Modifications in the nitric acid oxidation of D-mannose: X-ray crystal structure of N,N0-dimethyl D-mannaramide

Article abstract of DOI:10.1016/j.carres.2013.05.004

Nitric acid oxidation of D-mannose was carried out under an oxygen atmosphere using a computer controlled reactor. The process represents a catalytic oxidation of D-mannose with oxygen as the terminal oxidant. The crude oxidation product was esterified with methanolic HCl and the esterified product directly converted to crystalline N,N0-dimethyl-D-mannaramide with methylamine. Treatment of the diamide in aqueous sodium hydroxide gave solid disodium D-mannarate. The X-ray crystal structure of N,N0-dimethyl-D-mannaramide was determined as a model for the repeating D-mannaramide units of stereoregular poly(alkylene-D-mannaramides). Disodium D-mannarate was prepared as a precursor of esterified D-mannaric acid for use as a reactive diacid monomer to prepare poly-D-mannaramides.

Full text of DOI:10.1016/j.carres.2013.05.004

Carbohydrate Research 376 (2013) 29–36
Contents lists available at SciVerse ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Modifications in the nitric acid oxidation of D-mannose: X-ray crystal
structure of N,N0-dimethyl D-mannaramide
Chrissie A. Carpenter a, , Kenneth I. Hardcastle b, Donald E. Kiely a,
⇑
a Shafizadeh Rocky Mountain Center for Wood and Carbohydrate Chemistry, The University of Montana, Missoula, MT 59812, USA
b Emory University Crystallography Laboratory, Department of Chemistry, Emory University, Atlanta, GA 30322, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Nitric acid oxidation of
trolled reactor. The process represents a catalytic oxidation of
D
-mannose was carried out under an oxygen atmosphere using a computer con-
-mannose with oxygen as the terminal
oxidant. The crude oxidation product was esterified with methanolic HCl and the esterified product
directly converted to crystalline N,N0-dimethyl-
-mannaramide with methylamine. Treatment of the
diamide in aqueous sodium hydroxide gave solid disodium -mannarate. The X-ray crystal structure of
N,N0-dimethyl-
-mannaramide was determined as a model for the repeating -mannaramide units of
stereoregular poly(alkylene- -mannaramides). Disodium -mannarate was prepared as a precursor of
esterified -mannaric acid for use as a reactive diacid monomer to prepare poly- -mannaramides.
Received 19 March 2013
Received in revised form 29 April 2013
Accepted 1 May 2013
D
D
Available online 16 May 2013
D
D
D
Keywords:
D
D
D
-Mannaric acid
Nitric acid oxidation
N,N0-Dimethyl-
-mannaramide
Disodium -mannarate
D
D
Ó 2013 Elsevier Ltd. All rights reserved.
D
D
Crystal structure
1. Introduction
out at 25–30 °C as a catalytic process with oxygen serving as the
terminal oxidant.
D
-Mannaric acid (3) is an aldaric acid of interest to us as a chiral
The primary goal of the research described here was to evaluate
the computer controlled reactor parameters for controlled oxida-
diacid monomer for the preparation of stereoregular poly-D-mann-
aramides,1–3 members of a class of polymers labeled as polyhydr-
tion of
to find a suitable crystalline alternative to mannaro-1,4:6,3-dilac-
tone (4) as the isolated product from -mannose oxidation. While
the dilactone 4 can be used directly as a diacid monomer for poly-
D-mannose (2) to D-mannaric (3) acid. A second goal was
oxypolyamides.2 This diacid was first described in the late 19th
century as a nitric acid oxidation product of
mannitol (1)5 (Scheme 1) and typically isolated as crystalline
1,4:6,3-
-mannarodilactone (4).1,4,6–8
In a recent report from this laboratory,9 a modified nitric acid
oxidation method was employed for the conversion of -glucose
to -glucaric acid, a method that we have now applied for oxida-
tion of -mannose to -mannaric acid. Oxidation of sugars with ni-
D
-mannose (2)4 or
D
-
D
D
merization with diamines to make poly(alkylene D-mannara-
mides),1–3 it also undergoes competing decomposition reactions
in basic alcohol solution,10 a reaction medium typical for these
D
D
polymerizations. Thus, we were motivated to find a
D-mannaric
D
D
acid derivative that could be polymerized more efficiently and
tric acid are characterized by an early highly energetic exotherm, a
reaction characteristic that makes control of such oxidations chal-
lenging. These oxidations also generate large amounts of NOx
gases, in particular brown nitrogen dioxide and colorless nitric
with less decomposition than observed with 4. Disodium
D
-mann-
-mann-
arate (6), generated by basic hydrolysis of N,N0-dimethyl-
D
aramide (5) in aqueous sodium hydroxide (Scheme 2), was
chosen as a target molecule for eventual use in such a polymeriza-
tion process, the subject of a future report.
oxide. In the D-glucose to D-glucaric acid conversion recently de-
scribed,9 the nitric acid oxidation was carried out under computer
control in a closed reaction flask under a positive pressure of
oxygen. Oxygen was used to oxidize nitric oxide back to nitrogen
dioxide, which in water is converted into nitric acid. Through use
of a computer controlled reactor, the oxidation could be carried
2. Results and discussion
2.1. Oxidation reaction conditions
In order to control the highly exothermic nitric acid oxidation of
-mannose, oxidations were carried out in a Mettler Toledo RC-1
Labmax computer controlled reactor. The reactor is well suited
for this application as it is designed to control temperature, pres-
sure, and stirring rate in a closed reaction flask. In addition the
⇑
Corresponding author. Tel.: +1 406 543 6126; fax: +1 406 543 2304.
E-mail addresses: ccarpenter01@ugf.edu (C.A. Carpenter), donald.kiely@umon-
tana.edu, don@rivertop.com (D.E. Kiely).
D
Current address: Department of Chemistry, The University of Great Falls, 1301
20th Street South, Great Falls, MT 59405, USA.
0008-6215/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.carres.2013.05.004

30
C. A. Carpenter et al. / Carbohydrate Research 376 (2013) 29–36
OH
OH
OH
HO
OH
OH
OH
1
HO
H
O
6
OH
4
OH
2
- 2 H2O
O
3
5
[O]
OH
1
O
HO
O
HNO3
O
OH
OH
O
H
OH
3
4
OH
O
HO
HO
OH
2
H
Scheme 1. Preparation of
D-mannaro-1,4:6,3-dilactone by nitric acid oxidation of mannitol or
D
-mannose.
O
OH
OH
O
OH
OH
H
N
O
NaOH
H2O
Na
H3C
+
2 CH3NH2
Na
N
H
CH3
O
OH
OH
O
OH
OH
O
6
5
Scheme 2. Hydrolysis of N,N0-dimethyl-
D-manaramide (5) in aqueous sodium hydroxide to disodium D-mannarate (6).
D
-mannonic acid molar ratio value of about 4:1 based on GC
integration.
A final oxidation protocol (oxidation procedure 3) was carried
out that further pushed the oxidation and generated -mannaric
acid but no residual -mannonic acid. In this procedure the amount
of nitric acid was increased (3 mol–5 mol), the reactor temperature
was set to 30 °C during the -mannose solution addition, then
reactor setup allowed for programed addition of
D-mannose in
aqueous solution to nitric acid (70%). The -mannose solution also
D
contained a small amount of sodium nitrite as a reaction activator.
To facilitate regeneration of nitric acid from the NOx gases pro-
duced in the oxidation, the reaction flask was kept under a positive
pressure of oxygen. Oxygen rapidly oxidizes nitric oxide to nitro-
gen dioxide which then regenerates nitric acid through a series
of reactions11 as shown in Figure 1.
D
D
D
raised to 40 °C and kept at this temperature to complete the oxida-
tion step. Although no attempt was made to optimize the oxidation
The initial oxidation conditions employed for
tion were patterned after those employed for
D
-mannose oxida-
-glucose oxida-
-mannose
of
three procedures described, oxidation procedure 3 gave the most
complete conversion of -mannose to -mannaric acid. A prelimin-
D-mannose using the computer controlled reactor system, of the
D
tions,9 that is, a 4:1 molar ratio of nitric acid (70%) to
D
D
D
at reaction temperatures of 25–30 °C (oxidation procedure 1).
The course of the oxidation was monitored by GC–MS, samples
being analyzed as their per-O-trimethylsilyl derivatives. The two
ary report of this oxidation was reported in the PhD dissertation of
C.A. Carpenter.12
stage addition of -mannose was carried out at 25 °C, followed
D
by raising the temperature to 30 °C and maintaining that temper-
2.2. GC–MS monitoring of the oxidation
ature for 6 h until the end of the reaction (total reaction time 9.4 h).
A significant amount of
after the addition to nitric acid was complete, but there was none
present in the final reaction mixture that contained -mannonic
acid and -mannaric acid as the major and minor products, respec-
tively. In contrast, when the same experimental protocol was used
for -glucose oxidation, almost all of the -glucose was consumed
by the end of the addition and the final product mixture was dom-
inated by -glucaric acid. Consequently, when employing the same
mild reaction conditions, nitric acid oxidation of -glucose is faster
than that of -mannose at both the anomeric and terminal carbons.
D-mannose (ca. 70%) was still present right
All oxidation mixture samples were analyzed as their per-O-tri-
methylsilyl derivatives. GC–MS Method 1 is a GC–MS temperature
program that was routinely used in this laboratory for carbohy-
drate analysis.13 However the GC–MS Method 1 protocol failed to
separate per-O-trimethylsilylated D-mannonic acid and D-mannaric
acid so a new temperature program protocol, GC–MS Method 2,
was developed. Method 2 utilized a much slower temperature
D
D
D
D
D
ramp rendering baseline separation of the
D-mannonic acid and
D
D-mannaric acid peaks. A GC–MS chromatogram employing Meth-
D
od 2 (Fig. 2) illustrates separation of
D
-mannonic and
D-mannaric
A second experimental protocol (oxidation procedure 2) was
then applied wherein the reaction mixture temperature for the en-
tire reaction period was 30 °C. Increasing the reaction mixture
acids in an oxidation mixture.
The mass spectra of D-mannaric acid and D-mannonic acid are
readily differentiated by their mass fragmentation patterns.14
Many of the middle to small mass range fragments are the same
but the [Mꢀ15]+ peaks, arising from a loss of CH3 from a TMS group
are unique to each species and were used for compound identifica-
temperature by only 5 °C during the
sulted in all the -mannose being consumed right after the
addition was completed, and produced a final
D-mannose addition stages re-
D
D-mannaric acid to
tion; for
D-mannaric acid m/z = 627 and for D-mannonic m/z = 613.
2.3. IC monitoring of the oxidation
2 NO (g) + O2 (g)
2 NO2 (g) + H2O (l)
3 NO2 (g) + H2O (l)
3 HNO2 (l)
2 NO2 (g)
HNO3 (l) + HNO2 (l)
2 HNO3 (l) + NO (g)
(1.1)
(1.2)
(1.3)
(1.4)
GC–MS was the principal analytical technique used to monitor
the oxidation method during the computer controlled reactor pro-
cess development stage of this project. Thereafter an ion chromato-
HNO3 (l) + 2 NO (g) + H 2O (l)
Figure 1. Common NO and NO2 reactions in water/O2.

C. A. Carpenter et al. / Carbohydrate Research 376 (2013) 29–36
31
Figure 2. Gas chromatogram of an oxidation mixture using GC–MS Method 2 showing separation of per-O-TMS-
D-mannaric acid (peak 1) from per-O-TMS-
D-mannonic acid
(peak 2).
graphic method (IC), developed for analysis of
D
-glucose nitric acid
2.5. Nanofiltration separation of inorganic nitrate from
mannaric acid and other hydroxyacid salts
D-
oxidation product mixtures,9 was effectively applied to
D-mannose
oxidation reaction mixtures.
Mannaric acid was previously isolated as D-mannaro-1,4:6,3-
dilactone (4) after concentration of the syrupy acidic product
mixture.6,15 However, it is difficult to completely remove nitric
acid by any evaporation process from a carbohydrate oxidation
mixture because nitric acid forms a negative azeotrope with
water. In addition uncontrolled oxidation of carbohydrate sub-
strate continues during the evaporation process. Consequently,
the nanofiltration and diffusion dialysis methods recently em-
2.4. N,N0-Dimethyl-
mannarate (6) from
D
-mannaramide (5) and disodium
D-
D-mannose oxidation
A route to disodium
avoided sodium hydroxide hydrolysis of base labile 4 and pro-
ceeded through N,N0-dimethyl-
-mannaramide (5). It was rea-
soned that basic hydrolysis of acyclic 5 would be straightforward
and not be accompanied by the b-elimination reactions and col-
ored product formation associated with base hydrolysis of dilac-
tone 4.
An oxidation product mixture from oxidation procedure 3 was
concentrated at 40 °C under reduced pressure to a syrup which
was then treated with methanolic hydrogen chloride to give a
mixture of the methyl ester lactone (7) and acyclic dimethyl
mannarate (8). Aminolysis of 7 and 8 with methylamine in meth-
anol produced the insoluble solid acyclic diamide 5 (52.0%). So-
D-mannarate (6) was undertaken that
D
ployed to remove nitric acid from the
tion mixture9 were applied to the
oxidation mixtures. The nanofiltration protocol was used to sep-
arate inorganic nitrate from salts of -mannaric acid and other
hydroxyl acids generated in -mannose oxidation. Typically, an
D
-glucose nitric acid oxida-
D
-mannose nitric acid
D
D
oxidation mixture was partially concentrated at 40 °C and re-
duced pressure to a syrup which was then made basic with so-
dium hydroxide to generate an aqueous solution of sodium
nitrate and product carboxylic acids sodium salts. The solution
was applied to a nanofilter system as previously described,9 gen-
erating a permeate stream containing the bulk of the sodium
dium hydroxide hydrolysis of
(Scheme 3).
5
then yielded solid 6 (98%)
nitrate and
a retentate stream with most of the disodium
HO
OMe
O
H
O
OH OH
O
OH OH
O
Concentrate
- H2O
O
Cl
MeOH
O
O
OH
O
OMe
HO
HO
MeO
O
OH
H
+
HO
H
O
OH OH
O
OH OH
O
H
OH
8
Crude 4
Residual HNO3
Crude 3
+
7
Concentrate
- H2O
O
OH OH
O
OH
OH
Aq. NaOH
H
N
MeNH2
MeOH
H3C
O
Na
N
H
CH3
Na
O
OH OH
O
OH
OH
O
6
5
Scheme 3. Preparation of disodium D-mannarate (6).

32
C. A. Carpenter et al. / Carbohydrate Research 376 (2013) 29–36
Figure 3. Ion chromatogram of the diffusion dialysis organic acid stream from a D-mannose oxidation mixture.
Figure 4. Ion chromatogram of the diffusion dialysis reclaimed nitric acid stream from a D-mannose oxidation mixture.
2.6. X-ray crystallographic analysis of N,N0-dimethyl-
mannaramide (5)
D
-
D
-mannarate and minor hydroxyacid sodium salts. Ion chroma-
tography was used to monitor the course of the nanofiltration
separation.9
N,N0-dimethyl-
D-mannaramide (5) recrystallized from water
yielded crystals of good qualityand size for single-crystalX-ray crys-
tal structure analysis. This molecule is of interest to us as a precursor
A diffusion dialysis protocol was used to separate nitric acid
from D
-mannose oxidation products at room temperature.9 Two
aqueous streams were generated, a reclaimed acid stream and a
product stream. The streams were analyzed using ion chromatog-
raphy9 as shown in Figure 3 (the organic acid stream) and Figure 4
(the nitric acid stream). Both streams contain the same compo-
nents, that is, nitric, mannaric, tartaric, and oxalic acids, plus an
unknown component (ca. 22 min), but in very different amounts.
The reclaimed acid stream accounts for ca. 92% of the nitric acid
and ca. 34% of organic acids, with the product stream accounting
for ca. 8% of nitric acid and ca. 66% of organic acids. Reclamation
of additional organic acid products can be achieved by repeating
the diffusion dialysis step, but was not applied in this example.
The separations described here were not optimized but do serve
to further illustrate the use of diffusion dialysis to remove nitric
acid from a carbohydrate acid feedstock under low pressure at
room temperature.
for the preparation poly(
formational model for the repeating
polyamides.1–3 The ortep drawing of crystalline N,N0-dimethyl-
D
-mannaramides) and potentially as a con-
D
-mannaramide unit in these
D-
mannaramide (5) is shown in Figure 5. The absolute structure
parameter refined to an indeterminate value of 0.38 (0.11), is typical
of a small molecule with no atoms heavier than oxygen. The absolute
stereochemistry was confirmed by direct comparison from the
known configuration of the precursor, D-mannose.
Diamide 5 has extensive intermolecular hydrogen bonding
(Fig. 6), with seven hydrogen bonds between pendant hydroxyl
groups, two hydrogen bonds each involving O(2), O(4), and O(5)
and one associated with O(3). The terminal amido hydrogens,
H(1) and H(2), are hydrogen bonded to carbonyl oxygens O(6)
and O(1), respectively.

C. A. Carpenter et al. / Carbohydrate Research 376 (2013) 29–36
33
Figure 5. Ortep drawing showing the labeling of the atoms in the crystal structure of N,N0-dimethyl-
D-mannaramide (5).
Figure 6. The X-ray crystal structure of 5 with associated hydrogen bonds.
Diamide 5 does not have destabilizing 1,3-hydroxyl oxygen
interactions in an extended conformation so it is not surprising that
the crystallineform of 5 adopts this conformation. Consequently, the
crystal structure of 5 is a reasonable conformational model for the
methylsilane (0.00 ppm) in DMSO-d6 as a solvent. FIDs obtained
were processed using ACD/SpecManager 1D NMR software.
3.1. Nanofiltration system9
repeating
D-mannaramido units of poly(D
-mannaramides).1–3
Reverse osmosis (RO) purified water was used as the solvent in
a laboratory built nanofiltration unit with appropriate valves,
pump, lines, pressure gauge, and a GE DL2540F membrane.
3. Experimental
Nitric acid solution reaction mixtures were concentrated under
reduced pressure using a system consisting of a Buchi Rotovapor R-
205, Buchi Vacuum Controller V-800, Buchi Heating Bath B-490,
Brinkmann Model B-169 Vacuum Aspirator, and a Thermo Haake
compact refrigerated circulator DC30-K20 in conjunction with a
Thermo Haake EK45 immersion circulator cooling coil. Elemental
analyses were preformed by Atlantic Microlab, Inc., Norcross, Geor-
gia. The freeze-drier used was a LABCONCO Freezone 4.5. Millipore
water was obtained from a Millipore Simplicity 185 system. Re-
verse osmosis (RO) water was generated in house. Melting points
were obtained with a Fisher-Johns melting point apparatus. NMR
spectra were recorded using a 400 MHz Varian Unity Plus spec-
trometer, at 400 MHz for proton spectra, and at 100 MHz for car-
bon spectra. Chemical shifts are expressed in parts per million
relative to t-butyl alcohol (1.203 ppm) in D2O as a solvent or tetra-
3.2. Diffusion dialysis system
Diffusion dialysis was carried out with a Mech-Chem Diffusion
Dialysis Acid Purification System laboratory scale Model AP-L05
consisting of two metering pumps, the first being the acid-reclaim
pump and the second being the acid-reject pump.
3.3. GC–MS system
GC–MS was performed on an Agilent 6890N Network GC Sys-
tem with an on-column injector fitted with a 5%-phenyl 95%-dim-
ethylpolysiloxane column (Phenomenex ZB-5, 30 m ꢁ 32 mm
ꢁ 0.25
lm) installed with direct interface to an Agilent 5973 Net-
work MS Detector.

34
C. A. Carpenter et al. / Carbohydrate Research 376 (2013) 29–36
3.4. GC–MS temperature programs
3.8. GC–MS sample analysis
Two temperature programs were employed for sample analysis:
GC–MS Program 1; begins at 32 °C, +12.5 °C/min to 250 °C, (hold
5 min), +20°/min to 300 °C, (hold 5 min), total acquisition time of
30.10 min; GC–MS program 2: begins at 40 °C, +6.0 °C/min to
120 °C, +2.0 °C/min to 210 °C, +8.0 °C/min to 310 °C, (hold
10 min), total acquisition time of 80.83 min.
Samples were taken during the course of each oxidation for GC–
MS analysis: oxidation procedure 1, end of stages 5 and 7, stage 8
(h 1 and 2), final reaction mixture; oxidation procedure 2, end of
stages 5 and 7, stage 8 (h 1 and 2), end of stage 8; oxidation proce-
dure 3, end of stages 5 and 7, stage 8 (h 1,2, 3, 4, and 5), final reac-
tion mixture. The aqueous sample (2 drops) was cooled (ice bath),
transferred to a 7 mL vial, made basic with sodium hydroxide (2 M,
8 drops) and xylitol (30 mg/mL, 2 drops) added as a qualitative
internal standard. Samples were then dried on a freeze-drier. Tri-
SilÓ Reagent (1.0 mL) was added to dried sample (10.0 mg) in a
7 mL vial and the mixture heated (50 °C, 1 h). Solid residue in the
trimethylsilylation reaction mixture was removed by centrifuga-
tion and the supernatant transferred (pipette) to heptane (3 mL).
The heptane solution was centrifuged and the resulting superna-
tant removed for GC–MS analysis (GC–MS program 2).
3.5. Ion chromatography9
IC was performed on a Dionex ICS-2000 Ion Chromatography
System consisting of a Dionex IonPacÒ AS II analytical column
and a sodium hydroxide EluGenÒ cartridge in conjunction with
Chromeleon software. Samples were analyzed using a 35 mM so-
dium hydroxide isocratic elution method with a flow rate of
1.5 mL/min running with the suppressor current at 186 mA.
3.6. Nitric acid oxidation reactor setup
3.9. Isolation of
nanofiltration
D-mannaro-1,4:6,3-dilactone (4) after
Oxidation of D-mannose was carried out using a Mettler Toledo
RC-1 Labmax reactor fitted with a top-loading balance, a liquid
feed pump, a Sierra oxygen flow valve, a mechanically driven stir-
ring rod, a thermometer, a 2 L thermal jacketed flask, an FTS recir-
culating chiller, and a pressure manifold fitted with pressure relief
valves, and a pressure gauge. Operation of the reactor was con-
trolled using a personal computer with CamilleTG v1.2 software.
The product from oxidation procedure 3 taken directly from the
Mettler Toledo Labmax reactor, was concentrated to a thick syrup
at 40 °C. The syrup was dissolved in water (200 mL), the solution
concentrated to a thick syrup, the syrup dissolved in water
(100 mL) and the solution stirred (40 °C, 1.5 h). Sodium hydroxide
(5 M) was added (ice bath) to bring the solution to pH 9.5 which
was diluted with RO water to a volume of 3650 mL and subjected
to a nanofiltration protocol using an Osmonics DL2514 membrane
(2.500 diameter, 1400 length). When the permeate volume reached
500 mL, RO water (500 mL) was added to the feedstock. The typical
rate of permeate flow was 65.15 mL/min. This was repeated until
3000 mL of permeate had been collected and 3000 mL of RO water
had been added. The typical permeate flow rate when removing
the last 500 mL was 43 mL/min. Filtration continued after the last
500 mL of RO water was added to the feedstock and the permeate
flow slowed to a trickle.
3.7. Procedures for nitric acid oxidation of D-mannose
3.7.1. Oxidation procedure 1 (4:1 M ratio of nitric acid to D-
mannose at 25 °C)
The oxidation was carried out using the Labmax reactor. Reac-
tion parameters for the oxidation were programed in a series of
stages: Stage 1: The reactor temperature was set at 25 °C, the stir-
ring rod speed was set at 200 rpm and maintained throughout the
remaining stages. Concentrated nitric acid (70%, 3.00 mol, 187 mL)
was added to the reaction vessel, the vessel was closed to the
atmosphere, and time set for 1 min. Stage 2: Oxygen was added
to bring the pressure to 0.25 bar and pressure maintained until
Samples were taken for IC during the course of diffusion dialy-
sis: (1) feedstock solution, (2) reclaimed acid 4 h, (3) product 4 h,
(4) reclaimed acid 5 h, (5) product 5 h, (6) reclaimed acid 9 h, (7)
product 9 h, (8) reclaimed acid 21 h, (9) product 21 h, (10) re-
claimed acid collective pot, and (11) product collective pot. Sam-
ples were diluted 100-fold for analysis by IC.
the end of stage 6. Stage 3:
D-Mannose [43.3 g of an aqueous
62.5% solution containing sodium nitrite (8.7 mmol, 0.6 g)] was
added over 30 min. Stage 4: A stabilization period with no change
The retentate solution of enriched disodium D-mannarate (6)
in reaction conditions was employed (10 min). Stage 5:
D
-Mannose
was concentrated to 200 mL and then treated with an excess of
Amberlite IR-120 H+ form resin for 1 h to give an aqueous solution
of crude diacid (3). The resin was removed by filtration and thor-
oughly rinsed with water. The solution was concentrated to a thick
[173.9 g of an aqueous 62.2% solution containing sodium nitrite
(8.7 mmol, 0.6 g)] was added over 90 min. Stage 6: A second stabil-
ization period (5 min) was employed. Stage 7: The reactor temper-
ature was set to 30 °C and pressure to 0.5 bar over 1 h. Stage 8.
Reaction conditions were maintained over 6 h. Stage 9: The reac-
tion mixture was cooled to 25 °C over 10 min and the vessel
opened to the atmosphere.
syrup, seeded with
vacuum at room temperature using a vacuum pump for 24 h. The
dried syrup was triturated with EtOH/EtOAc (3:1) to give solid
D-mannaro-1,4:6,3-dilactone, and dried under
D-
mannaro-1,4:6,3-dilactone (4, 43.18 g, 248 mmol, 33.1%): mp
183–185 °C, lit. mp 170–175 °C.7 1H NMR (D2O) d 4.9 (d, 2H, H-3,
H-4) d 5.31 (d, 2H, H-2, H-5). 13C NMR (D2O) d 70.36 (C3, C4) d
77.46 (C2, C5) d 176.29 (C1, C6).
3.7.2. Oxidation procedure 2 (4:1 M ratio of nitric acid to D-
mannose at 30 °C)
The oxidation was carried out under the conditions described in
oxidation procedure 1 with the following modification: Stage 1.
The reactor temperature was set at 30 °C and maintained through-
out the entire reaction.
3.10. N,N0-Dimethyl
D-mannaramide (5)—direct isolation from
oxidation procedure 3
An oxidation mixture from oxidation procedure 3 was concen-
trated to a thick syrup at 40 °C. Water (200 mL) was added to dis-
solve the dried oxidation mixture and the solution concentrated to
dryness at 40 °C. The water dilution and concentration procedure
was repeated four times. The final concentrate was dissolved in
methanol (500 mL) and acetyl chloride (2.67 mL, 37.5 mmol) was
added dropwise (ice bath) with stirring. The resulting solution
3.7.3. Oxidation procedure 3 (6.67:1 M ratio of nitric acid to D-
mannose at 30 °C
The oxidation was carried out under the conditions described in
oxidation procedure 1 with the following modifications: Stage 1.
The reactor temperature was set at 30 °C. Stage 7. The reactor tem-
perature was set to 40 °C and pressure to 0.5 bar over 1 h.

C. A. Carpenter et al. / Carbohydrate Research 376 (2013) 29–36
35
was stirred at room temperature (24 h) after which it was concen-
trated to a thick syrup (30 °C). The syrupy residue was dissolved in
methanol (200 mL) and the solution concentrated at 30 °C to re-
move residual HCl. The resulting syrup residue was dissolved in
methanol (200 mL) and methylamine (187.98 mL, 1.51 mol) was
added. The solution was stirred at room temperature (24 h) with
a precipitate beginning to form within 30 min. The precipitate
was removed by filtration, washed with methanol (3 ꢁ 10 mL),
and dried under vacuum to give N,N0-dimethyl-d-mannaramide
(5, 92.8 g, 393 mmol, 52.4%): mp 219–221 °C. 1H NMR (D2O) d
2.74 (s, 6H, NH–CH3) d 3.90 (d, 2H, H-2, H-5, J2,5 = 6.35 Hz) d 4.17
(d, 2H, H-3, H-4, J3,4 = 5.86 Hz). 13C NMR (DMSO-d6) d 25.31
(NH–CH3) d 70.99 (C3, C4) d 71.54 (C2, C5) d 173.84 (C1, C6). Anal.
Calcd for C8H16N2O6 (236.22): C, 40.68; H, 6.83; N, 11.86. Found: C,
40.97; H, 6.98; N, 11.79.
trated at 30 °C to remove residual HCl. The syrupy residue was then
dissolved in abs ethanol (400 mL), methylamine (188 mL, 1.51 mol)
was added and the solution stirred at room temperature (24 h). The
precipitate formed was removed by filtration, washed with abs etha-
nol (3 ꢁ 10 mL) and dried under vacuum to give N,N0-dimethyl-
D-
mannaramide (5, 62.4 g, 264 mmol, 35.2%, 60.0% based on oxidized
D-mannose diffusion dialysis product stream).
3.12. Disodium -mannarate (6)
D
Sodium hydroxide (2 M, 22.56 mL, 450 mmol) was added drop-
wise to a solution of N,N0-dimethyl-
-mannaramide (1, 3.52 g,
D
1.50 mmol) dissolved in water (70 mL) and the resulting solution
heated at 60 °C with stirring for 48 h. The solution was then con-
centrated under vacuum at room temperature and triturated with
methanol. Disodium D-mannarate was isolated by vacuum filtra-
3.11. N,N0-Dimethyl
diffusion dialysis
D-mannaramide (5)—isolation after
tion as an amorphorous white solid (6, 3.71 g, 1.46 mmol, 98%).
1H NMR (D2O) d 4.07 (d, 2H, H-2, H-5, J2,5 = 5.86 Hz) d3.90 (d, 2H,
H-3, H-4, J3,4 = 5.86 Hz). Anal. Calcd for C6H8Na2O8ꢂ(H2O)2
(290.13): C, 24.84; H, 4.17. Found: C, 24.62; H, 3.66. The disodium
mannarate was observed as a single component at retention time
10.964 min by ion chromatography and 44.77 min by GC–MS
([Mꢀ15]+ of the trimethylsilyl derivative m/z 627).
Two oxidation procedure 3 reaction mixtures were combined
and diluted to a volume of 2000 mL (RO water). The Mech-Chem
Diffusion Dialysis Acid Purification System acid reject pump was
set at 30% (pump length) and 30% (pump seed) and the acid re-
claim pump was set at 40% (pump length) and 40% (pump seed).
The acid tank was filled with the diluted aqueous oxidation mix-
tures and the water tank filled with RO water. The acid purification
unit was turned on with the pumps set as indicated and the pro-
cess initiated. Over a period of 21 h the acid recovery stream and
product recovery stream were collected. The acid recovery stream
was enriched in nitric acid (ca. 92% via IC analysis) and the product
3.13. Collection of X-ray diffraction data and solution of the
crystal structure for N,N0 dimethyl-
D-mannaramide (5)
All software used was from Bruker AXS, Inc. Data collection,
indexing and initial cell refinements were all carried out using
Table 1
recovery stream contained was enriched in the oxidized D-man-
Crystal data and structure refinement for N,N0-dimethyl-D-mannaramide (5)
nose product. The product recovery stream was concentrated to a
thick syrup and dissolved in water (500 mL). Based on IC analysis,
the enriched oxidized carbohydrate product represented ca. 66% of
Identification code
N,N0-Dimethyl-
D-
mannaramide (5)
total oxidized
stock solution. The solution was then divided into two equal ali-
quots, and which were processed with methanol and
D-mannose product and 8% nitric acid from the feed-
Empirical formula
Formula weight
Temperature
Wavelength
Crystal system
Space group
C8H16N2O6
236.23
173(2) K
1.54178 Å
Monoclinic
P2(1)
A
B
absolute ethanol, respectively.
3.11.1. Methanol workup
Unit cell dimensions
a = 8.1098(3) Å
b = 5.6630(2) Å
c = 11.8507(4) Å
518.09(3) Å3
2
a
= 90°
b = 107.838(1)°
= 90°
Aliquot A was concentrated to a thick syrup which was dis-
solved in methanol (600 mL) and acetyl chloride (26. 79 mL,
377 mmol) added dropwise (ice bath) to the solution with stirring.
The solution was stirred at room temperature for 24 h and then
concentrated to a thick syrup under reduced pressure at 30 °C.
The syrup was dissolved in methanol (600 mL), the addition of
acetyl chloride with stirring was repeated as was concentration
of the solution under reduced pressure. The syrupy residue was
dissolved in methanol (200 mL) and concentrated again at 30 °C
to remove residual HCl. The syrupy residue was dissolved in meth-
anol (200 mL) and methylamine (188 mL, 1.51 mol) was added.
This solution was stirred at room temperature (24 h) and the pre-
cipitate removed by filtration was washed with methanol
c
Volume
Z
Density (calculated)
Absorption coefficient
F(000)
Crystal size
Theta range for data
collection
1.514 mg/m3
1.119 mmꢀ1
252
0.42 ꢁ 0.24 ꢁ 0.11 mm3
3.92–65.97°
Index ranges
ꢀ9 6 h 6 9, ꢀ6 6 k 6 6,
ꢀ13 6 l 6 13
3953
1541 [R(int) = 0.0095]
96.3%
Reflections collected
Independent reflections
Completeness to
theta = 65.97°
Absorption correction
(3 ꢁ 10 mL) and dried under vacuum to give N,N0-dimethyl-
D-
Semi-empirical from
equivalents
0.8868 and 0.6507
mannaramide (5, 54.7 g, 232 mmol, 30.9%, 52.3% based on oxidized
Max. and min.
transmission
D
-mannose diffusion dialysis product stream).
Refinement method
Full-matrix least-squares on
3.11.2. Ethanol workup
F2
Data/restraints/
parameters
Goodness-of-fit on F2
Final R indices
[I>2sigma(I)]
R indices (all data)
Absolute structure
parameter
1541/1/152
Aliquot B was concentrated to a thick syrup; the syrup was dis-
solved in abs ethanol (800 mL) and acetyl chloride (26.79 mL,
377 mmol) added dropwise (ice bath) with stirring. The resulting
solution was stirred at room temperature (24 h) and then concen-
trated to a thick syrup at 30 °C. The syrup was dissolved inabs ethanol
(400 mL) and the addition of acetyl chloride and concentration steps
carried out as with the methanol workup procedure. The syrupy res-
idue was dissolved in abs ethanol (400 mL) and the solution concen-
1.093
R1 = 0.0252, wR2 = 0.0676
R1 = 0.0253, wR2 = 0.0678
0.38(19)
Largest diff. peak and hole 0.143 and ꢀ0.200 e Åꢀ3

36
C. A. Carpenter et al. / Carbohydrate Research 376 (2013) 29–36
APEX16 software. Frame integration and final cell refinements were
done using SAINT17 Version 6.45A software. Structure solution,
refinement, graphics, and generation of publication materials were
performed by using SHELXTL18 V6.12 software. A suitable crystal of
obtained free of charge from the Director, Cambridge Crystallo-
graphic Centre, 12 Union Road, Cambridge, CB2 1EZ, UK. (Fax:
+44 1223 336033, email: deposit@ccdc.cam.ac.uk or via: http://
www.ccdc.cam.ac.uk.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.carres.2013.05.
004.
N,N0-dimethyl-
D-mannaramide (5) was coated with Paratone N oil,
suspended in a small fiber loop and placed in a cooled nitrogen gas
stream at 173K on a Bruker D8 APEX II CCD sealed tube diffractom-
eter with graphite monochromated CuK (1.54178 Å) radiation.
a
Data were measured using a series of combinations of phi and
omega scans with 10 s frame exposures and 0.5° frame widths.
The structure was solved using Direct methods and difference
Fourier techniques (SHELXTL,18 V6.12, Bruker and Mercury
1.4.219 software). Hydrogen atoms were placed at their expected
chemical positions using the HFIX command and were included
in the final cycles of least squares with isotropic Uij0s related to
the atom0s ridden upon. All non-hydrogen atoms were refined
anisotropically. Scattering factors and anomalous dispersion cor-
rections are taken from the International Tables for X-ray Crystal-
lography.20 Crystal data and structure refinement for N,N0-
References
1. Hashimoto, K.; Wibullucksanakul, S.; Matsuura, M.; Okada, M. J. Polym. Sci., Part
A: Polym. Chem. 1993, 31, 3141–3149.
2. Kiely, D. E.; Chen, L.; Lin, T.-H. J. Polym. Sci., Part A: Polym. Chem. 1999, 38, 594–
603.
3. Carpenter, C.; Kiely, D. E. Abstracts of Papers, 62nd Northwest Regional Meeting
of the American Chemical Society, Boise, ID, United States, June 2007; NW-114.
4. Fischer, E. Ber. Dtsch. Chem. Ges. 1891, 539–549.
5. Easterfield, T. H. J. Chem. Soc., Trans. 1891, 59, 306–310.
6. Haworth, W. N.; Heslop, D.; Salt, E.; Smith, F. J. Chem. Soc. 1944, Part 1, 217–224.
7. Hirasaka, Y.; Umemoto, K. Chem. Pharm. Bull. 1965, 13, 325–329.
8. Kiliani, H. Ber. Dtsch. Chem. Ges. 1943, 76B, 540–541.
9. Smith, T. N.; Hash, K.; Davey, C.-L.; Mills, H.; Williams, H.; Kiely, D. E. Carbohydr.
Res. 2012, 350, 6–13.
10. Heslop, D.; Smith, F. J. Chem. Soc. 1944, 577–584.
dimethyl-D-mannaramide (5) are shown in Table 1.
11. Cotton, F. A.; Wilkinson, G. W. Advanced Inorganic Chemistry; Interscience
Publishers: New York, 1967. pp 331–353.
Acknowledgments
12. Carpenter, C. A; The University of Montana: Ph.D. dissertation, 2008.
13. Hinton, M. R.; Manley-Harris, M.; Hardcastle, K. I.; Kiely, D. E. J. Carbohydr.
Chem. 2013, 31, 68–85.
Partial support for this research was from the U.S. Department
of Agriculture (USDA-CSRESS), Grant Nos. 2005-34463-11561 and
2006-34463-16886 and is gratefully acknowledged.
14. Petersson, G. Org. Mass Spectrom. 1972, 6, 565–576.
15. Kiely, D. E.; Ponder, G. U.S. Patent 6,049,004, 2000; Chem. Abstr. 2000, 238407.
16. APEX II, 2005, Bruker AXS, Inc., Analytical X-ray Systems, 5465 East Cheryl
Parkway, Madison WI 53711–5373.
17. SAINT Version 6.45A, 2003, Bruker AXS, Inc., Analytical X-ray Systems, 5465
East Cheryl Parkway, Madison WI 53711–5373.
Supplementary data
18. SHELXTL V6.12, 2002, Bruker AXS, Inc., Analytical X-ray Systems, 5465 East
Cheryl Parkway, Madison WI 53711–5373.
19. Macrae, C. F.; Edgington, P. R.; McCabe, P.; Pidcock, E.; Shields, G. P.; Taylor, R.;
Towler, M.; van de Streek, J. J. Appl. Cryst. 2006, 39, 453–457.
20. International Tables for X-ray Crystallography, Volume C; Wilson, A. J. C., Ed.;
Kynoch Academic: Dordrecht, 1992. pp 219–222 and 500–502.
Complete crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic Data Centre,
CCDC No. 826399. These data may be obtained on request from
the Director, Cambridge Crystallographic Data Centre, 12 Union
Road, Cambridge CB2 IEZ, UK. Copies of this information may be