UNIVERSITY OF CINCINNATI
Date:______
I, ______, hereby submit this work as part of the requirements for the degree of: in:
It is entitled:
This work and its defense approved by:
Chair: ______
SYNTHESIS OF ORGANOARSENIC COMPOUNDS
FOR ELEMENTAL SPECIATION
A Dissertation submitted to the
Division of Research and Advanced Studies of the University of Cincinnati
in partial fulfillment of the requirements for the degree of
DOCTOR OF PHILOSOPHY (Ph.D.)
in the Department of Chemistry of the College of Arts and Sciences
2004
by
Michael Fricke
B.S., The Ohio State University, 1997 B.A., The Ohio State University, 1998
Committee: Professor John Thayer, Co-Chair Professor Joseph Caruso, Co-Chair Professor Anna Gudmundsdottir ABSTRACT
Ongoing toxicokinetic and biogenesis investigations require gram quantities of
naturally occurring arsenobetaines and dimethylribofuranosides. The principal synthetic
routes to these compounds are hazardous; at least one laboratory explosion has occurred.
New routes to these compounds are now reported that eliminate dangerous procedures
and materials inherent in these syntheses.
Previously reported syntheses of arsenobetaines have all involved trimethylarsine
as a synthetic precursor. Due to the high vapor pressure (BP 52oC), extreme toxicity and pyrophoric nature of this compound it is both difficult and expensive to handle in a laboratory. In seeking to avoid this trimethylarsine as a precursor in the synthesis of arsenobetaines, the novel arsines ethoxycarbonylmethyldimethylarsine and
2-ethoxycarbonylethyl(dimethyl)arsine have been synthesized and isolated. The higher vapor pressure of these arsines permits purification by vacuum distillation and allows for safer handling. Subsequent conversion to arsenobetaines is simple and this route is preferred to those previously reported on the basis of material costs, operator time and safety.
The synthesis of the desired dimethylribofuranoside has involved the hydrogen peroxide oxidation in ether of a parent arsine to provide the desired arsine oxide moiety.
This reaction is hazardous. New routes to arsine oxides have been explored including the reduction of novel arsinoyl imidazolides and rearrangements involving Meyer chemistry that dates from the year 1883. Finally, the oxidation of arsines using the mild oxidant dimethyldioxirane was attempted.
I The dimethyldioxirane oxidation of triphenylarsine provided the desired triphenylarsine oxide quickly and quantitatively. After demonstrating this reaction on the small scale oxidation of triphenylarsine, the dimethyldioxirane oxidation was employed to replace the oxidation step in the synthesis of the required dimethylribofuranoside. This alternative oxidation worked well and the hazardous oxidation of arsines with hydrogen peroxide in ether can be abandoned.
II
III
Dedicated To
Mom and Dad
Without their love and patience none of this would have been possible.
IV ACKNOWLEDGEMENTS
I would like to thank a number of people who have contributed to my graduate
school experience. First I would like to thank Professor John Alexander. Unfortunately,
I was one of the last individuals to have the opportunity to learn on a graduate level from
this accomplished chemist. With his untimely passing, we lost a patient man who went
out of his way to teach those that others considered to be lost causes. He leaves a legacy
of excellence and generosity that I hope to emulate in some small part in my future
endeavors. JJA is missed but can never be forgotten.
I would like to thank Professor John Thayer for doing his best to fill some very
big shoes in his capacity as my replacement advisor. Dr. Thayer was an easy choice
because of his close friendship to Dr. Alexander and my own experience assisting him in
the teaching of freshman chemistry. He brought a fresh perspective to my research and
valuable experience with organoarsenic chemistry.
Professor Joseph Caruso or “Doc” has been a pillar since my childhood. His
example in no small part contributed to my early interest in chemistry and his guidance
played an even larger part in my opportunity to pursue this interest. Doc’s chemistry is
an enlightened network of the world’s top analytical scientists coupled with a laboratory ethos that emphasizes individual creativity and performance. Doc has cultured a workspace where he never has need to speak a harsh word to any of his students. He has been close to the ideal advisor.
Professor Anna Gudmundsdottir and I both started our work at the University of
Cincinnati in 1998 and she has been a continual presence for me throughout graduate school - teaching my 8:00 AM class on advanced organic chemistry during my first
V quarter through the final signature approving this thesis. I am indebted to her for the
honest advice she provided in the early years and then on experimental setups and finally
on what was necessary to finish my thesis. Through her efforts, my science has been
invaluably strengthened.
Professor William Cullen of the University of British Columbia has been a source
of insight since our first contact in Graz at the ICEBAMO meeting of 2000. Our collaboration has involved endless correspondence and the recent privilege of spending one month working in his laboratory. I wish him a long and enjoyable retirement but expect I’ll be hearing of his continued efforts in arsenic chemistry long after the renovation of his laboratory building provides the impetus to finally hang up his lab coat
and goggles.
Professor David Hart of the Ohio State University has made himself available to
me for consultation ever since I enrolled in his undergraduate honors organic laboratory
course in the spring of 1994. Although he finds organometallic chemistry a bit
inconsistent, his expertise in the synthesis of natural products has provided easy solutions
to problems arising in my own research. He will be among my first choices for advice on
future research.
Professor Andrew Benson at the Scripps Institute of Oceanography has provided
considerable perspective on my studies. Since his own Ph.D. defense at the California
Institute of Technology in 1941, Dr. Benson has had a distinguished career including the
use of radioactive carbon dioxide to discern the path of carbon in photosynthesis and
more recently he has been interested in the biochemistry of arsenic species in marine
organisms. I am grateful for the opportunity to hear the personal accounts of his
VI scientific exploration and am humbled by his kind words regarding my own accomplishments.
Dr. Jack Creed has been an able mentor at the U.S. Environmental Protection
Agency, where I have spent the last year working across the street from the campus in
Cincinnati. I will never be able to repay the opportunity Jack provided me or the patience he showed by permitting me to juggle my post-doctoral research while at the same time working to finish my thesis. The slight correlation between the two does not begin to explain the seemingly infinite freedom Jack allowed me to do what had to be done. It can not be understated that I would not have been able to complete my research after Dr.
Alexander died had Jack not provided me with a desk and a bench.
Finally, I would like to thank Dr. Maria Montes-Bayón for being an excellent post-doctoral fellow with the Caruso group and several graduate students including
Dr. Jason Day, Tyre Grant, Sasi Kannamkumarath and Eric Mack for teaching me practical laboratory skills. I had the opportunity to teach Sasi to drive and I think they all consider this some small repayment for their help.
VII 1
TABLE OF CONTENTS
Page
LIST OF FIGURES 6
LIST OF CHEMICALS 7
CHAPTER 1
SYNTHETIC CHEMISTRY FOR ELEMENTAL SPECIATION 12
1.1 Arsenic 13
1.2 Trace Level Elemental Speciation of Arsenic 17
1.3 Plasma Mass Spectroscopy 20
1.4 Electrospray Ionization Mass Spectroscopy 22
1.5 Organometallic Standards 25
1.6 Structural Configuration 26
1.7 Toxicokinetic, Toxicodynamic and Biogenesis Studies 26
1.8 Research 28
CHAPTER 2
SYNTHESIS OF ARSENOBETAINES VIA SODIUM
DIMETHYLARSENIDE 30
2.1 Introduction 31
2.2 Experimental 33
2.2.1 Reagents 33
2.2.2 Instrumentation 33
2.3 Synthetic Procedures 34
2.3.1 Dimethyliodoarsine (2) 34 2
Page
2.3.2 Sodium Dimethylarsenide (3) 35
2.3.3 Ethoxycarbonylmethyl(dimethyl)arsine (4) n=1 36
2.3.4 Ethoxycarbonylmethyl(trimethyl)arsonium iodide (5) n=1 36
2.3.5 Arsenobetaine (trimethylarsoniumacetate) (6) n=1 37
2.3.6 2-Ethoxycarbonylethyl(dimethyl)arsine (4) n=2 38
2.3.7 2-Ethoxycarbonylethyl(trimethyl)arsonium iodide (5) n=2 38
2.3.8 Arsenobetaine-2 (trimethylarsoniumpropionate) (6) n=2 39
2.4 Discussion 39
CHAPTER 3
ARSINE OXIDES REVISITED 41
3.1 Introduction 42
3.2 Cause of Explosion 44
3.3 Dioxirane – An Alternate Oxidant 46
3.4 Meyer Reaction 47
3.5 Arsinoyl imidazolides 48
CHAPTER 4
SYNTHESIS OF TRIPHENYLARSINE OXIDE BY OXIDATION
OF TRIPHENYLARSINE WITH DIMETHYLDIOXIRANE 50
4.1 Introduction 51 3
Page
4.2 Experimental 53
4.2.1 Reagents 53
4.2.2 Instrumentation 53
4.3 Synthetic Procedures 54
4.3.1 Dimethyldioxirane (DMD) (9) 54
4.3.2 Determination of Concentration of DMD 55
4.3.3 Triphenylarsine oxide (12) 56
4.4 Discussion 56
CHAPTER 5
SYNTHESIS OF (R)-2,3-DIHYDROXYPROPYL 5-DEOXY-
DIMETHYLARSINOYL-β-D-RIBOSIDE (As328) 58
5.1 Introduction 59
5.2 Experimental 62
5.2.1 Reagents 62
5.2.2 Instrumentation 62
5.2.3 Chromatography 63
5.3 Synthetic Procedures 64
5.3.1 Methyl 2,3-O-Isopropylidene-β-D-riboside (14) 64 4
Page
5.3.2 Methyl 2,3-O-Isopropylidene-5-O-toluenesulfonyl-β-D-riboside (15) 65
5.3.3 Methyl 5-Bromo-5-deoxy-2,3-O-isopropylidene-β-D-riboside (16) 66
5.3.4 (R)-2,3-Dibenzyloxypropyl 5-Bromo-5-Deoxy-2,3-O-isopropylidene-
β-D-riboside (18) 66
5.3.5 Dimethyliodoarsine (2) 67
5.3.6 Sodium dimethylarsenide (3) 67
5.3.7 (R)-2,3-Dibenzyloxypropyl 5-Deoxy-5-dimethylarsino-2,3-O-
isopropylidene-β-D-riboside (19) 68
5.3.8 (R)-2,3-Dihydroxypropyl 5-Deoxy-5-dimethylarsino-2,3-O-
isopropylidene-β-D-riboside (20) 68
5.3.9 Dimethyldioxirane (9) 69
5.3.10 (R)-2,3-Dihydroxypropyl 5-Deoxy-5-dimethylarsinoyl-2,3-O-
isopropylidene-β-D-riboside (21) 70
5.3.11 (R)-2,3-Dihydroxypropyl 5-Deoxy-5-dimethylarsinoyl-β-D-riboside
(As328) (22) 71
5.4 Discussion 73
CHAPTER 6
DISCUSSION 74
6.1 Discusion 75
CHAPTER 7
FT-IR SPECTRA 76
5
Page
CHAPTER 8
MASS SPECTRA 88
CHAPTER 9
NMR SPECTRA 98
6
LIST OF FIGURES
Page
1.1 Arsenic Species 19
1.2 Toxicity of Arsenic Species 20
1.3 Schematic of an ICP-MS 21
1.4 ICP-MS Chromatogram of Arsenic Species 22
1.5 Z Spray Source for ESI-MS 24
2.1 New Route to Arsenobetaines 32
. 3.1 Structure of TMAO HI 48
3.2 Activation of Dimethylarsinic acid 49
4.1 Mechanism of Dioxirane Oxidation of Arsines 52
4.2 Dioxirane Oxidation of Thioanisole 53
4.3 GC-MS Analysis of Conversion of Thioanisole to Phenylmethylsulfoxide 55
5.1 Four Major Arsenosugars 60
5.2 Synthesis of As328 61
5.3 Anion chromatography of Synthetic As328 72
5.4 Chromatographic profile of Anionic Arsenic Standards 72
7
LIST OF CHEMICALS
O 1
H3C AsOH
CH 3
CH3 2 As H3C I
CH3 3
NaAs
CH3
4 CH3
As CO2Et ( ) H3C n
CH3 5 H3C As+ CO Et 2 H C ( ) 3 n
CH3 6 H3C + As CO2Et
H3C ( )n
8
7 CH3 H + - As [ HSO3 ] H C I 3
8 Ph As 3
O 9 CH3 O
CH 3
10 S CH 3
O 11 S CH3
12 Ph3AsO
HO 13 OH O
OH OH 9
14 HO OMe O
O O
15 TsO OMe O
O O
16 Br OMe O
O O
17 HO OBn
OBn
10
18 Br O OBn O OBn
O O
19 Me2As O OBn O OBn
O O
Me2As 20 O OH O OH
O O
O
21 Me2As O OH O OH
O O 11
O 22 Me2As O OH O OH
OH OH
12
CHAPTER 1
INTRODUCTION:
SYNTHETIC CHEMISTRY FOR ELEMENTAL SPECIATION
SYNTHETIC BIOORGANOMETALLIC CHEMISTRY
13
1.1 ARSENIC
For over 5000 years, human exposure to arsenic has resulted from their
industrious activities. In 1991, an ancient coppersmith, now known as Otzi, was
discovered frozen in the Italian Alps and arsenic analysis provided the key to discovering
the iceman’s occupation.1 Having died around 3000 B.C. from a bow and arrow wound,
the well preserved mummy had traces of arsenic in his hair indicative of chronic arsenic
poisoning. These unusually high levels of arsenic are most readily explained by an
exposure over a long period to arsenic vapor or arsenic dust produced in the smelting of
copper. Indeed, the copper axe the iceman carried with him contained 0.2% arsenic.
Arsenic analysis of the axe indicated the copper was from a nearby source and meant the
Copper Age had begun in the region at least a thousand years earlier than previously
realized.
Humans have been aware of arsenic in the environment since antiquity and three
forms of arsenic, white (As2O3), red (As2S2) and yellow (As2S3) were known in 1250
A.D. when Albertus Magnus obtained the free form of the element by heating soap with
2 opriment (arsenic trisulfide, As2S3). Arsenic has long maintained a reputation as a poison although it is ubiquitous in the environment, ranking 20th in abundance of the elements in the earth’s crust.3 In the 1500s, experiments by Paracelsus led to the
observation of various physiological responses to different doses of arsenic and other
poisons and mark the foundation of the science of toxicology.4 In the year 1858,
1 Spindler, K. The man in the ice : the discovery of a 5,000-year-old body reveals the secrets of the Stone Age; Toronto: Doubleday, Canada, 1994. 2 Emsley, J. The Elements, 3rd. ed.; Clarendon Press: Oxford, 1998. 3 National Research Council. Arsenic; National Academy of Sciences: Washington D.C., 1977. 4 Deichmann, W.B.; Henschler, D.; Keil, G. Archives of Toxicology, 1986, 58, 207 14
Alexander Borodin observed the toxic effects of arsenicals on different organs in
experiments involving the intentional poisoning of dogs.5
The first synthetic organometallic compound was an arsenical. In 1760, attempts by French chemist Louis Claude Cadet de Gassicourt to invent an invisible ink led to the creation of a noxious liquid that was soon named dicacodyl (evil smelling).6 This
7 compound was identified as As2(CH3)4 in 1848 through the efforts of Bunsen. An important milestone in organometallic chemistry occurred in 1893 when an Italian chemist named Gosio was consulted on the unexplained deaths of over a thousand children. In the classical study that followed, Gosio determined the cause of the deaths to be an arsenic containing pigment found in the rooms of the victims and surmised the action of mildew was causing the release of an arsine gas.8 Forty years later, Challenger identified the volatile metabolite as trimethylarsine and established the basis for biomethylation as a natural means for detoxification of toxic metals.9
Arsenic compounds have found wide use both sinister and beneficial. The
invention of a sensitive test for arsenic in 1836 by Scottish chemist James Marsh
effectively ended the previous widespread use of arsenic as a poison and his test was used
in the first forensic evidence of poisoning ever presented to a jury.10 The development of an arsenical based blistering agent codenamed Lewisite at the end of World War One led to mass production of massive stocks of this weapon by both sides of combatants during
5 Benson, A.; Rosenberg, H. Alexander Borodin M.D. presented at the 6th International Conference on Environmental and Biological Aspects of Main-Group Organometallics, Université de Pau et des Pays de l’Adour, Pau, France 2003 (Lecture 5). 6 Seyferth, D. Organometallics, 2001, 20, 1488. 7 Davenport, D.; ChemMatters, 1984, 2, 14. 8 Gosio, B. Archives Italiennes de Biologie, 1893, 18, 253. 9 Challenger, F.; Higginbottom, C.; Ellis, L. Journal of the Chemical Society, 1933, 95. 10 Geber, S.M.; Saferstein, R. (eds.) More Chemistry and Crime; American Chemical Society: Washington, 1997. 15
World War Two.11 An intensive British research effort resulted in the discovery of
British Anti-Lewisite (BAL).12 The active ingredient of BAL, 2,3-dimercaptopropanol,
was the world’s first rationally designed chelating agent and worked by sequestered the
arsenical rendering it harmless.
Lead arsenate, PbHAsO4, was the primary insecticide used in fruit orchards prior
to the discovery of DDT in 1947.13 Although inorganic arsenic compounds are no longer used in agriculture, arsenic levels in some soils continue to be high from agricultural loading prior to the widespread banning of the practice that occurred over the last two
decades.14 Organic arsenicals such as dimethylarsinic acid, disodium methylarsonate, and monosodium methylarsenate are still used as pesticides, principally on cotton. The use of copper arsenate in wood preservatives represents the largest use of arsenic in the
U.S. and is currently embroiled in calls to ban this practice for fear of toxin release.15
Arsenic has found other uses in the industrial areas of glass, bullets, semiconductors, lasers and pharmaceuticals.
Arsenic has engendered near continuous interest in medicine since the first recorded use by Hippocrates, who administered opriment and realgar (As2S2) for the
treatment of ulcers.16 The arsenical based drugs Salvarsan and neo-Salvarsan were used
to treat syphilis (Treponema pallidum) before the discovery of penicillin. Convinced that
the toxicity of arsenic could be harnessed to treat bacterial diseases, the German chemist,
Paul Ehrlich, set about a search for his “magic bullet” that could kill the bacteria without
11 Compton, J.A.F. Military Chemical and Biological Agents: Chemical and Toxicological Properties, Telford Press: Caldwell, New Jersey, 1988. 12 Waters, L.I.; Stock, C. Science, 1945, 102, 601. 13 Shepard, H.H. The Chemistry and Action of Insecticides; McGraw-Hill: New York, 1951. 14 Welch, A.H.; Westjohn, D.B.; Helsel, D.R.; Wanty, R.B. Ground Water, 2000, 38, 589. 15 Arsenic in the Environment; Nraigu, J.O. Ed.; John Wiley: New York, 1994. 16 Miller, W.H.; Schipper, H.M.; Lee, J.L.; Singer, J.; Waxman, S. Cancer Reseach, 2002, 62, 3893. 16
harming the host. His 606th organoarsenic compound became the active ingredient in
Salvarsan, which proved an effective treatment for syphilis. Its release in 1910 marks the beginning of the use of chemotherapeutic disease specific medications.17 The investigation of arsenicals in medicine has recently generated renewed interest with the report of the use of arsenic trioxide to treat acute promyelocytic leukemia.18
Arsenic contaminated drinking water and arsenic accumulated in plant and animal food sources are the two major pathways of toxic arsenic exposure in humans. The symptoms of moderate arsenic exposure or poisoning include muscle spasms, garlic odor of the breath, diarrhea, cardiovascular collapse, aplastic anemia and death.19 Arsenic has long been known to be carcinogenic from chronic low level exposure.20
Because small amounts are consumed on a routine basis through food and
drinking water, government guidelines are set for controlling arsenic to minimize health
risks associated with arsenic exposure. The toxicity of arsenic has recently prompted the
United States Environmental Protection Agency (EPA) to lower the standard for maximum allowable arsenic in drinking water from 50 ppb to 10 ppb.21 This regulation is defined as a total allowable limit because the chief form of arsenic in drinking waters is known to be mostly inorganic arsenic.
17 Aronson, S. Medicine, 1994, 77, 233. 18 Zhang, P.; Wang, S.Y.; Hu, X.H. Chinese Journal of Hematology, 1996, 17, 58. 19 Franzblau, A.; Lilis, R. Archives of Environmental Health, 1989, 44, 385. 20 Tseng, W.P.; Chu, H.M.; Fong, J.M.; Lin, C.S.; Yeh, S. Journal of the National Cancer Institute, 1968, 40, 453. 21 Federal Registrar: Safe Drinking Water Regulation – Update – Arsenic Rule, 2002. 17
1.2 TRACE LEVEL ELEMENTAL SPECIATION OF ARSENIC
‘Elemental speciation’ involves the determination of an element’s chemical
form(s) in a sample and seeks to elucidate the isotopic, inorganic, organic,
organometallic, or complex nature in which the element exists. The significance of this
area of research in bioorganometallic chemistry comes from the growing awareness that
the toxicity, mobility, biological activity and the potential for bioaccumulation of an
element varies widely with the chemical form of the element.
Elemental speciation is positioned to make a particularly significant impact to
food, environmental and health sciences.22 Maximum allowable elemental guidelines are
set by legislative regulations seeking to minimize risk to humans and the ecosystem.
Routine speciation analysis will allow these threshold limits to evolve from the current
basis of total elemental content into a more useful set of rules governed by species
specific regulations. Speciation studies have been reported for such toxic elements as
arsenic23, cadmium24, lead25, mercury26 and tin27. The essential elements iron28, selenium29, iodine30 and chromium31 have also been investigated for speciation.
Arsenic speciation studies are part of the broad effort to evaluate arsenic exposure and determine health effects. In contrast to arsenic exposure from drinking water, plant
22 Trace Element Speciation for Environment, Food and Health; Ebdon, L., Ed.; Royal Society of Chemistry: Cambridge, 2001. 23 Corr, J.J.; Larsen, E.H. Journal of Analytical Atomic Spectroscopy 1996, 11, 1215-1224. 24 Crews, H.M.; Dean, J.R.; Ebdon, L.; Massey, R.C. Analyst 1989, 1989, 895. 25 Ebdon, L.; Hill, S.J.; Rivas, C. Spectrochimica Acta 1998, 53, 289. 26 Slaets, S.; Adams, F.; Pereiro, I.R.; Lobinski, R. Journal of Analytical Atomic Spectroscopy 1998, 14, 851. 27 Kumar, U.T.; Dorsey, J.G.; Caruso, J.A.; Evans, E.H. Journal of Chromatography A 1993, 654, 261. 28 Cook, J.D.; Dassenko, S.A.; Lynch, S.R. American Journal of Clinical Nutrition 1991, 54, 717. 29 Zheng, J.; Goessler, W.; Kosmus, W. Trace Elements and Electrolytes 1998, 15, 70. 30 Dermelj, M.; Slevkovec, Z.; Byrne, A.R.; Stegnar, P.; Rossbach, M. Fresenius Journal of Analytical Chemistry 1990, 338, 559. 31 Pantsar-Kallio, M.; Manninen, P.K.G. Journal of Chromatography A 1996, 750, 89. 18
and animal food products contain arsenic in a variety of forms and require speciation
analysis to determine exposure.32 The diversity of natural arsenic species in organisms
occurs because of different means of detoxification used throughout the food chain.
Many organisms are capable of some degree of arsenic metabolism and the
biomethylation of arsenic is a common means of detoxification.33 Scopulariopsis
brevicaulis, the mold implicated in the generation of Gosio gas was simply employing a
methylation and volatilization scheme to remove copper arsenite from its environment.34
The determination of arsenic exposure and risk is made more complex by the observation that the toxicity of arsenic species in humans is known to decrease with increasing organic nature.35 Arsenocholine and arsenobetaine are relatively nontoxic or a metabolically inert species. Arsenic has even been reported as a possible essential element at trace levels.36 Figure 1.1 shows several arsenic species determined in plants and animals and Figure 1.2 demonstrates the wide variance in acute toxicity of arsenic
37 species with median lethal dose values (LD50) determined by Kaise et al.
Speciation research to determine the distribution of toxic and non-toxic arsenic
species in the food chain is ongoing and efforts have focused on determining sources of
arsenic exposure, biological pathways and toxic effects. Feldmann et al. has investigated
the arsenic metabolism in seaweed-eating sheep from northern Scotland.38 Pergantis et
32 Thayer, J.S. Organometallic Compounds and Living Organisms; Academic: Orlando, FL, 1984. 33 Wood, J.M. Science 1974, 183, 1049-1052. 34 Cullen, W.R.; Reimer, K.J. Chemical Reviews 1989, 89, 713-764. 35 Edmonds, J.S.; Francesconi, K.A. Marine Pollution Bulletin 1993, 26, 665. 36 Fairweather-Tait, S.J.; Hurrell, R.F. Nutrition Research Reviews 1996, 9, 295. 37 Kaise, T.; Yamauchi, H.; Horiguchi, Y.; Tani, T.; Watanabe, S.; Hirayama, T.; Fukui, S. Applied Organometallic Chemistry, 1989, 3, 273. 38 Feldmann, J.; John, K.; Pengprecha, P. Fresenius Journal of Analytical Chemistry 2000, 368, 116-121. 19
al. used Fast Atom Bombardment Tandem Mass Spectrometry to characterize
arsenosugars of biological origin.39
O O O OH OH CH OH As As As 3 As HO OH HO OH H3C CH3 H3C OH
Arsenate As(V) Arsenite As(III) Trimethylarsine oxide Monomethylarsonic acid
O CH3 O
OH H3C + H3C As As As CH2OH H3C CH3 H3C CH3 H3C
Dimethylarsinic acid Tetramethylarsonium ion 2-(Dimethylarsinyl)ethanol
CH3 CH3 CH3
H3C - H3C H3C As+ CO2 As+ CH2OH As+ - H3C H3C H3C CO2
Arsenobetaine Arsenocholine Arsenobetaine-2
O CH3
As As+ H3C O OR H3C O OR
H3C H3C
HO OH HO OH
Dimethylarsinylribosides Trimethylarsonioribosides
Figure 1.1 Arsenic Species
39 Pergantis, S.A.; Francesconi, K.A.; Goessler, W.; Thomas-Oates, J.E. Analytical Chemistry 1997, 69, 4931-4937. 20
Arsenic Compound LD50 (mg/Kg) Animal ______Inorganic arsenite [As (III)] 4.5 rat Inorganic arsenate [As (V)] 14-18 rat Monomethylarsonic acid 1,800 mouse Dimethylarsinic acid 1,200 mouse Trimethyarsine oxide 10,600 mouse Arsenobetaine 10,000 mouse Arsenocholine 6,000 mouse
Figure 1.2. Toxicity of Arsenic Species.
1.3 PLASMA MASS SPECTROMETRY
Conventional methods of natural product identification, i.e. X-ray crystallography
and Nuclear Magnetic Resonance (NMR) are not applicable for characterizing and
detecting many organometallic species found in biological and environmental samples
because of the trace or ultra-trace levels of these compounds. Instead, High-Performance
Liquid Chromatography-Inductively Coupled Plasma-Mass Spectrometry (HPLC-ICP-
MS) has been used to routinely determine chemical speciation and quantification in many
samples.
ICP-MS analysis offers the capability for detection and quantification
approaching part-per-trillion levels, specific detection by one or multiple atomic ion
monitoring and linearity (≥106 orders of magnitude). The inability of the plasma source to provide information concerning molecular structure is a significant limitation of ICP-
MS. ICP-MS is an elemental or atomic technique by virtue of its destruction of all molecular information in the “harsh” plasma source. An overhead schematic of the
Agilent 7500 ICP-MS (Agilent Technologies, Palo Alto, CA, USA) used in this research is shown in Figure 1.3. A sample is nebulized by argon flow and the resulting fine droplets are carried into the high temperature argon plasma region (4000 – 10,000 K) 21 generated by argon excitation by an induction coil powered by a radio-frequency generator. After near complete atomization of elemental species, the sample undergoes excitation and ionization before entering a mass analyzer.
Elemental speciation using ICP-MS requires the coupling of a separation technique such as High-Performance Liquid Chromatography (HPLC).40 By matching reproducible chromatographic retention times between analytes and reference standards, species can be identified and quantified by MS response. ICP-MS chromatogram showing the separation over time of several arsenic species is shown in Figure 1.4.
Figure 1.3. Schematic of an ICP-MS (from Agilent HP 4500 ICP-MS operating software)
40 Ponce de León, C.A.; Montes-Bayón, M.; Caruso, J.A. Journal of Chromatography A 2002, 974, 1-21. 22
Figure 1.4. ICP-MS Chromatogram of Arsenic Species
1.4 ELECTROSPRAY IONIZATION MASS SPECTROMETRY
Electrospray Ionization-Mass Spectrometry (ESI-MS) has become widely used in
biological, organic, pharmaceutical, and medical research.41 In contrast to ICP, ESI is capable of high yield transfer of molecular ions from solution into the gas phase without chemical degradation. Because of this ‘soft ionization’ and the ready coupling of this source to mass spectrometry, ESI-MS is emerging as a compliment to NMR and X-ray crystallography in molecular structure identification in organometallic and inorganic solution chemistry.42
41 Electrospray Ionization Mass Spectrometry; Cole, R.B. Ed.; John Wiley: New York 1997. 42 Crociani, L.; Anacardio, R.; Traldi, P.; Corain, B. Inorganica Chimica Acta 1998, 282, 119-122. 23
ESI-MS is a powerful tool for elemental speciation because of the molecular
information it generates but has the notable limitation that ESI-MS typically has lower
sensitivity than ICP-MS with detection limits several orders of magnitude higher for most
metallic elements. In many cases the trace level of analytes may prevent ESI-MS
characterization without additional expense in operator time or instrument requirements.
Current speciation analysis requires information from both ESI and ICP sources.
A schematic of the Z Spray ESI source (Micromass, Manchester, UK) used in this
research is shown in Figure 1.5. A voltage is applied between the metal capillary needle
and a counter electrode. A sample solution containing positive and negative electrolytes
is fed into the needle by HPLC or direct infusion (syringe pump). These ions then
distribute to counteract the imposed field. If the capillary is acting as the positive electrode, positive ions will collect at the tip of the needle until a critical point is reached when a fine jet containing these ions escapes the tip and disperses. Desolvation occurs rapidly under low pressure conditions and the sample is drawn into the mass analyzer as individual molecular ions. The orthogonal positioning of the sampling and extraction cones relative to the needle source makes the Z Spray less susceptible to clogging from high salt content samples.
Tandem mass spectrometry is a powerful analytical technique in which a single
mass molecular ion is selected, collisionally fragmented, and the resulting fragment ions are mass measured. A mass resolving filter allows selectivity in the choice of precursor ions to within a mass to charge (m/z) range variable down to 1 dalton. The selected ions then undergo Collision Induced Dissociation (CID) in which a collision gas (usually argon or nitrogen) is bled slowly into the chamber containing the accelerated molecular 24 analyte ions. Collisions between the molecular ions and the gas cause the molecular ions to fragment into smaller ions. By changing the voltage drop across the differential aperture entrance to the CID chamber, it is possible to vary collision energies and identify a number of structurally significant fragments.
Figure 1.5. Z Spray Source for ESI-MS (from Micromass Promotional Material)
25
1.5 ORGANOMETALLIC STANDARDS
A Reference Material (RM) is a material with sufficiently well established properties sufficient for the calibration of an instrument, method or material. For speciation analysis to become the basis for legislation or legal proceedings, instrumental methodology must be traceable to a certified reference material (CRM) of known species distribution. A CRM has a determined value that is certified or traceable to documentation issued by a certifying body such as the U.S. National Institute of
Standards and Technology (NIST). This important aspect of quality control is often limited because of the unavailability of CRMs containing species of interest. It is often necessary to synthesize organometallic species to be used as RMs when no commercial source exists.
For speciation by HPLC-ICP-MS, the matching of retention times is often difficult because chromatographic mechanisms are highly dependant upon a variety of factors including composition and pH of matrices and buffered mobile phase. Speciation by HPLC-ICP-MS ideally requires access to an exhaustive library of organometallic compounds in every type of matrix. In practice, this requires a ready supply of standards for routine calibration. Reference materials should ideally be stable for long periods, homogeneous, representative of analytical samples in matrix composition and of similar concentration to the analytes.
26
1.6 STRUCTURAL CONFIRMATION
Unidentified peaks in HPLC-ICP-MS chromatograms present a major challenge to speciation studies. The characterization strategy of these unknowns is highly specialized and may involve the concentration and purification of several unknown analyte-containing fractions collected from multiple chromatographic runs. This time consuming process can continue until a sample is prepared that can be analyzed within the sensitivity of ESI-MS or more conventional means. An ‘exact’ molecular weight determination by ESI-MS combined with a molecular fragmentation pattern and prior knowledge of the system can be enough to assign a tentative structure to an unknown compound. Final confirmation of structure occurs when a synthetic sample is prepared and matched with an unknown’s chromatographic retention times an all other analytical characteristics.
1.7 TOXICOKINETIC, TOXICODYNAMIC AND BIOGENESIS STUDIES
Toxicokinetics, toxicodynamics and biogenesis studies are conducted to evaluate biological effects of toxins on organism and the metabolic products of ingested toxins.
Toxicokinetic and toxicodynamic research seeks to elucidate the mechanisms of biological action of chemicals and the correlation of adverse biological effects with concentration of species in affected tissue. In biogenesis studies, the metabolic fate of a chemical is of interest. By increasing the level of exposure to a specific chemical, the metabolic degradation or reaction products can similarly be amplified in concentration to the point of instrumental detection and characterization. This type of research provides an opportunity for the synthetic chemist because large quantities are required of the 27
chemical of interest. Synthesis of chemicals for toxicokinetic and biogenesis studies is
the often the only means of preparing quantities sufficient for required experimental
design.
As an example, the Norwegian National Institute of Nutrition and Seafood
Research has sought to establish a toxicokinetic model for arsenobetaine in Atlantic salmon and Atlantic cod. These two species contain distinct fat content and have varying arsenic concentrations in different tissues. Their project goal includes the description of:
1) the transfer of arsenobetaine from feed to fish 2) the bioavailability of arsenobetaine and 3) the accumulation, distribution and retention of arsenobetaine in fish. This knowledge will be important in the documentation of food safety.
An immediate problem facing this group was that arsenobetaine is not commercially available and gram quantities of arsenobetaine were necessary for the required experimental design. Processing and extracting arsenobetaine from natural sources was both time-consuming and not readily scalable to produce large amounts.
Under a special agreement, we were able to provide these researchers with 15 grams of arsenobetaine bromide
The synthesis of the arsenosugar 22 described in Chapter 5 was necessary to provide gram quantities for studies seeking to model metabolic pathways of this naturally occurring arsenical. One study seeks to determine the uptake and bioavailability of this arsenosugar in the human body. By extracting and collecting fluid from the lower intestines of mice, a biologically active solution will be used to study the pathways of biodegradation of arsenosugar 22 in mammals. In the second experiment, a toxicity test on cell cultures will be used to predict acute toxicity for arsenosugar 22. 28
1.8 RESEARCH
The diversity of naturally occurring main-group organometallic species presents a
great challenge for researchers interested in understanding the complex mobility and
biotransformation of elements through our ecosystem. The isolation, identification and
classification of organometallic natural products in marine and terrestrial plants and
animals is truly in its infancy as a science. The opportunity for the application of
knowledge attained in this field to pharmacological medicine is immense. The use of organometallic containing drugs is an under explored area and holds the potential for treatments for some of the worst diseases of our time. Concurrently, an understanding of the toxicity of organometallic species in food organisms is essential in the development
of exposure models and dietary estimates.
The objectives of this research are: 1) to provide critical synthetic knowledge in
the production of unavailable standards for analytical speciation laboratories 2) to
produce gram-quantities of unavailable organometallic compounds for research
investigating toxicity, toxicokinetics, and biogenesis of these species 3) to determine
areas in need of original synthetic bioorganometallic research by collaboration and
communication with the international speciation community 4) to identify novel
organometallic species recently identified in nature and to target these compounds for
efforts leading towards their total syntheses 5) to develop the basic synthetic reactions
and methodology necessary to synthesize these compounds 6) to work towards a higher
understanding of bioorganometallic transformation in nature and 7) to improve safety by
working to minimize dangerous procedures and materials inherent to organometallic
synthesis. 29
In this capacity the following discoveries have been made: 1) the development of an alternate synthetic route that provides arsenobetaines while avoiding trimethylarsine as a precursor. This work led to the creation and characterization of several novel arsine compounds. Fortuitously, this route also provides for a simpler route in the synthesis of isotopically labeled arsenobetaines and is being utilized by other researchers for this purpose. 2) The rediscovery of the Meyer synthesis of trimethylarsine oxide of 1883 and the development of this reaction using modern synthetic methodology as an alternative to the extremely hazardous route that had been employed for the synthesis of trimethylarsine oxide during the last forty years. 3) The invention of a new method for the general oxidation of arsines using dioxiranes. This discovery had immediate applications and was used in the penultimate step of the synthesis of arsenosugar 22 described in this thesis. 4) The activation of arsinic acids with carbonyldiimidazole to form novel arsinoyl imidazolides. Ultimately, these compounds were difficult to isolate and were not readily reactive with Grignard or organolithium reagents as desired. The reaction of arsinoyl azolides with other nucleophiles is a continuing interest.
30
CHAPTER 2
SYNTHESIS OF ARSENOBETAINES VIA SODIUM DIMETHYLARSENIDE
31
2.1 INTRODUCTION
First determined by Cannon et. al. as a non-toxic arsenic contaminant in the
western rock lobster in 197743, arsenobetaine is the most common arsenic species found in marine food organisms and is rapidly excreted from the human body. In 2000,
Francesconi identified a second arsenobetaine in marine organisms by liquid chromatography – mass spectrometry.44
There is an increasing demand for these compounds not only for use as standards for routine identification and quantification but for subsequent biogenetic studies on arsenic metabolic pathways. However, arsenobetaines are unavailable commercially.
Several synthesis have been published,43,44,45,46 but all of these routes involve the precursor trimethylarsine. Trimethylarsine is the noxious vapor known as “Gosio-gas” responsible for epidemic poisonings during the 19th and 20th Centuries.47 Evidence of trimethylarsine gas generation from sheepskins intended for use in baby bedding has recently been reported as a cause of cot death or Sudden Infant Death Syndrome
(SIDS).48 Due to the high vapor pressure (BP 52oC) and toxicity of trimethylarsine, this compound is difficult and expensive to isolate. Trimethylarsine is further disinclined as a synthetic intermediate as it has recently been reported involved in an injurious laboratory
43 Cannon, J.R.; Edmonds, J.S.; Francesconi, K.A.; Raston, C.L.; Saunders, J.B.; Skelton, B.W.; White, A.H. Tetrahedron Letters, 1977, 18, 1543. 44 Francesconi, K.A.; Khokiattiwong, S.; Goessler, W.; Pedersen, S.; Pavkov, M. Chemical Communications, 2000, 1083. 45 Minhas, R.; Forsyth, D.S.; Dawson, B. Applied Organometallic Chemistry, 1998, 12, 635. 46 Legarde, F.; Asfari, Z.; Leroy, M.J.F.; Demesmay, C.; Ollé, M.; Lamotte, A.; Leperchec, P.; Maier, E.A. Fresenius Journal of Analytical Chemistry, 1999, 363, 12. 47 Thayer J.S. Organometallic Chemistry: An Overview. VCH Publishers: New York, 1988; 122. 48 Cullen, W.R. Sixth SIDS International Conference, Auckland, February 2000 32
explosion and low molecular weight arsines, such as trimethylarsine, are known to be
pyrophoric when exposed to the atmosphere.49
In seeking a synthesis of the arsenobetaines that avoids trimethylarsine, the novel
trivalent arsenic compounds ethoxycarbonylmethyl(dimethyl)arsine 4, n=1 and 2-
ethoxycarbonylethyl(dimethyl)arsine 4, n=2 were synthesized via sodium
dimethylarsenide. Several alternative routes to provide arsines were not explored.50,51
The arsines were then converted to quaternary arsonium salts 5 by the addition of methyl
iodide, which gave the corresponding arsenobetaines 6 by known procedure.43 Sodium dimethylarsenide52,53,54 3 was produced from dimethyliodoarsine52,55 2 which was obtained by the reduction of dimethylarsinic acid. This synthetic route is shown in
Scheme 2.1. Analytical techniques used to characterize and determine purity of synthetic
products include IR, 1H NMR, 13C NMR, ESI-MS, GC-MS and HPLC-ICP-MS.
O CH CH Br CO Et 3 3 2 CH3 HCl, H2O Na (CH2)n H C AsOH As NaAs 3 NaHSO , KI THF As CO2Et 3 H C I CH THF 3 3 H3C (CH2)n CH3
1 2 3 4
CH CH3 H C 3 - H C H3CI 3 Dowex-2 [OH ] 3 + - As+ CO Et As CO Et Toluene 2 2 H C (CH2)n H3C (CH2)n 3 56
Figure 2.1. New Route to Arsenobetaines.
49 Synthetic Methods of Organometallic and Inorganic Chemistry; Herrman, W., Ed.; Verlag: Stuttgart, 1996, 203. 50 Zhurnal Obsheki Khimii, 1934, 4, 192. 51 Tzschach, A.; Voigtländer, W. Journal of Organometallic Chemistry, 1977, 137, 31. 52 Feltham, R.D.; Kasenally, A.; Nyholm, R.S. Journal of Organometallic Chemistry, 1967, 7, 285. 53 Phillips, J.R.; Vis, J.H. Canadian Journal of Chemistry, 1967, 45, 675. 54 Doyle, R.J.; Salem, G.; Willis, A.C. Journal of the Chemical Society (Dalton). 1995, 1867. 55 Burrows, G.J.; Turner, E.E. Journal of the Chemical Society, 1920, 117, 1373. 33
2.2 EXPERIMENTAL
2.2.1 REAGENTS
HPLC grade solvents were used. Tetrahydrofuran, toluene and pentane were
purified further by distillation under argon from sodium and benzophenone for use in the
preparation of sodium dimethylarsenide. Dimethylarsinic acid was purchased from
Strem (Newburyport, MA, USA). Dowex 2 anion exchange resin was purchased from
Sigma (St. Louis, MO, USA) and was exercised with 5% HCl and 5% NaOH followed by
water wash. 18 MΩ cm-1 distilled, deionized water (Sybron Barnstead, Boston, MA) was used with this resin and washes. All other reagents were purchased from commercial vendors and used as received.
2.2.2 INSTRUMENTATION
Total arsenic was determined on an Agilent 7500 ICP-MS with the Agilent 1100 liquid chromatograph for speciation (Agilent Technologies, Palo Alto, CA, USA). The mass spectrometer employed was a Waters hybrid Q-TOF2 quadrupole time-of-flight mass spectrometer from Micromass (Manchester, U.K.) with nitrogen as the nebulizing gas and the Z Spray ion source. 1H and 13C-NMR spectra were obtained on a 250 MHz
AC-250 (Bruker, Billerica, Maryland. Infrared spectroscopy was done using a 1600
Series FT-IR (Perkin Elmer, Wellesley, Maryland). GC-MS were preformed on a HP
6890 Gas Chromatograph-Mass Spectrometer (Hewlett-Packard, Palo Alto, California) with GC sample introduction. The GC column used was a Zebron ZB-5 (Newport Beach,
California) with the following dimensions: 15 mm X 0.25 mm X 0.25 mm and was conditioned for use up to 320 oC. 34
2.3 SYNTHETIC PROCEDURES
All reactions and subsequent procedures were carried out in a well-ventilated
fume hood and gloves were worn when handling arsenic chemicals. All glassware that
came in contact with arsenic was rinsed in the fume hood prior to cleaning.
2.3.1 Dimethyliodoarsine (2)
Dimethylarsinic acid 1 (50 g, 362 mmol) was added to a 1000 mL Erlenmeyer
flask and dissolved in 200 mL of water. Potassium iodide (165 grams, 994 mmol) and
sodium bisulfite (5 g, 48 mmol) were added followed by rapid addition (2 min) of 250
mL of concentrated hydrochloric acid. The reaction was observed to be occurring as the
color turned yellow. This mixture was allowed to stir for 24 hours at room temperature
with occasional addition of sodium bisulfite (total of 23 grams, 221 mmol was added
over the course of the reaction) to maintain catalytic amounts and a yellow tinge to the
aqueous layer. The resulting yellow oil was separated and dried over calcium chloride.
Distillation gave 75 grams (90%) of 2 as a yellow oil. b.p. 149-151oC. 1H NMR (250
13 MHz, CDCl3): δ 2.01. C-NMR (66 MHz, CDCl3): δ 15.4. GC-MS: m/z calcd. for
+ C2H6AsI (M+H ) 231.9, found 232. FT-IR (neat): 2972.2 (m), 2904.7 (m) 2790.6 (w),
2471.4 (w), 1809.8 (w), 1407.6 (m), 1253.2 (m), 899.0 (m), 830.9 (m), 576.8 (m) cm-1.
These values were identical to literature values.52,55
35
2.3.2 Sodium Dimethylarsenide (3) A dry 1 liter three neck round bottom flask equipped with a water condenser and a
mechanical stirrer was flushed with argon and sodium (9.0 g, 390 mmol) was vigorously
stirred in 150 mL of refluxing toluene. After 2 hours, heating was discontinued and the
mixture was allowed to cool to room temperature with continued stirring. The toluene
was removed by cannulla and the resulting sodium granules were washed 3X with
pentane. These washes were removed by cannulla. 150 mL of THF was added and the
flask cooled by CO2/acetone bath.
Dimethyliodoarsine 2 (26.8g, 116 mmol) dissolved in 65 mL of THF was added with stirring over the course of one hour. The reaction was observed to occur as the mixture became golden and the sodium fused into a porous mass. The reaction was allowed to warm to room temperature over the course of several hours and then was refluxed for 30 minutes to produce a dark brown-green mixture. At this point, the condenser and mechanical stirrer were removed and replaced by a male/male u-joint coupled to a joint containing a course frit and leading into a second round bottom. The solution of sodium dimethylarsenide was separated from excess sodium and the sodium iodide byproduct under a positive argon pressure by filtration thru the frit. Clogging of the frit was avoided by slowly decanting, which allowed the solids to settle inside the top round bottom. Vacuum was applied to aid the solution through the frit. Sodium dimethylarsenide was not isolated further or characterized.
36
2.3.3 Ethoxycarbonylmethyl(dimethyl)arsine (4) n=1
Ethyl bromoacetate (19.3 g, 115 mmol) was slowly added (30 minutes) to the above green product 2 as the reaction was stirred and cooled by ice bath. This reaction was allowed to stir for two days under argon at room temperature. The resulting white residue (sodium bromide) was removed by filtration under argon and rinsed with de- gassed pentane. The combined solvents were concentrated and 4 was purified by fractional vacuum distillation as a clear oil (10.8 g, 51%). b.p. 61-68oC, 7 torr. 1H NMR
(250 MHz, CDCl3): δ 4.13 (q, J=7, 2H), 2.39 (s, 2H), 1.26 (t, J=7 Hz, 3H), 1.05 (s, 6H)
13 ppm. C NMR (66 MHz, CDCl3): δ 129.6, 131.8, 132.4, 133.2 ppm. HRMS: m/z
+ + calcd. for C6H14AsO2 (M+H ) 193.0210, found 193.0184. FT-IR (neat): 3418 (w), 2981
(m), 2911.3 (m), 2813 (w), 1721 (br s), 1445 (m), 1416 (m), 1389 (m), 1365 (m), 1250
(br s), 1149 (m), 1098 (s), 1035 (m) cm-1.
2.3.4 Ethoxycarbonylmethyl(trimethyl)arsonium iodide (5) n=1 This arsine 4, n=1 (9.8g, 51 mmol) was dissolved in 20 mL of toluene and an excess of methyl iodide (9.1g, 64 mmol) slowly added with stirring under argon.
Immediately, a white precipitate began to form. After 24 hours, 20 mL of pentane was added and the precipitate was collected by filtration and washed with pentane. The resulting white solid was dried and recrystallized from methanol to yield 15.5 g (99%) of ethoxycarbonylmethyl(trimethyl)arsonium iodide 5, n=1 as a white microcrystalline
1 precipitate. H NMR (250 MHz, D2O): δ 4.27 (q, J = 7 Hz, 2H), 3.64 (s, 2H), 2.03 (s,
13 9H), 1.30 (t, J=7 Hz, 3H) ppm. C NMR (66 MHz, D2O) δ 10.59, 15.94, 32.87,
+ + 66.37, 170.46 ppm. HRMS: m/z calcd. for C7H16AsO2 (M+H ) 207.0366, found 37
207.0382. FT-IR (KBr): 2983.6 (m), 1732.2 (s), 1423.6 (m), 1368.7 (m), 1349.8 (m),
1240.3 (s), 1193.2 (s), 1114.6 (m), 1054.4 (m), 1023.0 (m), 941.3 (s), 861.5 (m), 651.4
(m), 593.2 (w) cm-1. These spectroscopic data are identical to literature values.43
2.3.5 Arsenobetaine (trimethylarsoniumacetate) (6) n=1
Following the procedure described by Cannon et. al.43, a solution was made by
dissolving the iodide 5, n=1 (15g, 50 mmol) in water apply to 250 grams of Dowex 2
(OH-). Upon evaporation and drying, elution with 400 mL of water gave white crystals
which were dissolved in a minimum of hot ethanol (15 mL) and 200 mL of acetone was
added to initiate recrystalization. Argon was bubbled into the solution and the container
sealed. After 48 hours, the white crystals were removed by filtration and dried under high
1 vacuum to yield 5.58 g (64%) of arsenobetaine 6, n=1. H NMR (250 MHz, D2O) δ :
13 1.85 (s, 9H), 3.29 (s, 2H) ppm. C NMR (66 MHz, D2O) δ : 9.87, 32.81, 172.17 ppm.
+ + HRMS: m/z calcd. for C5H12AsO2 (M+H ) 179.0053, found 179.0040. FT-IR (KBr):
3005.1 (m), 2946.5 (m), 2890.2 (m), 1708.8 (br s), 1384.9 (s), 1357.8 (m), 1267.9 (m),
1253.3 (m), 1157.0 (s), 934.4 (s), 896.3 (s), 839.5 (m), 638.3 (m), 596.8 (m), 536.9
(w) cm-1. These spectroscopic data are identical to literature values.43
HPLC-ICP-MS showed arsenobetaine as the only arsenic-containing peak in the
chromatogram with >99% total recovery of arsenobetaine (as As). As arsenobetaine is
known to be deliquescent43, water is a likely impurity. The total yield for this synthesis from dimethylarsinic acid 1 is 33%.
2.3.6 2-Ethoxycarbonylethyl(dimethyl)arsine (4) n=2 38
The same procedure as for 4, n=1 was used with ethyl bromopropionate replacing
ethyl bromoacetate. 2-Ethoxycarbonylethyl(dimethyl)arsine 4, n=2 was isolated by
fractional vacuum distillation as a clear oil. b.p. 67-75oC, 7 torr. 1H NMR (250 MHz,
CDCl3): δ 4.14 (q, J = 7 Hz, 2H), 2.45 (t, J = 9 Hz, 2H), 1.68 (t, J =9 Hz, 2H), 1.26 (t, J =
13 7 Hz, 3H), 0.96 (s, 6H) ppm. C NMR (66 MHz, CDCl3): δ 9.13, 14.27, 22.52, 31.58,
+ + 60.52, 173.93 ppm. HRMS: m/z calcd. for C7H16AsO2 (M+H ) 207.0366, found
207.0375. FT-IR (neat): 3453 (w), 2977 (m), 2907 (m), 1735 (br s), 1419 (m), 1369 (m),
1341 (m), 1201 (m), 1150 (m), 1035 (m) cm-1. The yield for this reaction was 65%.
2.3.7 2-Ethoxycarbonylethyl(trimethyl)arsonium iodide (5) n=2
In similar fashion, treatment of the parent arsine 4, n=2 with methyl iodide gave
2-Ethoxycarbonylethyl(trimethyl)arsonium iodide 5, n=2 as white microcrystalline
1 precipitate in 99 % yield. H NMR (250 MHz, D2O): δ 4.20 (q, J = 7
Hz, 2H), 2.89 (t, J =8 Hz, 2H), 2.62 (t, J =7 Hz, 2H), 1.91 (s, 9H), 1.25 (t, J =7 Hz, 3H)
13 ppm. C NMR (66 MHz, D2O): δ 9.76, 15.54, 22.56, 22.96, 64.80 ppm. HRMS:
+ + m/z calcd. for C8H18AsO2 (M+H ) 221.0523, found 221.0539. FT-IR (KBr): 2984 (m),
1732 (s), 1424 (m), 1385 (m), 1369 (m), 1350 (m), 1240 (m), 1193 (m), 1115 (w), 1054
(m), 1023 (m) cm-1.
39
2.3.8 Arsenobetaine-2 (trimethylarsoniumpropionate) (6) n=2
Arsenobetaine-2 6, n=2 was isolated after treatment of the iodide 5, n=2 on
- 1 Dowex 2 (OH ). H NMR (250 MHz, D2O): δ 2.58 (t, J = 6 Hz, 2H), 2.45 (t, J =6 Hz,
13 2H), 1.80 (s, 9H) ppm. C NMR (66 MHz, D2O):
+ + δ 10.38, 25.23, 32.91, 181.97 ppm. HRMS: m/z calcd. for C6H14AsO2 (M+H )
193.0210, found 193.0182. FT-IR (KBr): 2994 (m), 2916 (m), 1593 (br s), 1384 (s), 1297
(m), 1187 (w) cm-1. The yield for this step was 59%. The total yield for the synthesis of
6, n=2 was 34%. These spectroscopic data are identical to literature values.44
2.4 DISCUSSION
Dimethyliodoarsine 2 was synthesized with slight modification from previous
procedures. Catalytic amounts of sodium bisulfite replaced the continuous bubbling of sulfur dioxide into the reaction. When molar equivalents of sodium bisulfite are used, a dangerous situation occurs when, during distillation, a salt forms inside the condenser, blocking the distillation pathway and creating a closed heated system. This salt is the acid salt 7. During heating, the acid salt 7, which is dissolved in the yellow oil product,
releases sulfur dioxide and water, which re-react with the product in the condenser to
form the salt.
CH3 H + - As [ HSO3 ] H3C I 7
Using only catalytic amounts of sodium bisulfite minimizes salt formation in the condenser. Because sodium bisulfite is in aqueous equilibrium with sulfur dioxide, sodium bisulfite must be added several times over the course of the reaction in order to 40
maintain catalytic amounts. As an indicator that the reaction is preceding and bisulfite is
present, the aqueous layer will be yellow. The reaction is at an end when further addition
of sodium bisulfite fails to turn the aqueous layer from clear to yellow.
The trivalent arsine compounds 4 must be kept under inert atmosphere because air exposure allows a reaction that produces dimethylarsinic acid 1. Fractional vacuum
distillation immediately prior to methylation removes these products along with
byproducts from the sodium reaction, i.e. tetramethyldiarsane and the arsonium
compounds resulting from the double addition of the bromide esters. While the synthesis
can continue without distillation, this stage provides the best opportunity to remove these
side products. Without distillation, these impurities lower the recovery during subsequent
recrystalization steps and impurities are manifest in the final arsenobetaine compounds as
identified by additional arsenic peaks obtained by HPLC-ICP-MS chromatogram.
41
CHAPTER 3
ARSINE OXIDES REVISITED
42
3.1 INTRODUCTION
The title compounds have been of interest since 1854, when triethylarsine oxide
was reported obtained from the slow evaporation of an ether solution of triethylarsine.56
This inauguration of arsine oxide chemistry now appears premature. The authors most likely misidentified their water insoluble product as triethylarsine oxide which is in fact extremely hygroscopic. In 1966, Zingaro et al. found that exposing arsines to the atmosphere produces conversion to the arsinic acids.57 This conversion of arsines has also been observed in this laboratory. Inconsistencies in earlier work on these compounds were, in part, the result of the misassignment of arsinic acids as arsine oxides.
The difficulty in preparing arsine oxides was overcome by Zingaro with the oxidation of arsines using mercuric oxide or a solution of 30% aqueous hydrogen peroxide in ether under an inert atmosphere to prevent contact with molecular oxygen.
Using this procedure, trimethylarsine oxide (TMAO) has been synthetically produced by reaction of trimethylarsine with aqueous hydrogen peroxide in tetrahydrofuran. TMAO is an ubiquitous form of organic arsenic and is regularly determined in the environment and in foods. The Zingaro procedure had been used for forty years in the production of TMAO on large scales up to 25 grams without incident.
THERE ARE MAJOR PROBLEMS WITH THIS PROCEDURE AND HYROGEN
PEROXIDE OXIDATION OF ARSINES SHOULD BE IMMEDIATELY
ABANDONED!
A recent damaging explosion occurred during this reaction causing injury and equipment destruction. An experienced researcher and experimentalist following standard
56 Landolt, H. Annalen der Chemie und Pharmacie, 1854, 89, 301. 57 Merijanian, A.; Zingaro, R.A. Inorganic Chemistry, 1966, 5, 187. 43
procedures was standing near a rotary evaporator containing the reaction product when it
violently exploded cutting him badly in the face and neck with flying glass. Fortunately
he was wearing eye protection and a lab coat. Possible causes for this explosion are
speculative. This accident has prompted the total reevaluation of synthetic arsine oxide
techniques.
The recent synthesis of the arsine oxide containing sugar 22 shown in Figure 1 is
the structurally simplest representative of a family of arsenosugars that have been
identified in food sources. This synthesis included a reluctant step involving the oxidation of a parent arsine precursor with hydrogen peroxide in THF for lack of an alternate route.58 Risk assessment and arsenic exposure concerns for arsenic-containing food sources have generated toxicokinetic and biogenesis interest in these arsenosugars creating demand for gram quantities of these compounds and hastened the need for alternate routes to arsine oxides. Other naturally occurring arsine oxides including trimethylarsine oxide, 2-dimethylarsinoylethanol, and dimethylarsinoyl acetate have been in demand subsequent to the accident.
An exhaustive search of the literature did provide several alternative routes to arsine oxides. The dehydration of (HO)2AsPh3 has been reported to give triphenylarsine
oxide.59 Arsine oxides are also available from dihaloarsanes.60 Tetraalkylarsonium compounds have been converted to arsine oxides by treatment with bromine water.61
Finally catalytic oxidation of simple arsines has been reported with various complexes of
58 Stick, R.A.; Stubbs, K.A.; Tilbrook, D.M.G. Australian Journal of Chemistry, 2001, 54, 181. 59 Zuckerkandl, F.; Sinai, M. Berichte der Deutschen Chemischen Gesellschaft., 1921, 54, 2484. 60 Zuckerland, F.; Sinai, M. Berichte der Deutschen Chemischen Gesellschaft, 1921, 54, 2484. 61 Benson, A. A.; Proceedings of the National Academy of Sciences, 1989, 86, 6131. 44
ruthenium, rhodium and iron.62 Triphenylarsine has been converted to its oxide by
63 reaction with the harsh oxidant KMnO4. Investigations into these routes revealed an
entirely new route would be necessary for use in the synthesis of arsenosugars. The goal
has been two-fold - the development of an alternative attachment for the arsine oxide to
sugars and the development of a general method for the synthesis of simple arsine oxides
such as TMAO that do not involve an oxidation.
3.2 CAUSE OF EXPLOSION64
The cause of the accident has been examined at length. The reaction had been
done many times before and the explosion was not the result of a scale-up problem because greater quantities had been used in the past. The possibility of the formation of
ether peroxides had been considered and tetrahydrofuran, freshly distilled from sodium,
was used throughout the procedure. Because of the possibility of problems with excess hydrogen peroxide, the reaction solution had been diluted with water and warmed to decompose any remaining hydrogen peroxide.
After the reaction, the product was extracted into an aqueous layer and washed
with fresh ether to remove any ether peroxides that were present. Care was taken to
remove some water with each wash. Finally, water was removed on a Rotavap using a
rotary pump to reduce pressure. The solution was not taken to dryness, which is the
source of most problems with ether peroxides. With 40 mL remaining, the Rotavap was
destroyed by the considerable force of the explosion. While the formation of organic
62 Vančová, V.; Ondrejkovičová, I. Collection of Czechoslovak Chemical Communications, 1991, 56, 2869. 63 National Academy of Science Letter (India), 1985, 8, 271. 64 Extensive discussions with Professor William Cullen at the University of British Columbia have included review of original accounts of the accident as reported to the university. I am deeply indebted to Dr. Cullen for access to this material. 45
peroxides can not be discounted, experimental precautions would seem to rule out this
explanation.
The presence of trimethylarsine in the accident mixture presented it own set of
hazards. Trimethylarsine is the noxious vapor known as “Gosio-gas” responsible for
epidemic poisonings during the 19th and 20th Centuries.65 Due to the high vapor pressure
(BP 52oC) and exceeding toxicity of trimethylarsine, this compound is difficult and expensive to isolate and manipulate. Low molecular weight arsines, such as trimethylarsine, are also known to be spontaneously flammable when exposed to the
atmosphere.66 At the point of explosion, trimethylarsine would only remain at trace
amounts because the hydrogen peroxide oxidation is very effective. Because the accident
under consideration involved a true explosion (one generating a shockwave) and was not simply the mixture bursting into flames, the trimethylarsine starting material can be ruled out as the sole cause.
Two theories have been advanced to explain the cause of this explosion. Peroxy
derivatives of arsenic are essentially unknown; however, the antimony derivative
Me3Sb(OOH)2 was synthesized using the same hydrogen peroxide/ ether treatment and on one occasion exploded with similar results.67 This compound was synthesized from trimethylstibine using the same hydrogen peroxide/ ether treatment. One example of an organophosphorus peroxide has also been reported.68
65 Thayer J.S. Organometallic Chemistry: An Overview. VCH Publishers: New York, 1988; 122. 66 Synthetic Methods of Organometallic and Inorganic Chemistry; Herrman, W., Ed.; Verlag: Stuttgart, 1996, 203. 67 Dodd, M.; Grundy, S.L.; Reimer, K.J.; Cullen, W.R. Applied Organometallic Chemistry, 1992, 6, 207. 68 Danney, R.L.; Waller, R.L. Journal of the American Chemical Society, 1965, 87, 4805. 46
A second theory has developed with the discovery of a 1968 report of an adduct
between triphenylarsine oxide and hydrogen peroxide.69 Crystals of this compound were stable and decomposed by loosing hydrogen peroxide at 135oC to yield the oxide. The formation of an analogous adduct between trimethylarsine oxide and hydrogen peroxide may be the culprit behind the accident. Crystals of the TMAO analog would be more compact and might not share the stability observed for crystalline triphenylarsine oxide – hydrogen peroxide. The observation of tertiary arsine oxides as active and selective catalysts for organic epoxidation with hydrogen peroxide provides further evidence for a strong association of arsine oxides with hydrogen peroxide.70
3.3 DIOXIRANE – AN ALTERNATVE OXIDATION
Dimethyldioxirane is an oxygen transfer reagent and has found widespread use in organic synthesis since its discovery in 1985.71 The use of this dioxirane has been developed as an alternative to the synthetically useful oxidation of arsines with hydrogen peroxide. Dioxiranes are cyclic peroxides capable of oxygen transfer to a wide range of compounds under mild conditions.72 Dimethyldioxirane (DMD) is close to an ideal
oxidant because it is selective, mild, efficient and can be readily made from commercially
available starting materials.
Dioxirane is a milder reagent than hydrogen peroxide and can not participate in
the chemistry of hydrogen peroxide proposed responsible for the recent accident.
Hydrogen peroxide can react via two pathways – oxidation by oxygen transfer and
69 Howell, G.V.; Williams, R. L. Journal of the Chemical Society (A), 1968, 117. 70 van Vliet, M.C.A.; Arends, I.W.C.E.; Sheldon, R.A. Tetrahedron Letters, 1999, 40, 5239. 71 Murray, R.W.; Jeyaraman, R. Journal of Organic Chemistry, 1985, 50, 2847. 72 Murray, R.W. Chemical Reviews, 1989, 89, 1187. 47
addition of an –OOH directly. Because dioxirane is only capable of a single oxygen transfer, the only way to form the peroxide would be the two-step process of oxidizing the arsine to the oxide and the subsequent oxidation of the oxide to the peroxide. This process is unlikely when only one equivalent amount of DMD is utilized. As a final advantage, formation of the hydrogen peroxide adduct is impossible because this oxidant is not present. Arsine oxidations by dimethyldioxirane are now reported in Chapters Four and Five of this thesis.
3.4 MEYER REACTION
The prospect of using the Arbuzov-Michaelis or Michaelis-Becker reaction for the
synthesis of arsine oxides proved futile but did lead to the consideration of Meyer
chemistry.73 The reaction developed by Meyer74 in 1883 has found extensive use in the synthesis of alkylarsonates and dialkylarsinic acids75 but its usefulness in the preparation of arsine oxides has been largely overlooked. The Meyer reaction for the preparation of
TMAO in excellent (80%) yield has recently been reported in collaboration with
Professor William Cullen (University of British Columbia, Vancouver, Canada).76 The
Meyer reaction for TMAO has two main advantages in that it avoids the perils of an oxidation step and trimethylarsine as a precursor. The X-ray structure of the acid salt resulting from the Meyer synthesis is shown in Figure 3.2. This salt is converted to
73 Abalonin, B.E. Russian Chemical Reviews, 1991, 60, 2593. 74 Meyer, G. Berichte der Deutschen Chemischen Gesellschaft, 1883, 16, 1439. 75 Schmeisser, E.; Gessler, W.; Kienzl, N.; Francesconi, K.A. Analytical Chemistry, 2003; ASAP Web Release Date 06-Dec-2003. 76 Fricke, M.; Sun, H.; Creed, J.; Thayer, J.; Caruso, J.; Cullen, W. The Chemistry of Arsine oxides Related to the Synthesis of Arsenosugars presented at the 6th International Conference on Environmental and Biological Aspects of Main-Group Organometallics, Université de Pau et des Pays de l’Adour, Pau, France 2003 (Lecture 8). 48
TMAO by anion exchange chromatography and highly pure TMAO via Meyer chemistry
is available by sublimation.
+ OH - I H3C As CH3
CH3 . Figure 3.1. Structure of TMAO HI.
3.5 ARSINOYL IMIDAZOLIDES
The preparation of imidazolides of phosphoric acids for the study of high-energy phosphates has been described77 and the use of arsinoyl imidazolides has been
investigated for the synthesis of arsine oxides.76 The activation of dimethylarsinic acid with carbonyldiimidazole is shown in Figure 3.2. The reaction proceeds rapidly in acetonitrile as the emission of CO2 bubbles is observed. However, the reactive product, dimethylarsinoyl imidazolide, is not easily isolatable nor is suitable for alkylation by reaction with Grignard or organolithium reagent. The insolubility of this compound in
1,4-dioxane, tetrahydrofuran or diethyl ether precludes formation of arsine oxides and yields of the desired compounds are less than 10%. The reaction of the broad class of arsinoyl azolides with other nucleophiles, primarily thiols, is a continuing process.
77 Cramer, F.; Schaller, H.; Staab, H.A. Chemische Berichte, 1961, 94, 1612. 49
O C NNN N O H C O H3C 3 As As H CCN N OH 3 H C H3C 3 N N NH
CO2 Figure 3.2. Activation of Dimethylarsinic acid.
50
CHAPTER 4
SYNTHESIS OF TRIPHENYLARSINE OXIDE BY OXIDATION OF TRIPHENYLARSINE WITH DIMETHYLDIOXIRANE
51
4.1 INTRODUCTION
Although all oxidation reactions have the potential to explode, dioxiranes are
capable of oxygen transfer to a wide range of compounds under mild conditions.78 A single report describing an accident involving dioxiranes was not discovered despite
extensive searching. Dimethyldioxirane (DMD) is the cyclic peroxide analog of acetone
and is close to an ideal oxidant for arsines because it is selective, mild, efficient and can
be readily made from commercially available starting materials.79 Dioxirane has been described as “the wonder oxidant”80 and is reported to make oxidation chemistry easy.81
. . DMD is produced in situ by the reaction of Caroate (2KHSO5 KHSO4 K2SO4) with acetone. Caroate is a readily available reagent sold under the commercial name Oxone.
The oxidation of arsines by dioxiranes is unknown although several oxidations involving heteroatoms have been reported. Dioxirane oxidation of sulfur-containing organic compounds has been reported by Murray et al.82 Ethyl(methyl)dioxirane has recently been reported as an efficient reagent for the oxidation of nucleoside phosphites
83 into phosphates. The reaction should proceed by SN2 nucleophilic attack of the arsine
electron pair on the dioxirane peroxide bond as shown in Figure 4.1.84
78 Murray, R.W. Chemical Reviews 1989, 89, 1187. 79 Personal communication with Professor David Hart, The Ohio State University, 2003. 80 Waldemar, A. (ed.) Peroxide Chemistry, Wiley: Weinheim, Germany, 2000, p. 25 81 Adam, W.; Hadjiarapoglou, L.; Dioxiranes: Oxidation Chemistry Made Easy. In Topics in Current Chemistry v164, Waldemar, A. (ed.), Springer-Verlag: Berlin, 1993. 82 Murray, R.W.; Jeraraman, R.; Pillay, M.K. Journal of Organic Chemisry, 1987, 52, 746. 83 Kataoka, M.; Hattori, A.; Okino, S.; Hyodo, M.; Asano, M.; Kawai, R.; Hayakawa, Y. Organic Letters, 2001, 3, 815. 84 Adam, W.; Golsch, D.; Görth, F.C. Chemistry: A European Journal, 1996, 2, 255. 52
H3C
O CH3 O
As CH Ph + 3 Ph Ph3 As O Ph3AsO Ph - O + acetone CH3
Figure 4.1. Mechanism of Dioxirane Oxidation of Arsines.
Because DMD is typically used as its acetone solution, the oxidation of thioanisole 10 by reaction with DMD shown in Figure 2.2 has been used to determine the reactive equivalent of dioxirane solution.85 The characteristics of this reaction make it ideal for DMD concentration determination. The reaction is well studied and is quantitative and rapid. The oxidation product, phenylmethylsulfoxide 11, is volatile and
GC-MS has been used to quantify the conversion of 10 to 11. Because it is possible to further oxidize the product 11 to its sulfone, thioanisole is used in large excess (>5X).
The difference in chemical shift of methyl groups also allows the reaction products to be
1 characterized by H NMR integration of the corresponding methyl singlets.82 Monitoring
the ratio of the areas of these singlets provides a good approximation of the conversion
yield of the reaction.
85 Murray, R.V.; Singh, M. Organic Synthesis, 199, 74, 288. 53
CH3 O O CH3 S S O CH3 CH3
10 11
Figure 4.2. Dioxirane Oxidation of Thioanisole.
4.2 EXPERIMENTAL
4.2.1 REAGENTS
Triphenylarsine was purchased from Strem (Newburyport, MA, USA). Oxone
was purchased Aldrich (Milwaukee, Wisconsin). Deionized distilled water 18 MΩ water
(Millipore, Bedford, MA) was used throughout the experiment. All other reagents were
purchased from commercial vendors and used as received.
4.2.2 INSTRUMENTATION
1H and 13C-NMR spectra were obtained on a 250 MHz AC-250 (Bruker, Billerica,
Maryland). Infrared spectroscopy was done using a 1600 Series FT-IR (Perkin Elmer,
Wellesley, Maryland). High-resolution mass spectrometry was performed on a Fourier
Transform Ion Cyclotron Resonance Mass Spectrometer (FTICRMS) with a 4.7 T FTICR magnet equipped for electrospray (IonSpec, Lake Forest, California). GC-MS were preformed on a HP 6890 Gas Chromatograph-Mass Spectrometer (Hewlett-Packard, Palo
Alto, California) with GC sample introduction. The GC column used was a Zebron ZB-5 54
(Newport Beach, California) with the following dimensions: 15 mm X 0.25 mm X 0.25
mm and was conditioned for use up to 320 oC.
4.3 SYNTHETIC PROCEDURES
These reactions were carried out in a fume hood and gloves were worn when
handling arsenic chemicals. Any glassware that came in contact with arsenicals was
rinsed into waste in the fume hood prior to cleaning. Glassware was cleaned by soaking
in an alcohol/ potassium hydroxide bath, soap scrubbing as necessary followed by 18 Ω
water and acetone rinse. Because of the risk of explosion in attempting an untested
means for oxidizing arsines, an explosion shield was erected during the full course of this
reaction and handling was kept to a minimum. The small size of the attempted reaction
was an additional precaution designed to minimize the risk of an explosion.
4.3.1 Dimethyldioxirane (DMD) (9)
DMD was prepared by a method described by Adam. 81 Water (170 mL), acetone
(130 mL, 1800 mmol) and sodium bicarbonate (39 g, 464 mmol) were stirred in a three-
neck 250 mL round bottom flask in an ice bath at 0 oC. Oxone (80 g, 130 mmol) was added in four portions at three minute intervals. Following the final addition, the ice bath was removed and the reaction flask was left open to the air. A two-neck 250 mL round
bottom flask was cooled by dry ice/ acetone bath to -78 oC and connected to the reaction flask by high density polyethylene tubing. Distillation was accomplished by the steady pull of air through the reaction flask into the second flask by connection to the house vacuum. Condensation gave 46 mL of 9 as its acetone solution. The molarity of this 55
solution was determined to be 0.02 M by reaction with 10 and analysis by GC-MS as described in the following section. The yield for this reaction was 0.7 % from oxone.
4.3.2 Determination of the Concentration of DMD (9)
Thioanisole 10 (20 µL, 0.17 mmol) was transferred by automatic pipet into a glass sample jar and 1 mL of a solution of 9 was added and stirred for 4 hours. The sample was diluted 1:100 and the GC-MS analysis of a 1 µL injection gave a 9.8% conversion of
10 to 11 by integration of the peak M+. = 124 for 10 eluting at seven minutes and M+. =
140 for 11 eluting at nine minutes. This GC-MS analysis is shown in Figure 4.3 and the concentration of 9 was determined to be 0.02 M. This solution was used immediately in the following reaction.
GC-MS Analysis of the reaction of 9 with 10
Abundance Ion 124.00 (123.70 to 124.70): THIOAN41.D (+,-)
100000
90000
80000
70000
60000
50000
40000
30000
20000
10000
0 Time--> 6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 Abundance Ion 140.00 (139.70 to 140.70): THIOAN41.D (+,-)
10000 9000 8000 7000 6000 5000 4000 3000 2000 1000 0 Time--> 6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00
Figure 4.3. GC-MS Analysis of Conversion of Thioanisole to Phenylmethylsulfoxide.
56
4.3.3 Triphenylarsine oxide (12)
Behind an explosion shield, triphenylarsine 8 (0.187g, 0.61mmol) was dissolved
and stirred in a 250 mL round bottom flask with 32 mL of the 0.02M DMD solution
described above (1.05 equivalents). The flask was filled with argon and covered with
Teflon tape and stirred for four hours. At this point, a slow stream of argon was used to
remove volatiles over the course of the night. The next day, the reaction was cautiously
approached. NOTE: Most peroxide explosions occur after an evaporation step and
this point of the procedure includes the highest risk. Recrystalization of the resulting
white residue from a minimum amount of acetone followed by the addition of water (2
mL) gave 12 as fine white crystals (146 mg, 74%) after drying in an oven at 160 oC. m.p.
o 1 196-200 C. H NMR (250 MHz, CDCl3): δ 7.48−7.62 (m, 9H), 7.71-7.74 (m,6H) ppm.
13 + C NMR (CDCl3): δ 129.6, 131.8, 132.4, 133.2 ppm. HRMS: m/z calcd. for (M+H )
+ C18H15AsOH 323.0417, found 323.0344. FT-IR (KBr): 3399.4 (br m), 1663.6 (w),
1439.6 (m), 1384.6 (s), 1087.0 (m), 882.0 (s), 743.7 (m), 691.1 (m), 472.7 (m) cm-1.
These spectroscopic data are identical to literature values.62,63,69
4.4 DISCUSSION
The synthesis of triphenylarsine oxide by oxidation of triphenylarsine with
dimethyldioxirane worked as expected and demonstrates a safer means for the oxidation
of arsines than previously reported. Determination of DMD concentrations by GC-MS
analysis of the reaction product of DMD with thioanisole was more consistent than when
concentration was determined by peak integration in NMR. Because all oxidation 57 reactions have the potential to explode, this reaction should only be attempted by qualified individuals using proper safety precautions.
58
CHAPTER 5
SYNTHESIS OF (R)-2,3-DIHYDROXYPROPYL 5-DEOXY-DIMETHYLARSINOYL-β-D-RIBOSIDE (As328)
59
5.1 INTRODUCTION
The identification two decades ago of arsenic-containing ribosides in
macroalgae86 has generated intense interest in these species. These arsenosugars are now believed to be widespread in the marine ecosystem and to play a vital role in the biotransformation of arsenic by marine organisms.87 Ongoing research involving arsenic- containing carbohydrates is multi-faceted and has stretched across several disciplines.
Arsenic speciation analysis for arsenosugars is important for developing better
dietary exposure estimates and to distinguish between toxic and nontoxic forms. Several
studies have sought to provide arsenosugar species specific data on a variety of foods.
McSheehy et al. determined 15 organoarsenic species including 8 ribofuranosides in the
kidney of the clam (Tridacna derasa) by multidimensional HPLC-ICP-MS and HPLC-
ESI-MS. Creed et al. determined the molecular weight of three arsenosugars found in
ribbon kelp using ion chromatography-electrospray ionization-mass spectrometry.88
Schmeisser et al. have experimented with hydride generation and the resulting volatile
analytes formed from arsenosugars and the implications for arsenic speciation analyses.89
Recently, arsenosugars have been determined in fresh water algae and in terrestrial plants.90 Although fifteen arsenosugars have been isolated and identified as natural products91, the four arsenosugars shown in Figure 5.1 comprise the major arsenic-
containing carbohydrates determined to occur in nature.
86 Edmonds, J.S.’ Francesconi, K.A., Nature, 1981, 18, 1543. 87 Edmonds, J.S.; Francesconi, K.A.; Stick, R.V. Natural Products Report, 1993, 10, 43. 88 Gallagher, P.A.; Wei, X.; Shoemaker, J.A; Brockhoff, C.A.; Creed, J.T. Journal of Analytical Atomic Spectroscopy, 1999, 14, 1829. 89 Schmeisser, E.; Goessler, W.; Kienzl, N. Francesconi, K.A. Analytical Chemistry, in press. 90 Francesconi, K.A.; Kuehnelt, D.; Environmental Chemistry of Arsenic; Frankenberger, W.T. (ed.); Marcel Dekker: New York, 2002, Chapter 2. 91 Francesconi, K.A.; Edmonds, J.S.; Arsenic in the sea. In Oceanography and Marine Biology, An Annual Review; Ansell, A.D.; Gibson, R.N.; Barnes, M. (eds.); UCL Press: London, 1993. 60
Relatively little work has been done to determine the toxicity of arsenosugars due
largely to the unavailability of even microgram amounts of these compounds. In an
attempt to address a need for gram amounts, the total synthesis of the first arsenosugar
was first reported in 1987.92 An alternate route with significantly higher yields of this
compound was reported in 199693 and a hybrid of the two routes in 2001.94 Microgram amounts of five trimethylarsoniorobosides have also been synthesized by derivation of dimethylarsinoylribosides isolated from natural sources.95
Four Major Arsenosugars
O O
Me2As Me2As O SO3H O OSO3H O O OH OH
OH OH OH OH
O O
Me As O 2 Me2As O OH O O P O O O OH OH HO OH OH
OH OH OH OH
Figure 5.1
The synthesis of arsenosugar (As328) 22 shown in Figure 5.2 was necessitated by
ongoing studies concerning the metabolic fate and biodegradation observed upon
digestion of this compound.
92 McAdam, D.P.; Perera, A.M.A.; Stick, R.V. Australian Journal of Chemistry, 1987, 40, 1901. 93 Liu, J.; O’Brien, D.H.; Irgolic, K.J. Applied Organometallic Chemistry, 1996, 10, 1. 94 Stick, R.V.; Stubbs, K.A.; Tilbrook, D.M.G. Australian Journal of Chemistry, 2001, 54, 181. 95 Francesconi, K.A.; Edmonds, J.S.; Stick, R.V. Applied Organometallic Chemistry, 1994, 8, 517. 61
Figure 5.2. Synthesis of As328. 62
5.2 EXPERIMENTAL
5.2.1 REAGENTS
HPLC grade solvents were used. Tetrahydrofuran (THF) was distilled under
nitrogen from sodium and benzophenone prior to use. Dimethylarsinic acid was purchased from Strem (Newburyport, MA, USA) and was >98% pure. Buffers, eluents, standards and washes were prepared using distilled deionized 18 MΩ water (Millipore,
Bedford, Massachusetts), distilled water. D-(-)-Ribose 13 was purchased from Acros
(Morris Plains, New Jersey) and was >99.5% pure. The chiral alcohol, (S)-(-)-1,2-Di-O- benzylglycerol 17, was purchased from Aldrich (Milwaukee, Wisconsin) and was 96%
pure. All other reagents were purchased from commercial vendors and used as received.
5.2.2 INSTRUMENTATION
1H and 13C-NMR spectra were obtained on either a 250 MHz AC-250 (Bruker,
Billerica, Maryland) or a 400 MHz Avance (Bruker). Infrared spectroscopy was done
using a 1600 Series FT-IR (Perkin Elmer, Wellesley, Maryland). GC-MS were
preformed on a HP 6890 Gas Chromatograph-Mass Spectrometer (Hewlett-Packard, Palo
Alto, California) with GC sample introduction. The GC column used was a Zebron ZB-5
(Newport Beach, California) with the following dimensions: 15 mm X 0.25 mm X 0.25 mm and was conditioned for use up to 320 oC.
The HPLC used in this study was a HP 1100 (Agilent, Bellvue, Washington).
The ICP-MS was a HP 4500 ICP-MS (Agilent) with the instrumental operating
parameters as follows: rf power = 1210 W, carrier gas flow = 1.26 L min –1, plasma gas 63
flow = 15 L min –1, and the spray chamber temperature was 5 oC. The IC and ICP-MS
were fully integrated.
5.2.3 CHROMATOGRAPHY
Silica gel (chromatographic grade) used for flash chromatography was purchased from Fisher and was 60-200 mesh from Lot #034281. Thin Layer Chromatography
(TLC) was performed on glass plates coated with silica gel with a UV indicator at 254 nm and were purchased from Aldrich. TLC were visualized either with a UV lamp, in an iodine vapor chamber or by spraying with a solution containing 0.5g of Ce(SO4)2, 1.25g
96 of (NH4)6Mo7O24 in 50 mL of 10% H2SO4. Sprayed plates were then developed by
heating from a hot air gun.
The HPLC column used for cation chromatography was an Ionospher C (Varian,
Palo Alto, California). The mobile phase was 10 mM pyridinium formate with the pH adjusted to 2.7 with formic acid. The flowrate and injection volume were 1 mL/min and
100 µL respectively. The identification of the final sugar 22 was made by retention time match to a standard isolated from ribbon kelp (Alaria marginata).97 Amberlite IRA-400 anion exchange resin used for ion exchange chromatography was purchased from Sigma
(St. Louis, Missouri) and was exercised with 100 mL of 5% HCl and then 100 mL of 5%
NaOH followed by water wash until neutral.
96 Skaanderup, P.R.; Poulsen, C.S.; Hyldtoft, L.; Jørgensen M.R.; Madsen, R. Synthesis, 2002, 12, 1721. 97 Gamble, B.M.; Gallagher, P.A.; Shoemaker, J.A.; Wei§, X.; Schwegel, C.A.; Creed, J.T. The Analyst, 2002, 127, 781. 64
5.3 SYNTHETIC PROCEDURES
All reactions and subsequent procedures were carried out in a well-ventilated
fume hood and gloves were worn when handling arsenic chemicals. Any glassware that
came in contact with arsenicals was rinsed into waste in the fume hood prior to cleaning.
Glassware was cleaned by soaking in an alcohol/ potassium hydroxide bath, scrubbing
with Contrad 70 (Curtisn Matheson Scientific, Houston, Texas) as necessary followed by
18 Ω water and acetone rinse. Glassware used for moisture sensitive reactions were dried in an oven at 160 oC. All synthetic products were prepared following the procedure
described by Stick94 with the following exceptions: arsine 20 was oxidized to the arsine
oxide 21 using dimethyldioxirane 9 not 30 % aqueous hydrogen peroxide in
tetrahydrofuran as reported. The bromoriboside 16 was synthesized from tosylate 15
using potassium bromide instead of tetrabutylammonium bromide and was purified by
fractional vacuum distillation instead of by preparative chromatography.
5.3.1 Methyl 2,3-O-Isopropylidene-β-D-riboside (14)
D-(-)-Ribose 13 (27 g, 180 mmol), acetone (100 mL), methanol (100 mL), and concentrated HCl were heated to reflux for 3 hours. The resulting orange oil was cooled to room temperature and poured into water (270 mL). Chloroform (400 mL) was used to extract the product. The combined organic layers were dried over Na2SO4 and the solvent removed by rotary evaporator leaving 14 as a light orange oil (30.5 g, 83 %). 1H
NMR (250 MHz, CDCl3): δ 1.32 (s, 3H), 1.49 (s, 3H), 3.21-3.25 (dd, 1H), 3.44 (s, 3H),
3.61-3.69 (m, 2H), 4.43, (d, 1H), 4.59 (d, 1H), 4.84 (d, 1H), 4.98 (s, 1H) ppm. FT-IR
(neat): 3466 (br s), 2988 (s), 2940 (s), 2835 (m), 2162 (w), 2025 (w), 1952 (w), 1739 (w), 65
1690 (w), 1459 (m), 1374 (s), 1273 (m), 1240 (m), 1210 (s), 1161 (m), 1093 (s), 1042 (s),
1009 (m), 962 (m), 870 (m), 761 (m), 649 (m), 580 (m), 516 (m) cm-1. These spectroscopic data are identical to literature values.98
5.3.2 Methyl 2,3-O-Isopropylidene-5-O-toluenesulfonyl-β-D-riboside (15)
To the above oil 14 (30.5 g, 149mmol), pyridine (54 mL) was added and cooled in
an ice bath. p-Toluenesulfonyl chloride (35 g, 184 mmol) was added and the was
allowed to warm to room temperature and was stirred overnight. 2.7 mL of water was
added and stirred for 30 minutes. Chloroform (100 mL) was added and the solution was
washed with a 0.05 M H2SO4 solution (160 mL) followed by three washes with 0.2 M
NaOH solution (170 mL). The organic layer was dried over Na2SO4 and an orange oil was obtained after the removal of solvent. Recrystalization from ethanol (60 mL) gave
o 1 15 as a white solid (31.6 g, 59%). m.p. 82-83 C. H NMR (250 MHz, CDCl3): δ 1.29 (s,
3H), 1.45 (s, 3H), 2.46 (s, 3H), 3.24 (s, 3H), 4.02 (d, 2H), 4.31 (t, 1H), 4.51-4.60 (m, 2H),
4.93 (s, 1H), 7.36 (d, 2H), 7.81 (d, 2H) ppm. FT-IR (KBr): 3421 (w), 2947 (m), 1597
(m), 1452 (w), 1359 (s), 1273 (m), 1247 (m), 1212 (s), 1180 (s), 1092 (s), 1044 (m), 1021
(m), 970 (s), 960 (s), 903 (m), 867 (m), 854 (m), 834 (m), 813 (m), 682 (m), 664 (m), 557
-1 (s), 524 (m) cm . These spectroscopic data are in agreement with literature values.94
98 Lerner, L.M. Carbohydrate Research, 1977, 53, 177. 66
5.3.3 Methyl 5-Bromo-5-deoxy-2,3-O-isopropylidene-β-D-riboside (16)
The tosylate 15 (28.7 g, 80 mmol) was combined with sodium bromide (14.4 g,
140 mmol) and refluxed for two hours in N,N-dimethylformamide (DMF) (280 mL). The color changed to purple. The mixture was allowed to cool to room temperature and was stirred overnight. DMF was removed by evaporation leaving a brown solid. Fractional vacuum distillation gave 16 as a nearly colorless oil (9.83 g, 46%). b.p. 105-115 oC, 3
1 torr. H NMR (250 MHz, CDCl3): δ 1.33 (s, 3H), 1.49 (s, 3H), 3.28-3.45 (m, 2H), 3.37 (s,
3H), 4.39 (m, 1H), 4.61 (m, 1H), 4.77 (m, 1H), 5.02 (s, 1H) ppm. FT-IR (neat): 2989
(m), 2938 (m), 2835 (m), 2061 (w), 1730 (w), 1681 (w), 1441 (m), 1374 (m), 1272 (m),
1240 (m), 1210 (m), 1195 (m) 1162 (m), 1106 (br s), 1074 (s), 1053 (m), 1029 (m), 974
(m), 959 (m), 931 (m), 869 (s), 819 (m), 775 (w), 680 (w), 652 (w), 581 (w), 515 (w)
-1 cm . These spectroscopic data are in agreement with literature values.94
5.3.4 (R)-2,3-Dibenzyloxypropyl 5-Bromo-5-Deoxy-2,3-O-isopropylidene-β-D-
riboside (18)
The above bromoriboside 16 (4 g, 15 mmol) was combined with (S)-(-)-1,2-Di-O- benzylglycerol 17 (4.3 g, 16 mmol) and p-toluenesulfonic acid monohydrate (0.33 g, 1.7 mmol) in a 50 mL round bottom flask and heated to 80 oC. The reaction was sealed at 30
torr and stirred at 80 oC for 20 hours. Chloroform (100 mL) was added and washed with
60 mL of a saturated sodium bicarbonate solution. The solution was dried over Na2SO4 and after evaporation a dark oil remained. Flash chromatography (650 mL 12:1 petroleum ether: ethyl acetate - then 550 mL 10:1 petroleum ether: ethyl acetate – finally
800 mL 7:1 petroleum ether: ethyl acetate) gave 18 as a slightly yellow oil (2.36 g, 31%). 67
1 H NMR (250 MHz, CDCl3): δ 1.32 (s, 3H), 1.47 (s, 3H), 3.26-3.36 (m, 2H), 3.49-3.56
(m, 3H), 3.69-3.81 (m, 2H), 4.33-4.39 (m, 1H), 4.52-4.75 (m, 6H), 5.10 (s, 1H), 7.27-
13 7.34 (m, 10H) ppm. C NMR (100 MHz, CDCl3): δ 24.90, 26.36, 32.46, 67.33,
69.68, 72.27, 73.43, 82.55, 85.11, 86.64, 108.42, 112.60, 127.65, 127.77, 128.38,
138.02, 138.27 ppm. These spectroscopic data are in agreement with literature values.94
5.3.5 Dimethyliodoarsine (2)
This compound was prepared as described previously in section 2.3.1 of this
thesis and distilled prior to use.
5.3.6 Sodium dimethylarsenide (3)
In a 100 mL Schlenk tube, sodium (0.80 g, 35mmol) was stirred under argon in boiling toluene (40 mL) for 30 minutes. With stirring, the mixture was cooled to room temperature and the toluene removed from the resulting sodium granule by cannulla. Dry
THF (10 mL) was used to wash the sodium and was also removed by cannulla. The sodium was covered with dry THF (60 mL) and with stirring dimethyliodoarsine 2 (3.2 g,
14 mmol) was slowly added. After three hours, stirring was discontinued and the
reaction separated into a top green layer containing 3 and a dark bottom layer. Without
further purification or characterization, the top layer was used for the following reaction.
Care was taken to leave some of the top layer to insure none of the dark bottom layer was
transferred.
68
5.3.7 (R)-2,3-Dibenzyloxypropyl 5-Deoxy-5-dimethylarsino-2,3-O-isopropylidene-
β-D-riboside (19)
The bromoriboside 18 (2.32 g, 4.57 mmol) was dissolved in dry THF (10 mL) under argon inside a dry 250 mL two-neck round bottom flask. Sodium dimethylarsenide
3 was added by cannulla with stirring. The reaction immediately became orange and was allowed to stir overnight under a slow flow of argon. 20 mL of water were added and the resulting clear orange solution was extracted with ethyl acetate (3X, 50 mL). The combined organic extract was evaporated to yield a yellow oil. Flash chromatography
(10:1 petroleum ether: ethyl acetate) gave 19 as a colorless oil (1.51 g, 62 %). 1H NMR
(250 MHz, CDCl3): δ 0.95 (s, 3H), 0.96 (s, 3H), 1.31 (s, 3H), 1.58 (s, 3H), 1.77-1.87 (m,
2H), 3.51-3.59 (m, 3H), 3.69-3.76 (m, 1H), 3.85-3.91 (m, 1H), 4.24-4.31 (m, 1H), 4.47-
13 4.67 (m, 7H), 5.06 (s, 1H), 7.32 (m, 10H) ppm. C NMR (66 MHz, CDCl3):
δ 9.23, 25.1, 26.5, 34.8, 67.3, 72.2, 73.4, 77.0, 85.1, 85.7, 85.9, 108.8, 127.6,
128.4, 138.7 ppm. These spectroscopic data are in agreement with literature values.94
5.3.8 (R)-2,3-Dihydroxypropyl 5-Deoxy-5-dimethylarsino-2,3-O-isopropylidene-β-
D-riboside (20)
The benzyl deprotection of arsine 19 was carried out under the following
conditions. A dry 100 mL Schlenk tube was connected to a slow argon flow and cooled
to -78 oC by immersion in a dry ice / acetone bath. Anhydrous ammonia gas was
introduced into the tube and 30 mL of liquid ammonia was collected by condensation.
The arsine 17 (1.48 g, 2.8 mmol) dissolved in freshly distilled THF (15 mL) was added
along with pieces of sodium (80 mg, 3.5 mmol) as the mixture was stirred. The Schlenk 69
tube was sealed and argon was slowly bubbled out of an adjacent tube. The mixture
became blue and thick. After stirring for three hours at -78 oC, the Schlenk line was opened, the acetone/CO2 bath was removed and ammonium carbonate (40 mg) was
added. The mixture was allowed to sit and slowly warm to room temperature leaving a
dark brown residue. Water (40 mL) was added and the product was extracted into
methylene chloride (3X 50 mL). The combined organic extracts were dried over Na2SO4 and the solvent evaporated. Flash chromatography (1:1 petroleum ether: ethyl acetate)
1 gave 20 as a yellow oil (0.58 g, 59 %). H NMR (250 MHz, CDCl3): δ 1.00 (s, 3H),
1.02 (s, 3H), 1.32 (s, 3H), 1.48 (s, 3H), 1.65-1.70 (m, 1H), 1.79-1.88, (m, 1H), 3.60-
3.88 (m, 7H), 4.29-4.35 (m, 1H), 4.62-4.70 (m, 2H), 5.07 (s, 1H) ppm. 13C NMR (66
MHz, CDCl3): δ 9.18, 9.33, 25.01, 26.55, 34.69, 63.85, 70.65, 70.72, 85.03, 85.65,
85.84, 109.65, 129,63 ppm. These spectroscopic data are in agreement with literature
values.94
5.3.9 Dimethyldioxirane (9)
A two L, two neck flask was cooled in an ice bath and sodium bicarbonate (48.3
g, 575 mmol) was vigorously stirred into water (211 mL) and acetone (160 mL, 2200
mmol). Oxone (100 g, 162 mmol) was added over the course of ten minutes. Five minutes after the final addition, the pressure was pumped to 30 torr and 9 was collected as an acetone solution by vacuum distillation into a 500 mL, two neck flask cooled to -78
oC. The bubbling of the reaction mixture was intense and the reason for the large volume flask. The dioxirane solution was dried over potassium carbonate and stored over molecular sieves (A-4) at -20 oC. The concentration of the dioxirane solution was 70
determined by reaction with an excess of thioanisole 10 and GC-MS determination of 10 along with its oxidation product, phenylmethylsulfoxide 11, as described in Chapter 4 of this thesis. A 95 mL solution of 0.055 M 9 was available for use in the following reaction (3% yield from oxone). The molarity of dioxirane solutions was observed to slowly decrease when stored at -20 oC and therefore was determined by GC-MS analysis of the reaction with thioanisole immediately prior to further use. This procedure was an improvement over that described in Chapter 4.
5.3.10 (R)-2,3-Dihydroxypropyl 5-Deoxy-5-dimethylarsinoyl-2,3-O-isopropylidene-
β-D-riboside (21)
The arsine 20 (0.50 g, 0.14 mmol) was dissolved with stirring in 37 mL of a
0.0041 M solution of 9 (1.1 eqs) inside a 250 mL round bottom flask. The flask was filled
with argon and loosely sealed with a rubber septa. After four hours, stirring was halted
and volatiles were allowed to evaporate overnight under a slow stream of argon. The
white residue was dissolved in 1:1 methanol: chloroform (5 mL) and loaded onto a small
(200 mL) column of silica. Goggles and protective gloves and clothing were worn during
this transfer and the reaction flask was not held by hand because of the risk of explosion.
Flash chromatography (1:1 methanol: chloroform) gave 21 as a clear oil (0.42 g, 80 %).
1 H NMR (250 MHz, CDCl3): δ 1.32 (s, 3H), 1.49 (s, 3H), 1.73 (s, 3H), 1.89 (s, 3H),
2.17 (acetone impurity), 2.27 (dd, 1H), 2.98 (dd, 1H), 3.43-3.54 (m, 2H), 3.66-3.71 (m,
1H), 3.82 (1H), 4.06-4.13 (m, 1H), 4.41-4.78 (m, 3H), 5.12 (s, 1H) ppm. 13C NMR
(100 MHz, CDCl3): δ 14.74, 15.74, 24.75, 26.28, 64.25, 69.27, 70.90, 82.13, 85.01,85.24,
108.98, 112.77 ppm. These spectroscopic data are in agreement with literature values.94 71
5.3.11 (R)-2,3-Dihydroxypropyl 5-Deoxy-5-dimethylarsinoyl-β-D-riboside
(As328) (22)
The arsine oxide 21 (0.42 g, mmol) was dissolved and stirred in nine mL of a 9:1
(v:v) solution of trifluoroacetic acid in water. The acetal deprotection of the isopropylidene group took place over the course of ten minutes and the solution was quickly brought to dryness under high vacuum (3 torr). Water (20 mL) was added and the pH of the solution was made neutral by addition of an ammonium hydroxide solution
(~10 mL of a 1 M solution). The solvent was removed by evaporation and the product applied to 15 mL of exercised Amberlite IRA-400 anion exchange resin. The first 100 mL eluting from the resin was taken to dryness and gave 22 as a clear syrup (0.20 g, 56
%). This product was pure by NMR and was the only arsenic species detected by HPLC-
ICP-MS. The HPLC-ICP-MS analysis of synthetic 22 is shown in Figure 5.2 and its peak is identified by retention time match to a cation standard mix containing 22 as the third eluting peak following the marker peak as shown in Figure 5.3. 1H NMR (250 MHz,
CDCl3): δ 1.56 (s, 3H), 1.58 (s, 3H), 2.20 (dd, 1H), 2.35 (dd, 1H), 3.27-3.34 (m, 3H),
3.43-3.49, (m, 1H), 3.59-3.62 (m, 1H), 3.84 (d, 1H), 3.92-4.00 (m, 2H), 4.53 (s, 1H) ppm.
13 C NMR (66 MHz, CDCl3): δ 16.3, 16.7, 38.2, 64.6, 71.1, 72.4, 76.4, 77.9, 79.0, 109.6
ppm. These spectroscopic data are in agreement with literature values.94 The total yield for this synthesis from D-ribose was 1.1%.
72
IC-ICP-MS of As328
400000
350000
300000
250000 s 200000 ount C
150000
100000
50000
0 02468101214 Minutes
Figure 5.3. Anion chromatography of Synthetic As328.
Cationic As Standard Mix
20000
18000
16000
14000
12000 ts n 10000 Cou
8000
6000
4000
2000
0 02468101214 Minutes
Figure 5.4. Chromatographic profile of Anionic Arsenic Standards. 73
5.4 DISCUSSION
The synthesis of arsenosugar 22 is a long and complicated procedure. Distillation of dimethyldioxirane at 30 torr increased the yield of this oxidant. By replacing the oxidation reaction that previously utilized hydrogen peroxide in ether solvents, the synthesis is considerably safer if no less tedious. Alternate means of arsenic attachment to sugar moieties that do not involve an oxidation step is a continuing interest.
74
CHAPTER 6
DISCUSSION
75
6.1 DISCUSSION
The synthesis of arsenobetaines and arsine oxides has been improved and new
routes to these compounds are safer than previously reported. The novel arsines,
ethoxycarbonylmethyldimethylarsine and 2-ethoxycarbonylethyl(dimethyl)arsine allow
synthetic access to arsenobetaines while eliminating trimethylarsine as a precursor. The higher boiling points of these arsines make them considerably easier and safer to handle and manipulate. The previous oxidation of arsines with hydrogen peroxide in ether solvents is a dangerous endeavor and the new route utilizing dimethyldioxirane as an alternative oxidant removes several risks inherent in the oxidation of arsines. This reaction was first demonstrated on a simple compound, triphenylarsine, and following that success was utilized in the synthesis of the complex chemical arsenosugar As328.
76
CHAPTER 7
FT-IR SPECTRA
78 Ethoxycarbonylmethyl(dimethyl)arsine (4, n=1)
100
3418.2 90 2812.5 665.3
951.1 80 929.0 763.4
857.6 70 1388.9 1445.7 584.6 1364.8 60 889.8 2911.3 1149.6
%T 50 1416.2
2981.2 1035.1 40
30
20 1097.7
10 1250.5 1720.8
0
-8.1 4000.0 3000 2000 1500 1000 450.0 cm-1 79 %T 30.6 78.6 35 40 45 50 55 60 65 70 75 4000.0 3452. 8 2-Ethoxycarbonylethyl(dimethyl)arsine (4,n=2) 3000 2977. 2907. 4 3 2000 1735. cm -1 0 1500 1419. 1368. 3 1341. 7 1277. 0 2 1201. 1149. 3 9 1035. 1000 0 888. 842. 4 0 740. 7 581. 7 461. 470. 476. 481. 493. 450.0 1 6 2 0 9 80 Ethoxycarbonylmethyl(trimethyl)arsonium iodide (5, n=1) 105.3 104
102
100
98 600.0 96 2212.6
94
92 648.3 90 1444.2
88 1403.0 86 1364.9 %T 84 878.3 2872.0 2909.1 82 1181.9 1028.1 80
78
76
74 2990.6 1118.0 72
70 1279.4 937.2
68 1732.4 66.8 4000.0 3000 2000 1500 1000 450.0 cm-1 43.9 2-Ethoxycarbonylethyl(trimethyl)arsonium iodide (5, n=2)
81 42
40
38 593.2 36
34
32 651.4
30
1114.6 28 %T
26 1054.4 1023.0 24 2983.6 1349.8 861.5 1423.6 22 1384.6 1368.7 20
18 1240.3 1193.2 941.3 16
14 1732.2
12.4 4000.0 3000 2000 1500 1000 450.0 cm-1 94.4 Arsenobetaine (6, n=1) 82 90
536.9 85
80
596.8
75
70 %T 638.3
3005.1 1357.8 65 2946.5 2890.2 1267.9 1253.3 839.5 60
896.3
55 1708.8 1157.0 934.4
50 1384.9
45.3 4000.0 3000 2000 1500 1000 450.0 cm-1 78.5 Arsenobetaine-2 (6, n=2) 83 76
74
72
788.1 70 1187.1 648.9
68
2993.7 %T 66 1296.6 867.6 2915.9
64
62
60 941.6
58
56 1592.9 1383.9
54.0 4000.0 3000 2000 1500 1000 450.0 cm-1 Triphenylarsine oxide (12) 64.1 84 63 62
61
60 472.7 59 58 57 56 1663.6 55
54 1087.0 53 1439.6 691.1 %T 3399.4 743.7 52
51 50 49 882.0 48
47 46 45 1384.6 44 43
41.6 4000.0 3000 2000 1500 1000 450.0 cm-1 Methyl 2,3-O-Isopropylidene-β-D-riboside (14) 103.6
100 85 95
2161.5 1951.8 1738.9 90 1690.0 2024.7
85
80 649.0 515.7 75 580.4 70
761.1 %T 65
60
55
50 1458.8
45 2835.6 1273.1 961.6 1240.1 1160.9 1009.0 40 3466.4 1374.2 870.0 35 2988.1 1210.7 1043.8 2940.2 1092.9
28.2 4000.0 3000 2000 1500 1000 450.0 cm-1 Methyl 2,3-O-Isopropylidene-5-O-toluenesulfonyl-β-D-riboside (15)
103.3
86 102
100 3421.5
98
96
94
92
90 1597.1 1452.1 2947.4 681.7 %T 88 1272.8 903.2 524.1 1020.6 854.2 663.9 86 1247.0
84 867.0 556.5 1044.0 813.4 833.8 82 1211.6
80 1358.7 1092.0 78 1180.5 959.9 976.1 76
73.7 4000.0 3000 2000 1500 1000 450.0 cm-1 Methyl 5-Bromo-5-deoxy-2,3-O-isopropylidene-β-D-riboside (16) 133.1 130 87 125
120
115 1730.1 1681.9 110 680.0 105 652.3 515.1 2834.9 774.9 580.6 100 1441.0 819.5 95
90 931.1
85 2989.4 1272.1 %T 80 1240.0 973.9 75 2938.5 1194.7 958.8 70 1374.2 1162.2 65 1210.4 60
55 1053.3 1029.3 50 869.3 1073.8 45
40 1105.6 35 32.4 4000.0 3000 2000 1500 1000 450.0 cm-1 88
CHAPTER 8
MASS SPECTRA GC-MS Dimethyliodoarsine (2) 89
TI C: DMAI . D 3200000 3000000 2800000 2600000 2400000 2200000 2000000 1800000 1600000 1400000 1200000 1000000 800000 600000 400000 200000 0 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50 7.00 7.50 8.00 8.50 CH3
90 GC-MS Dimethyliodoarsine (2) As
H3C I Average of 2.485 to 2.932 min.: DMAI.D 232 550000 500000 450000 217 105 400000 89 350000 300000 250000 200000 150000 127 202 100000 75 50000 45 59 152 169 184 300 337 359 0 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 - -> CH3 OH+ Ethoxycarbonylmethyl(dimethyl)arsine (4, n=1) 91 + C6H14AsO2 As C Exact Mass: 193.0210 H C OCH CH 3 2 3 Found: 193.0184 92 H 3 C As CH 3 OH + OC H 2 CH 3 (2-Ethoxycarbonylethyl)dimethylarsine (4,n=2) C Exact Mass:207.0366 Found: 207.0375 7 H 16 AsO 2 + 93 Ethoxycarbonylmethyl(dimethyl)arsonium (5, n=1)
+ C7H16AsO2 CH3 Exact Mass: 207.0366 Found: 207.0382 H3C As CO2Et H3C (2-Ethoxycarbonylethyl)trimethylarsonium (5, n=2) 94
+ CH3 C8H18AsO2 H3C Exact Mass: 221.0523 As+ Found: 221.0539
H3C CO2Et 95 Arsenobetaine (6, n=1)
CH 3 + C5H12AsO2 H C Exact Mass: 179.0053 3 As CO H 2 Found: 179.0040 H3C 96 Arsenobetaine-2 (6, n=2)
CH3 + C6H14AsO2 H3C Exact Mass: 193.0210 As+ Found 193.0182 H3C CO2H 97 Triphenylarsine oxide (12)
OH+ + C18H16AsO As Exact Mass: 323.0417 Found : 323.0344 98
CHAPTER 9
NMR SPECRTRA Dimethyliodoarsine (2) 1H NMR 99 Dimethyliodoarsine (2) 13C NMR 100
CDCl3 Ethoxycarbonylmethyl(dimethyl)arsine (4, n=1) 1H NMR 101 102 Ethoxycarbonylmethyl(dimethyl)arsine (4,n=1 CDCl 3 )
13 C NMR 2-Ethoxycarbonylethyl(dimethyl)arsine (4, n=2) 1H NMR 103 2-Ethoxycarbonylethyl(dimethyl)arsine (4, n=2) 13C NMR 104
CDCl3 105 Ethoxycarbonylmethyl(trimethyl)arsonium HOD (5 , n=1) 1 H NMR 106 Ethoxycarbonylmethyl(trimethyl)arsonium (5 , n=1) 13 C NMR (2-Ethoxycarbonylethyl)trimethylarsonium (5, n=2) 1H NMR 107
HOD (2-Ethoxycarbonylethyl)trimethylarsonium (5, n=2) 13C NMR 108 Arsenobetaine (6, n=1) 1H NMR 109
HOD 110 Arsenobetaine (6,n=1) 13 C NMR Arsenobetaine-2 (6, n=2) 1H NMR HOD 111 112 Arsenobetaine-2 (6,n=2) 13 C NMR Triphenylarsine oxide (12) 1H NMR 113 Triphenylarsine oxide (12) 13C NMR 114 1 Methyl 2,3-O-Isopropylidene-β-D-riboside (14) H NMR 115
CHCl3 Methyl 2,3-O-Isopropylidene-5-O-toluenesulfonyl-β-D-riboside (15) 1H NMR 116
CHCl3 Methyl 5-Bromo-5-deoxy-2,3-O-isopropylidene-β-D-riboside (16) 1H NMR 117
CHCl3 (R)-2,3-Dibenzyloxypropyl 5-Bromo-5-Deoxy-2,3-O-isopropylidene-β-D-riboside (18) 1H NMR 118 (R)-2,3-Dibenzyloxypropyl 5-Bromo-5-Deoxy-2,3-O-isopropylidene-β-D-riboside (18) 13C NMR 119
CDCl3 120 1 (R)-2,3, Dibenzyloxypropyl 5-Deoxy-5-dimethylarsino-2,3-O-isopropylidene-β-D-riboside (19) H NMR 6 5 9 9 0. 0. (R)-2,3, Dibenzyloxypropyl 5-Deoxy-5-dimethylarsino-2,3-O-isopropylidene-β-D-riboside (19) 13C NMR 121 CDCl3 122 (R)-2,3,-Dihydroxypropyl CHCl 3 5-Deoxy-5-dimethylars ino-2,3-O-isopropylidene- β - D -riboside (20) 1 H NM R 123 (R)-2,3,-Dihydroxypropyl 5-Deoxy-5-dimethylars CDCl ino-2,3-O-isopropylidene- 3 β - D -riboside (20) 13 C NMR 1 (R)-2,3-Dihydroxypropyl 5-Deoxy-5-dimethylarsinoyl-2,3-O-isopropylidene-β-D-riboside (21) H NMR 124
Acetone 13
(R)-2,3-Dihydroxypropyl 5-Deoxy-5-dimethylarsinoyl-2,3-O-isopropylidene-β-D-riboside (21) C NMR
5
2 1
CdCl3 (R)-2,3-Dihydroxypropyl 5-Deoxy-5-dimethylarsinoyl-β-D-riboside (22) 1H NMR 126
HOD (R)-2,3-Dihydroxypropyl 5-Deoxy-5-dimethylarsinoyl-β-D-riboside (22) 13C NMR 127