated aqueous NaHCO3 (5 mL), the organic layer was separated
and the aqueous layer was extracted with ethyl acetate (3 x 5 mL). The combined organic layer was washed with
water (2 x 5 mL) and brine (2 x 5 mL), and dried (MgSO4). After filtration, the solvent was evaporated and the
residue crystallized directly from methanol to give artemiside 7 (452.1 mg, 69%). Repetition of the experiment
1
(x7) and using H NMR spectroscopy with 1,3,5-trimethoxybenzene as an internal standard to assess product
mixtures after workup indicated formation of artemiside 7 (71-75%), glycal 16 (27-29% based on reacting DHA),
and DHA (6-8%).
A solution of oxalyl bromide (1 mL, 10.33 mmol, 0.6 equiv.) in toluene (10 mL) was transferred dropwise
via cannula under nitrogen into a suspension of DHA 2 (5.044 g, 17.76 mmol) in toluene (30 mL) at 0 °C under
nitrogen. The mixture was stirred at 0 °C under nitrogen for 30 minutes with occasional gentle outgassing and then
treated via cannula with a solution of thiomorpholine (2 mL, 19.9 mmol) and triethylamine (2.5 mL, 18 mmol) in
toluene (10 mL). The mixture was warmed to room temperature and stirred for another 2 hours, and then processed
1
as described above to provide the product mixture, analysis of which by H NMR spectroscopy with 1,3,5-
trimethoxybenzene as an internal standard indicated formation of artemiside 7 (71%).
4.2.5. From DHA and thionyl chloride-DMSO
DMSO (25 µL, 0.2 equiv) was added into a stirred mixture of DHA 2 (500 mg, 1.76 mmol) in toluene (10
mL) under nitrogen. Thionyl chloride (127.7 µL, 1 equiv) was added slowly into the mixture at 0 °C, and the
resulting solution was stirred for 1 hour at 0 °C, and then gently flushed by passage of nitrogen over the reaction
mixture for 15 minutes. Thereupon, a neat mixture of triethylamine (294.5 µL, 1.2 equiv.) and thiomorpholine
(212.5 µL, 1.2 equiv.) was added slowly into the reaction mixture that was then stirred for 2 hours at 0 °C. Before
workup, the standard 1,3,5-trimethoxybenzene was added and the reaction mixture was quenched by addition of
saturated aqueous NaHCO3 (30 mL) and extracted with ethyl acetate (3 x 15 mL). The extracts were combined and
the organic layer was washed with brine (30 mL), and then dried (MgSO4) and filtered. The filtrate was evaporated
1
under reduced pressure to give the crude product mixture, analysis of which by H NMR spectroscopy with the

19
ACCEPTED MANUSCRIPT
internal standard 1,3,5-trimethoxybenzene indicated all DHA had been consumed, and that artemiside 7 (75%) and
glycal 16 (16%) were present.
For an example of a larger scale preparative reaction, thionyl chloride (1.022 mL, 1 equiv) was added
slowly to a stirred mixture of DHA (4.0 g, 14.085 mmol) in toluene (20 mL) containing DMSO (200 µL, 0.2 equiv)
at 0 °C. The resulting solution was stirred for 1.3 hour at 0 °C, and then gently flushed with argon for 20 minutes.
A 1:1 mixture of triethylamine (2.356 mL) and thiomorpholine (1.70 mL) (each 1.2 equiv) was added slowly into
the reaction mixture that was then stirred for 3 hours at 0 °C. Just before workup, a known amount of the internal
standard 1,3,5-trimethoxybenzene was added into the reaction mixture that was quenched by addition of saturated
aqueous NaHCO3 (180 mL). The mixture was extracted with ethyl acetate (3 x 30 mL) and the combined organic
layer was washed with saturated aqueous NaHCO3 (60 mL), then with brine (60 mL) and dried (MgSO4). The
filtrate was evaporated under reduced pressure to give the crude product that was crystallized directly from
methanol to give artemiside 7 (2.39 g, 46%). More artemiside was recovered from the first mother liquors by
evaporation and fractional crystallization, and then by chromatography of the final residue over silica gel (8:92
ethyl acetate-hexane).
5. Preparation of Artemisone 8
Data for artemisone and procedures for its preparation via oxidation of artemiside 7 with tetrapropylammon-
ium perruthenate-N-methylmorpholine-N-oxide are given in the Supporting Information 1 associated with ref. 37.
5.1. By oxidation of artemiside
30% H2O2-acetonitrile: Acetonitrile (0.21 mL, 4.07 mmol, 2.3 equiv.), and potassium carbonate (0.272 g, 1.96
mmol, 1.11 equiv.) were added to a stirred solution of artemiside 7 (500.0 mg, 1.36 mmol) in methanol (10 mL)
and dichloromethane (2 mL). The resulting mixture was cooled to 0 °C, and aqueous hydrogen peroxide (30%,
0.46 mL, 4.07 mmol, 2.3 equiv.) in methanol (2 mL) was added dropwise. After completion of the addition, the
reaction mixture was allowed to warm to room temperature and stirred for 21 hours. It was then filtered through a
filter paper. The filter paper was rinsed with dichloromethane (20 mL) with washings passing into the filtrate. The
filtrate was treated with brine (10 mL). The organic layer was separated, and the aqueous layer was extracted with
dichloromethane (3 x 20 mL). The combined organic layer was dried (MgSO4), filtered and evaporated to dryness
to leave a white crystalline residue that was essentially pure artemisone 8 (531.4 mg, 98%). The residue was
recrystallized from isopropanol to give artemisone (482.6 mg, 89%), as needles, m.p. 152-153 °C (lit.37 152.3-
152.7 °C),
Urea-H2O2-complex: Acetonitrile (0.12 µL, 2.3257 mmol, 6.6 equiv.) was added to a stirred solution of artemiside
7 (128.0 mg, 0.355 mmol), 98% urea-hydrogen peroxide (67.6 mg, 0.704 mmol, 2 equiv.) and potassium carbonate
(70.3 mg, 0.5069 mmol, 1.44 equiv.) in methanol (2 mL) at 0 °C. The mixture was allowed to warm up to room
temperature and stirred for 21 hours. The solvent was evaporated to leave a solid residue that was triturated
thoroughly with distilled water. The undissolved solid was collected by filtration and dried under high vacuum;
thereby was obtained artemisone 8 (131 mg, 94%).
5.2. By direct oxidation of the crude residue from the DHA-thiomorpholine-oxalyl bromide reaction
A solution of oxalyl bromide (2 mL, 20.66 mmol, 0.6 equiv,) in toluene (20 mL) was added dropwise via
cannula to a stirred mixture of DHA 2 (10.002 g, 35.2 mmol) in toluene (60 mL) at 0 °C under nitrogen. The
resulting solution was stirred at 0 °C under nitrogen for 30 minutes with gentle outgassing from time to time.
Thereafter, a solution of thiomorpholine (4 mL, 39.8 mmol, 1.13 equiv) and triethylamine (1.06 equiv, 5 mL, 36

20
ACCEPTED MANUSCRIPT
mmol) in toluene (1 mL) was added, and the reaction mixture was warmed up to room temperature and stirred for
another 2 hours. The precipitate was collected by filtration and washed with ethyl acetate into the filtrate. The
filtrate was evaporated under reduced pressure to leave a yellow residue that was treated sequentially with
methanol (160 mL), acetonitrile (11.8 mL, 232.3 mmol, 6.6 equiv), urea-hydrogen peroxide (6.7577 g, 70.4 mmol,
2 equiv.) and potassium carbonate (7.001 g, 50.69 mmol, 1.44 equiv.) at 0 °C. The resulting mixture was stirred
with warming to room temperature for 21 hours. The solvent was evaporated to leave a residue that was treated
with water as described above to leave a crystalline solid. This was collected by filtration, dried under vacuum and
recrystallized from isopropanol to give artemisone 8 (6.70 g, 48%).
5.3. From DHA and thiomorpholine-1,1-dioxide
Oxalyl chloride in toluene: Oxalyl chloride (1.8 mL, 20.6 mmol, 1.16 equiv.) was added to a stirred mixture of
DHA (5.027 g, 17.7 mmol) and DMSO (0.13 mL, 1.8 mmol) in toluene (50 mL) under gentle nitrogen flush at
room temperature. After 30 min, a solution of triethylamine (2.53 mL, 18.18 mmol, 1.01 equiv.) and
thiomorpholine-1,1-dioxide (2.84 mg, 21.1 mmol, 1.19 equiv.) in dichloromethane (5 mL) was transferred
dropwise via cannula into the mixture. This was stirred for another 105 min with the gentle nitrogen flush
continuing for the first 45 minutes. The reaction mixture was quenched with saturated aqueous NaHCO3 (50 mL),
and the aqueous layer was extracted with ethyl acetate (3 x 50 mL). The combined organic layer was washed with
water (2 x 50 mL) and then brine (2 x 50 mL) and dried (MgSO4). After filtration and concentration of the filtrate
1
under reduced pressure, analysis of the residue by H NMR spectroscopy with 1,3,5-trimethoxybenze as internal
standard indicated formation of artemisone 8 (70% based on reacting DHA) and the presence of DHA (3%) and
glycal 16 (20% based on reacting DHA). Crystallization of the residue from isopropanol provided crystalline
artemisone (2.98 g, 42%). Repetition of the reaction by using acetone (5 mL) instead of dichloromethane to
dissolve the thiomorpholine-1,1-dioxide provided similar yields of artemisone with the evident formation of a
small amount of an additional by-product derived from the DHA.
Oxalyl bromide in dichloromethane: A solution of oxalyl bromide (0.1 mL, 1.033 mmol, 0.6 equiv.) in
dichloromethane (1 mL) was added via cannula under nitrogen to a solution of DHA 2 (503.4 mg, 1.77 mmol) in
dichloromethane (1.5 mL) at 0 °C. The reaction mixture was stirred at 0 °C under nitrogen for 30 minutes with two
cycles of gentle outgassing and nitrogen flush over the mixture. Thereupon a solution of thiomorpholine-1,1-
dioxide (0.27 g, 1.99 mmol, 1.13 equiv.) and triethylamine (0.25 mL, 1.8 mmol, 1.06 equiv.) in dichloromethane
(0.5 mL) was added to the reaction mixture, and the reaction mixture was warmed up to room temperature and
stirred for another 4 hours. Saturated aqueous NaHCO3 (5 mL) was added to quench the reaction. The organic
layer was separated and the aqueous layer was extracted with dichloromethane (3 x 5 mL). The combined organic
layer was washed with water (2 x 5 mL) and brine (2 x 5 mL). The solution was dried (MgSO4) and filtered. The
solvent was evaporated. The yield of artemisone 56 (39%) was determined by NMR spectroscopy with 1,3,5-
trimethoxybenzene as an internal standard.
Oxalyl bromide in toluene: A solution of oxalyl bromide (0.1 mL, 1.033 mmol, 0.6 equiv.) in toluene (0.5 mL) was
added dropwise via cannula to a stirred suspension of DHA (501.0 mg, 1.76 mmol) in toluene (1.5 mL) at 0 °C
under nitrogen. The reaction mixture was stirred at 0 °C under nitrogen for 30 minutes with a cycle of gentle
outgassing. Thereupon a solution of thiomorpholine-1,1-dioxide (0.27 g, 1.99 mmol, 1.13 equiv.) and
triethylamine (0.25 mL, 1.8 mmol, 1.06 equiv) in dichloromethane (0.5 mL) was added dropwise via cannula into
the reaction mixture. The reaction mixture was warmed to room temperature with stirring for another 3 h.
Saturated aqueous NaHCO3 (5 mL) was added to quench the reaction. The organic layer was separated and the

21
ACCEPTED MANUSCRIPT
aqueous layer was extracted with ethyl acetate (3 x 5 mL). The combined organic layer was washed with water (2
x 5 mL) and brine (2 x 5 mL). The solution was dried (MgSO4) and filtered. The solvent was evaporated. The
1
yield of artemisone 56 (50%) was determined by H NMR spectroscopy with 1,3,5-trimethoxybenzene as an
internal standard.
5.4. Preparation of artemiside 7 from glycal 16.
HCl gas was passed over a stirred mixture of glycal 16 (241 mg, 0.91 mmol) in dichloromethane (5 mL)
at 0 °C. After 10 minutes, the system was flushed with nitrogen for 15 min. Then, under an atmosphere of
nitrogen, a mixture of thiomorpholine (0.10 mL, 1.0 mmol) and triethylamine (0.125 mL, 1.0 mmol) in
dichloromethane (1.3 mL) was added. After stirring 1 hour at room temperature, the reaction mixture was
quenched with brine (5 mL). The organic layer was separated and the aqueous layer was extracted with
dichloromethane (3 x 5 mL). The combined organic layer was dried (MgSO4) and then filtered. The solvent was
1
evaporated to give a pale yellow residue. The yield (74%) of artemiside 7 was determined by H NMR
spectroscopy with 1,3,5-trimethoxybenzene as an internal standard; recrystallization of the residue directly from
methanol provided artemiside 7 (197 mg, 59%).
5.5. Preparation of artemisone 8 from glycal 16.
HCl gas was passed over a stirred mixture of glycal 16 (94.4 mg, 0.355 mmol) in dichloromethane (2 mL) at
0 °C. After 10 minutes, the system was flushed with nitrogen for 15 minutes. Then, under an atmosphere of
nitrogen, a solution of thiomorpholine-1,1-dioxide (57.3 mg, 0.398 mmol) and triethylamine (0.05 mL, 0.36 mmol)
in dichloromethane (0.5 mL) was added. After stirring 4 hours at room temperature, the reaction mixture was
quenched with brine (2 mL). The organic layer was separated and the aqueous layer was extracted with
dichloromethane (3 x 2 mL). The combined organic layer was dried (MgSO4) and then filtered. The solvent was
evaporated to give a yellow residue. The yield of artemisone 8 (21%) was determined by NMR spectroscopy with
1,3,5-trimethoxybenzene as an internal standard; .
Acknowledgments: This project is funded in South Africa by the South African Medical Research Council (MRC)
with funds from the National Treasury under its Economic Competitiveness and Support Package. The South
African National Research Foundation is thanked for financial support to RKH (NRF UIDs 90682 and 98934).
RKH also thanks the North-West University for financial support. Work at HKUST was originally supported
through the HKSAR University Grants Committee Areas of Excellence Fund, the HKSAR Research Grants
Council through the Competitive Earmarked Research Grants Scheme, and Bayer AG, Leverkusen, Germany.
Disclaimer: Any opinion, finding and conclusion or recommendation expressed in this material is that of the
authors and the South African NRF does not accept any liability in this regard.
References
1
2
Zhang, J. F. A Detailed Chronological Record of Project 523 and the Discovery and Development of
Qinghaosu (Artemisinin). Yang Cheng Evening News Publishing Company, Yang Cheng, China, 2005;
Artemisinin-Based and Other Antimalarials: Detailed Account of Studies by Chinese Scientists Who
Discovered and Developed Them. Li Guoqiao, Li Ying, Li Zelin, Zeng Meiyi, Eds Chinese Edition; Arnold,
M., Arnold, K., Zeng Meiyi, Eds. English Edition; Academic Press, Elsevier: London; ISBN: 978-0-12-
813133-6; 2017.
Li Y, Wu YL. Curr. Med. Chem. 2003; 10:2197.

22
ACCEPTED MANUSCRIPT
3
4
Cui L, Su XZ. Expert Rev. Anti Infect. Ther. 2009; 7:999.
Co-operative Research Group on Qinghaosu. A Novel Type of Sesquiterpene Lactone – Qinghaosu. Ke Xue
Tong Bao 1977; 22:142; Liu JM, Ni MY, Fan JF, Tu YY, Wu ZH, Wu YL, Zhou WS. Acta Chimica Sinica
1979; 37:129; Li Y, Yu PL, Chen YX, Li LQ, Gai YZ, Wang DS, Zheng YP. Ke Xue Tong Bao 1979; 24:
667; Li Y, Yu PL, Chen YX, Li LQ, Gai YZ, Wang DS, Zheng YP. Acta Pharmaceutica Sinica 1981;
16:429; China Co-operative Research Group on Qinghaosu and its Derivatives as Antimalarials. The
chemistry and synthesis of qinghaosu derivatives. J. Tradit. Chin. Med. 1982; 2:9; Wu YL, Li Y. Med.
Chem. Rev. 1995; 5:569.
5
Liu X. Yao Xue Tong Bao 1980; 15:183; Liu X. Antimalarial Reduced Artemisinin Succinate. Faming
Zhuanli Shenqing Gongkai Shuomingshu CN 85100781 A 20 Aug 1986 (Chem. Abstr. 1987, 107, 78111j);
Li L, Dong YC, Zhu NJ. Jiegou Huaxue 1986; 5:73 (Chem. Abstr. 1987, 107, 236339).
Li Y, Yu PL, Chen YX, Ji RY. Acta Chimica Sinica 1982; 40:557.
6
7
Whilst the discovery of artemisinin resulted in the award in 2015 of the Nobel Prize for Physiology or
Medicine to Youyou Tu , the award is best seen as recognition of the achievements of all collaborating
groups under Project 523. Youyou Tu was associated with the Beijing Institute Materia Medica, and certain
descriptions of events in her book (Tu, Y-Y. Qinghao and Qinghaosu Derivative Drugs, Chemical Industry
Publication, Beijing, 2009) and in the publication Tu YY. Southeast Asian J. Trop. Med. Public Health
2004; 35:250 either do not acknowledge or do not correctly assign the responsibilities for the actual
discoveries made by other groups working under the Project 523. The detailed records of those discoveries
are presented in the publications of refs. 1-3.
8
Golenser J, Waknine J, Krugliak M, Hunt NH, Grau GE. Internat. J. Parasitol. 2006; 36:1427; O’Neill PM,
Barton VE, Ward SA. Molecules 2010; 15: 1705; Ding XC, Beck HP, Raso G. Trends Parasitol. 2011;
27:73.
9
Haynes RK, Cheu KW, N'Da D, Coghi P, Monti, D. Infect. Disorders – Drug Targets 2013; 13:217.
Xiao SH. Acta Tropica 2005; 96: 153; Sen R, Bandyopadhyay S, Dutta A, Mandal G, Ganguly S, Saha P,
Chatterjee M. J. Medical Microbiol. 2007; 56:1213; Hencken CP, Jones-Brando L, Bordón C, Stohler R,
Mott BT, Yolken R, Posner GH, Woodard LE. J. Med. Chem. 2010; 53:3594; McFarland MM, Zach SJ,
Wang X-F, Potluri L-P, Neville AJ, Vennerstrom JL, Davisa PH. Antimicrob. Agents Chemother. 2016;
60:7017.
10
11
12
Miller MJ, Walz AJ, Zhu H, Wu C, Moraski G, Möllmann U, Tristani EM, Crumbliss AL, Ferdig MT,
Checkley L, Edwards RL, Boshoff HI. J. Am. Chem. Soc. 2011; 133:2076 Goswami S, Bhakuni RS,
Chinniah A, Pal A, Kar SK, Das PK. Antimicrob. Agents Chemother. 2012; 9:4594; Zheng H, Colvin CJ,
Johnson BK, Kirchhoff PD, Wilson M, Jorgensen-Muga K., Larsen SD, Abramovitch RB. Nature Chem.
Biol. 2017; 13: 218.
Efferth T, Romero MR, Wolf DG, Stamminger T, Marin JJ, Marschall M. Clin. Infect. Dis. 2008; 47: 804;
Shapira MY, Resnick IB, Chou S, Neumann AU, Lurain NS, Stamminger T, Caplan O, Saleh N, Efferth T,
Marschall M, Wolf DG. Clin. Infect. Dis. 2008; 46:1455; Chou S, Marousek G, Auerochs S, Stamminger T,
Milbradt J, Marschall M. Antiviral. Res. 2011; 92; 364; Flobinus A, Taudon N, Desbordes M, Labrosse B,
Simon F, Mazeron MC, Schnepf N. J. Antimicrob. Chemother. 2014; 69: 34; Drouot E, Piret J, Boivin G.
Antivir. Ther. 2016, 21, 535.
13
14
Goswami S, Bhakuni RS, Chinniah A, Pal A, Kar SK, Das PK. Antimicrob. Agents Chemother. 2012; 9:
4594.
Krishna S, Bustamante L, Haynes RK, Staines HM. Trends Pharmacol. Sci. 2008; 29: 520; Nakase I, Lai H,
Singh NP, Sasaki T. Internat. J. Pharmaceutics 2008; 354:28; Crespo-Ortiz MP, Wei. MQ. J. Biomed.
Biotech. 2012, Article ID 247597, doi:10.1155/2012/247597; Efferth T. Seminars Cancer Biol. 2017; 46:65;
Wong YK, Xu C, Kalesh KA, He Y, Lin Q, Wong WS,. Shen HM, Wang J. Med. Res. Rev. 2017; 1-26.
Olliaro PL, Taylor WR. J. Postgrad. Med. 2004; 50:40 ; Eastman RT, Fidock DAH. Nat. Rev. Microbiol.
2009; 7:864 ; Wells TNC, Alonso PL, Gutteridge WE. Nat. Rev. Drug Disc. 2009; 8:879; Angus B. Expert
Rev. Clin. Pharmacol. 2014; 7:299; World Health Organization, World Malaria Report 2016, ISBN 978-92-
4-151171-1.
15

23
ACCEPTED MANUSCRIPT
16
van Agtmael MA, Cheng-Qi S, Qing JX, Mull R, van Boxtel CJ. Int. J. Antimicrob. Agents 1999; 12:151;
Khanh NX, De Vnes PJ, Ha LD, van Boxtel CJ, Koopmans R, Kager PA. Antimicrob. Agents Chemother.
1999; 43:690; Newton P, Suputtamongkol Y, Teja-Isavadharm P, Pukrittayakamee S, Navaratnam V, Bates
I, White N. Antimicrob. Agents Chemother. 2000; 44:972; Gordi T, Xie RJ, Huong NV, Huong DX,
Karlsson MO, Ashton MA. Br. J. Clin. Pharmacol. 2005; 59:189.
17
18
Davis TME, Binh TQ, Ilett KF, Batty KT, Phuöng HL, Chiswell GM, Phuong VDB, Agus C. Antimicrob.
Agents Chemother. 2003; 47:368.
Leskovac V, Theoharides AD. Comp. Biochem. Physiol.1991; 99:383; Lefevre G, Thomsen MS. Clin. Drug
Invest.1999; 18:467; Grace JM, Aguilar AJ, Trotman KM, Peggins JO, Brewer TG. Drug Metab. Dispos.
1998; 26:704; Navaratnam V, Mansor SM, Sit NW, Grace J, Li QG, Olliaro P. Clin. Pharmacokinetics
2000; 39:255; Lefevre G, Carpenter P, Souppart C, Schmidli H, McLean M, Stypinski D. Brit. J. Clin.
Pharmacol. 2002; 54:485.
19
20
Svensson USH, Ashton M. Br. J. Clin. Pharmacol. 1999; 48:528.
Haynes RK, Chan WC, Wong HN, Li KY, Wu WK., Fan KM, Sung HHY, Williams ID, Prosperi S, Melato
S, Coghi P, Monti D. ChemMedChem 2010; 5:1282.
21
22
Baker JK, McChesney JD, Chi HT. Pharmaceutical Res. 1993; 10:662.
Olliaro PL, Nair NK, Haynes RK, Wong HN. Tang MMK, Decosterd LA, Navaratnam V, unpublished
work.
23
24
Haynes RK, Chan HW, Lung CM, Ng NC, Wong HN, Shek LY, Williams ID, Gomes MF, Cartwright A.
ChemMedChem 2007; 2:1448.
Brewer TG, Peggins JO, Grate SJ, Petras JM, Levine BS, Weina PJ, Swearengen J, Heiffer MH, Schuster
BG. Trans. Roy. Soc. Trop. Med. Hyg.1994; 88 (Supp. 1):33; Brewer TG, Grate SJ, Peggins JO, Weina PJ,
Petras JM, Levine BS, Heiffer MH, Schuster BG. Am. J. Trop. Med. Hyg. 1994; 51: 251; Genovese RF,
Petras JM, Brewer TG. Ann. Trop. Med. Parasitol. 1995; 89:447; Genovese RF, Newman DB, Brewer TG.
Pharmacol. Biochem. Behav. 2000; 67:37; Nontprasert A, Pukrittayakamee S, Nosten-Bertrand M,
Vanijanonta S, White NJ. Am. J. Trop. Med. Hyg. 2000; 62:409; Smith SL, Sadler CJ, Dodd CC, Edwards G,
Ward SA, Park BK, McLean WG. Biochem. Pharmacol. 2001; 61:409.
25
26
Schmuck G, Haynes RK. Neurotox. Res. 2000; 2:37; Schmuck G, Roehrdanz E, Haynes RK, Kahl R.
Antimicrob. Agents. Chemother. 2002; 46:821.
Schmuck G, Temerowski M, Haynes RK, Fugmann B. Res. Adv. in Antimicrob. Agents & Chemother. 2003;
3:35.
27
28
Genovese RF, Newman DB. Arch Toxicol. 2008; 82:379.
Van Vugt M, Angus BJ, Price RN, Mann C, Simpson JA, Poletto C, Htoo SE, Looareesuwan S, White NJ,
Nosten FM. Am. J. Trop. Med. Hyg. 2000; 62:65; Kissinger E, Hien TT, Hung NT, Nam ND, Tuyen NL,
Dinh BV, Mann C, Phu NH, Loc PP, Simpson JA, White NJ, Farrar JJ. Am. J. Trop. Med. Hyg. 2000; 63:48;
Hien TT, Turner GDH, Mai NTH, Phu NH, Bethell D, Blakemore WF, Cavanagh JB, Dayan A, Medana I,
Weller RO, Day NPJ, White NJ. Lancet 2003; 362:295.
29
30
Toovey S, Jamieson A. Trans. R. Soc. Trop. Med. Hyg. 2004; 98:261; Comments - Trans. R. Soc. Trop.
Med. Hyg. 2004; 98, 268-269; Correspondence - Trans. Roy. Soc. Trop. Med. Hyg. 2005; 99:313-317;
Toovey S. Travel. Med. Infect. Dis. 2006; 4:71.
Johann-Liang R, Albrecht R. Clin. Infect. Dis. 2003; 36:1626; Gordi T, Lepist, EI. Toxicol. Lett. 2004;
147:99; Toovey S, Jamieson A. Response – Toxicol. Lett. 2004; 151:489; Toovey S. Lancet 2006; 367: 111;
Toovey S. Clinical Infect. Dis. 2006; 42; 1214; Toovey S. J. Am. Trop. Med. Hyg. 2006; 74:939; Toovey S.
Toxicol. Lett. 2006; 166:95.
31
32
Campos S, de la Cerda P, Rivera A. J. Pediatric Infect. Dis. 2008; 3:69.
Ramos-Martín V, González-Martínez C, Mackenzie I, Schmutzhard J, Pace C, Lalloo DG, Terlouw DJ. Am.
J. Trop. Med. Hyg. 2014; 91:62.
33
34
Vroman JA, Alvim-Gaston M, Avery MA. Curr. Pharm. Design 1999, 5, 101-138; Kim BJ, Sasaki T. Org.
Prep. Proc. Int. 2006; 38:1.
Haynes RK. Curr. Med. Chem. 2006; 13:509.

24
ACCEPTED MANUSCRIPT
35
36
Haynes RK. Curr. Opinion Infect. Dis. 2001; 14:719.
Haynes RK, Ho WY, Chan HW, Fugmann B, Stetter J, Croft SL, Vivas L, Peters W, Robinson BL. Angew.
Chem. Int. Ed. 2004; 43: 1381.
37
Haynes RK, Fugmann B, Stetter J, Rieckmann K, Heilmann HD, Chan HW, Cheung MK, Lam WL, Wong
HN, Croft SL, Vivas L, Rattray L, Stewart L, Peters W, Robinson BL, Edstein MD, Kotecka B, Kyle DE,
Beckermann B, Gerisch M, Radtke M, Schmuck G, Steinke W, Wollborn U, Schmeer K, Römer A. Angew.
Chem. Internat. Edit. 2006; 45:2082.
38
Schmuck G, von Keutz E, Haynes RK. Artemifone, A New Anti-Malarial Artemisinin Derivative: Lack Of
Neurotoxicity, XVI International Congress for Tropical Medicine and Malaria: Medicine And Health In The
Tropics; Marseille, France, 11-15 September 2005, Abstract Book, O-004, p. 26.
39
40
Ramharter M, Burkhardt D, Nemeth J, Adegnika AA, Kremsner PG. Am. J. Trop. Med. Hyg. 2006; 75:637.
Vivas L, Rattray L, Stewart LB, Robinson B, Fugmann B, Haynes RK, Peters W, Croft SL.J. Antimicrob.
Chemother. 2007; 59:658.
41
42
Marfurt J, Chalfein F, Prayoga P, Wabiser F, Wirjanata G, Sebayang B, Piera KA, Wittlin S, Haynes RK,
Möhrle JJ, Anstey NM, Kenangalem E, Price RN. Antimicrob. Agents Chemother. 2012; 56:5258.
Krudsood S, Wilairatana P, Chalermrut K, Leowattana W, Voith B, Hampel B, Looareesuwan S.
Artemifone, A New Anti-Malarial Artemisinin Derivative: Open Pilot Trial to Investigate the Antiparasitic
Activity of Bay 44-9585 in Patients with Uncomplicated P. Falciparum Malaria' XVI International
Congress for Tropical Medicine and Malaria: Medicine And Health In The Tropics; Marseille, France, 11-15
September 2005, Abstract Book, P054, p. 142.
43
44
Schmeer K, Breyer M, Hucke K, Gaefke D, Kern A, Schmuck G. Artemifone, a New Anti-Malarial
Artemisinin Derivative: Metabolism In Vitro in Liver Microsomes and Hepatocyte Sandwich Cultures of
Different Species Including Man. Presentation PO63, XVI International Congress for Tropical Medicine and
Malaria: Medicine And Health In The Tropics; Marseille, France, 11-15 September 2005, Abstract Book, O-
063, p. 144.
Nagelschmitz J, Voith B, Römer A. Artemifone, A New Anti- Malaria Artemisinin Derivative- Safety And
Pharmacokinetics In Healthy Male Subjects. XVI International Congress for Tropical Medicine and Malaria:
Medicine And Health In The Tropics; Marseille, France, 11-15 September 2005, Abstract Book, P047, p.
142; Nagelschmitz J, Voith B, Römer A. Artemifone, A New Anti- Malaria Artemisinine Derivative - Safety
and Pharmacokinetics of a 3 Days Treatment in Healthy Male Subjects. XVI International Congress for
Tropical Medicine and Malaria: Medicine And Health In The Tropics; Marseille, France, 11-15 September
2005, Presentation P048, Abstract Book p. 142; Steinke W, Beckermann B, Fischer H, Gerisch M,
Kohlsdorfer C, Römer A, Theis M, Welbers A. Artemifone, a New Anti-Malarial Artemisinin Derivative:
Pharmacokinetics in Rats and Dogs. XVI International Congress for Tropical Medicine and Malaria:
Medicine and Health in the Tropics; Marseille, France, 11-15 September 2005, Abstract Book, Presentation
P062, p.143.
45
46
47
Nagelschmitz J, Voith B, Wensing G, Roemer A, Fugmann B, Haynes RK, Kotecka BM. Rieckmann KH,
Edstein MD. Antimicrob. Agents Chemother. 2008; 52:3085.
Na-Bangchang K, Krudsood S, Silachamroon U, Molunto P, Tasanor O, Chalermrut K, Tangpukdee N,
Matangkasombut O, Kano S, Looareesuwan S. Southeast Asian J. Trop. Med. Public Health 2004; 35:575.
Like artemether, arteether was first prepared at the Institute Materia Medica in Shanghai in 1976, but
artemether was chosen by the Chinese for development, because of its superior efficacy in vivo against P.
berghei in mice (ref. 4). It is worth noting that arteether was selected for development in the West following
a decision by the Steering Committee of the Scientific Working Group on Malaria Chemotherapy of the
World Health Organization in Geneva, Switzerland in 1985 for the reasons that "arteether with an ethyl ether
group instead of the … methyl ether group in artemether would be more lipophilic, a possible advantage for
its accumulation in brain tissue. Arteether …. was expected to afford ethanol by cleavage of the ethyl ether
function and not methanol as does artemether"; see also ref. 61 below. The tenuous nature of the reasoning
aside, highlighted by the assumption that metabolism of artemether and arteether (that involves oxidative

25
ACCEPTED MANUSCRIPT
dealkylation) will provide methanol and ethanol respectively, the effort was eventually abandoned after
considerable expenditure.
48
https://www.pharmaasia.com/2017/11/fda-orphan-drug-designation-artemisone-artemis/; accessed February
2018.
49
50
Dunay IR, Haynes RK, Chan, WC, Sibley LD. Antimicrob. Agents Chemother. 2009; 53:4450.
Mazuz ML, Haynes RK, Shkap V, Fish L, Wollkomirsky R, Leibovich B, Molad T, Savitsky I, Golenser J.
Vet. Parasitol. 2012; 187:99.
51
52
53
Müller J, Balmer V, Winzer P, Rahman M, Manser V, Haynes RK, Hemphill A. Int. J. Antimicrob. Agents
2015; 46:88.
Sisto F, Scaltrito MM, Masia C, Bonomi A, Coccè V, Marano G, Haynes RK, Miani A, Farronato G,
Taramelli D. Internat. J. Antimicrob. Agents 2016; 48: 101.
Wolf D, Haynes RK. Methods and Compositions for Treating Viral Infections. U.S. Provisional Patent
Application No. 61/625,701, filed: April 18, 2012; granted May 2013; Oiknine-Djiana E, Weisblum Y,
Panet A, Wong HN, Haynes RK, Wolf DG. Antimicrob. Agents Chemother. 2018, 52, in press.
van Huijsduijnen RH, Guy RK, Chibale K, Haynes RK, Peitz I, Kelter G, Phillips MA, Vennerstrom JL,
Yuthavong Y, Wells TNC. PLoS ONE 2013; 8:e82962.
54
55
Gravett AM, Liu WM, Krishna S, Chan WC, Haynes RK, Wilson NL, Dalgleish AG. Cancer Chemother.
Pharmacol. 2011; 67:569.
56
57
Coghi P, Basilico N, Taramelli D, Chan WC, Haynes RK, Monti D. ChemMedChem 2009; 4;2045.
Laurent SAL, Robert A, Meunier B. Angew. Chem. Int. Ed. 2005; 44:2060; Bousejra-El Garah F, Meunier
B, Robert A. Eur. J. Inorg. Chem. 2008: 2133–2135.
58
59
Haynes RK. Angew. Chem. Int. Ed. 2005; 44:2064; Haynes RK, Chan WC, Lung CM, Uhlemann AC,
Eckstein U, Taramelli D, Parapini S, Monti D, Krishna S. ChemMedChem 2007; 2:1480.
Haynes RK, Cheu KW, Chan HW, Wong HN, Li KY, Tang MMK, Chen MJ, Guo ZF, Guo ZH, Sinniah K,
Witte AB, Coghi P, Monti D. ChemMedChem 2012; 7:2204.
60
61
62
Haynes RK, Wong HN, Wu Y, Wu WK, Cheu KW, Williams ID, Krishna S, Slavic K, Gravett AM, Liu
WM. ChemMedChem. 2016; 11:1469.
Brossi A, Venugopalan B, Dominguez-Gerpe L, Yeh HJC, Flippen-Anderson JL, Buchs P, Luo XD,
Milhous W, Peters W. J. Med. Chem.1988; 31:645.
Lin AJ, Klayman DL, Milhous WK. J. Med. Chem. 1987; 30:2147; O'Dowd H, Ploypradith P, Xie SJ,
Shapiro TA, Posner GH. Tetrahedron 1999; 55:3625; Posner GH, Ploypradith P, Parker MH, O’Dowd H,
Woo SH, Northrop J, Krasavin M, Dolan P, Kensler TW, Xie SJ, Shapiro TA. J. Med. Chem. 1999;
42:4275; Abouabdellah A, Bégué JP, Bonnet-Delpon D, Gantier JC, Nga TTT, Thac TD. Bioorg. Biomed.
Chem. Lett. 1996; 6:2717; Magueur G, Crousse B, Ourevitch M, Bonnet-Delpon D, Begue JP. J. Fluorine
Chem. 2006; 127:637.
63
Torok DS, Ziffer H, Meshnick SR, Pan XQ, Ager A. J. Med. Chem. 1995; 38:5045; Haynes RK, Wong HN,
Lee KW, Lung CM, Shek LY, Williams ID, Croft SL, Vivas L, Rattray L, Stewart L ChemMedChem 2007;
2:1464; Harmse R, Wong HN, Smit F, Haynes RK, N'Da DD. Curr. Med. Chem. 2015; 22:3607.
64 Abdel-Magid AF, Maryanoff CA, Carson KG. Tetrahedron Lett. 1990; 31:5595; McGill JM, LaBell ES,
Williams MA. Tetrahedron Lett. 1996; 37:3977.
65 Sato S, Sakamoto T, Miyazawa E, Kikugawa Y. Tetrahedron 2004; 60:7899.
66 Luo XD, Yeh HJC, Brossi A, Flippen-Anderson JL, Gilardi R. Helv. Chim. Acta 1984; 67:1515.
67
Haynes RK, Chan HW, Cheung MK, Lam WL, Soo MK, Tsang HW, Voerste A, Williams ID. Eur. J. Org.
Chem. 2002; 113-132.
68
Serrano JA, Jimmenez M, Roman E. J. Carbohydrate Chem. 1997; 16:1051; Ohkubo M, Nishimura T,
Kawamoto H, Nakano M, Honma T, Yoshinari T, Arakawa H, Suda H, Morishima H, Nishimura S.
Bioorg. Med. Chem. Lett. 2000; 10: 419; Chisholm JD, Van Vranken DL. J. Org. Chem. 2000; 65:7541;
Oscarson, S. "Conjugation of Monosaccharides - Synthesis of Glycosidic Linkages in Glycosides,
Oligosaccharides and Polysaccharides" in Carbohydrates; Finch, P. Ed. Springer, Dordrecht, 1999.; Ch. 4,
pp. 150-186.

26
ACCEPTED MANUSCRIPT
69
Knerr T, Ochs S, Severin T. Carbohydrate Res. 1994; 265:177; Chisholm JD, Vrankem DLV. J. Org.
Chem. 1995; 60: 6672-6673.
70
71
Jung ME, Yang EC, Vu BT, Kiankarimi M, Spyrou E, Kaunitz J. J. Med. Chem. 1999; 42: 3899.
Postema MHD. Tetrahedron 1992; 40; 8545; Toshima K, Tatsuta K. Chem. Rev. 1993; 93:1503; Jaramillo
C, Knapp S. Synthesis 1994; 1-20; Zhu X, Schmidt RR. Angew. Chem. Int. Ed. 2009; 48:1900.
Voerste A. The Preparation and Evaluation of Known and New Derivatives of Artemisinin and Totally
Synthetic Trioxanes as Antimalarials. Bayer AG, Leverkusen, 1996; Haynes RK, Lam WL, Chan HW,
Cheung MK Tsang, HW, Greif G. C-10 Ester Derivatives of Artemisinin for Treatment of Malaria,
Coccidiosis, and Neosporosis. 98305595.5; 14 July 1998; Europe; July 1999; Haynes RK, Lam WL, Chan
HW, Cheung MK Tsang, HW, Greif G. C-9 Sulfur-, Nitrogen-, and Carbon Linked Derivatives of
Artemisinin for Treatment of Malaria, Coccidiosis, and Neosporosis. 98305628.4; 14 July 1998; Europe;
International (PCT) PCT/GB99/02272; July 1999; Haynes RK, Lam WL, Chan HW, Cheung MK Tsang,
HW, Greif G. C-10 Ether Derivatives of Artemisinin for Treatment of Malaria, Coccidiosis, and
Neosporosis. 98305594.8; 14 July 1998; Europe July 1999.
72
73
Haynes RK, Chan HW, Cheung MK, Chung SK, Lam WL, Tsang HW, Voerste A, Williams ID. Eur. J.
Org. Chem. 2003; 2098-2114.
74
75
Uchiyama T, Hindsgaul O. Synlett. 1996: 499-501.
Lin AJ, Li LLQ, Klayman DL, George CF, Flippen-Anderson JL. J. Med. Chem. 1990; 33: 2610; Yang YH,
Li Y, Shi YL, Yang JD, Wu BA. Bioorg. Med. Chem. Lett. 1995; 5: 1791.
76
77
78
Schmidt AH, Russ M. Chem. Ber. 1981; 114: 1099; Iqbal J, Khan MA. Synth. Commun. 1989; 19:515.
Lancelin JM, Morin-Allory L, Sinaÿ P. J. Chem. Soc., Chem. Commun. 1984; 355-356.
Davis, B. G; Chambers, D,; Cumpstey, I.; France, R.; Gamblin, D. "Synthesis and Activation of
Carbohydrate Donors: Acetates, Halides, Phenyl Selenides and Glycals" in Carbohydrates, Osborn, H. M. I.,
Ed.; Best Synthetic Methods Series Academic Press, Oxford, England 2003, Ch. 3, pp. 69-120.
Snyder DC. J. Org. Chem. 1995; 60:2638.
79
80
Adams R, Weeks LF. J. Am. Chem. Soc. 1916; 38:2514; Barrett AGM, Braddock DC, James RA, Koike N,
Procopiou PA. J. Org. Chem. 1998; 63: 6273; Manley PJ, Bilodeau MT. Org. Lett. 2002; 4:3127.
Haiss P, Zeller KP. Angew. Chem. Internat. Edit. 2003; 42:303.
81
82
83
Rhoads SJ, Michel RE. J. Am. Chem. Soc. 1963; 85:585.
Thomas V. Lee, in Comprehensive Organic Synthesis, 1991,Vol. 7, pp. 291-303; section 2.8; Oxidation
Adjacent to Oxygen of Alcohols by Activated DMSO Methods; 2.8.3.6. DMSO-thionyl chloride; Pergamon,
Oxford, 1991.
84
85
86
Page PCB, Graham AE, Bethell D, Park BK. Synth. Commun. 1993; 23:1507.
Balicki R. Synth. Commun. 1999; 29:2235.
El-Feraly FS, Al-Yahya MA, Orabi KY. J. Nat. Prod. 1992; 55:878; Pu YM, Ziffer H. J. Labelled Comp.
Radiopharm. 1993; 33:1013.