Synthesis, Thermochemical, and Thermomechanical Characterization of High Conductivity 4-Azoniaspiro[3,4]octane Salts

Full text of DOI:10.1080/00304948.2018.1462061

Organic Preparations and Procedures International
The New Journal for Organic Synthesis
ISSN: 0030-4948 (Print) 1945-5453 (Online) Journal homepage: http://www.tandfonline.com/loi/uopp20
Synthesis, Thermochemical, and
Thermomechanical Characterization of High
Conductivity 4-Azoniaspiro[3,4]octane Salts
Seiichiro Higashiya, Adam P. Schulz, Donald M. DeRosa, Manisha V. Rane-
Fondacaro & Pradeep Haldar
To cite this article: Seiichiro Higashiya, Adam P. Schulz, Donald M. DeRosa, Manisha V. Rane-
Fondacaro & Pradeep Haldar (2018) Synthesis, Thermochemical, and Thermomechanical
Characterization of High Conductivity 4-Azoniaspiro[3,4]octane Salts, Organic Preparations and
Procedures International, 50:3, 323-331, DOI: 10.1080/00304948.2018.1462061
To link to this article: https://doi.org/10.1080/00304948.2018.1462061
Published online: 17 May 2018.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=uopp20

Organic Preparations and Procedures International, 50:323–331, 2018
Copyright Ó Taylor & Francis Group, LLC
ISSN: 0030-4948 print / 1945-5453 online
DOI: 10.1080/00304948.2018.1462061
Synthesis, Thermochemical, and Thermomechanical
Characterization of High Conductivity
4-Azoniaspiro[3,4]octane Salts
Seiichiro Higashiya, Adam P. Schulz, Donald M. DeRosa,
Manisha V. Rane-Fondacaro, and Pradeep Haldar
Energy and Environmental Technology Applications Center (E2TAC), Colleges
of Nanoscale Science and Engineering (CNSE), SUNY Polytechnic Institute,
257 Fuller Road, Albany, NY 12203, USA
Electrolytes based on the spiro-type quaternary ammonium salt, 1,10-spirobipyrrolidinium
tetrafluoroborate (SBP BF4, 4-azoniaspiro[4.4]nonane tetrafluoroborate, Figure 1) are
widely being investigated for electric double layer capacitor (EDLC) applications owing
to their high conductivity.1-6 The SBP cation comprises two symmetric 5-membered rings
joined together by a NC functionality. Replacing one of the 5-membered rings with a 4-
membered ring results in the 1-azetidyl-10-spiropyrrolidinium cation (AP, 4-azoniaspiro
[3.4]octane, Scheme 1). The highest conductivity among all the quaternary ammonium
salts has been reported for AP salts in several patents.7-10 However, there are no reports
on a facile synthesis of AP salts nor on the thermochemical properties of high purity AP
salts. In this paper, we report a novel method for the synthesis of AP salts (BF4, PF6 and
NTf2), and compare and contrast properties of AP BF4, AP PF6, AP NTf2 with these of
SBP BF4.
Figure 1. Structures and abbreviations of salt components.
AP Cl was synthesized from pyrrolidine and 1-bromo-3-chloropropane via a two-step
method as shown in Scheme 1.
Received October 30, 2016; in final form September 28, 2017.
Address correspondence to Manisha V. Rane-Fondacaro, Energy and Environmental
Technology Applications Center (E2TAC), Colleges of Nanoscale Science and Engineering
(CNSE), SUNY Polytechnic Institute, 257 Fuller Road, Albany, NY 12203.
E-mail: ranemv@gmail.com
Color versions of one or more figures in this article can be found online at
www.tandfonline.com/uopp.
323

324
Rane-Fondacaro et al.
Scheme 1.
We evaluated and modified two methods for intramolecular cyclization of 1-(3-chlor-
opropyl)pyrrolidine under diluted homogeneous and non-homogeneous conditions; both
had been previously described in a patent for the preparation of AP Cl and AP PF6,
respectively.11 Under homogeneous conditions, the reactant concentration was deliber-
ately maintained at a low level to minimize formation of byproducts, particularly the
“intractable” byproduct.12 Despite addition of 1-(3-chloropropyl)pyrrolidine dissolved in
EtOH in a dropwise manner to boiling EtOH for maintaining low concentration, a signifi-
cant amount of byproducts was formed that reduced the conversion rates and required
additional purification. Non-homogeneous conditions using hot H2O as a solvent exclu-
sively produced AP Cl; however, this required optimization of the amount of boiling
water. Using the amount of H2O stated in the original patent11 frequently lead to forma-
tion of homogeneous mixtures upon addition of 1-(3-chloropropyl)pyrrolidine that
resulted in low yields with large amounts of byproducts. Using five times more H2O than
the original procedure ensured formation of two-phases, and the existence of non-homo-
geneous condition upon addition of 1-(3-chloropropyl)pyrrolidine.
Since 1-(3-chloropropyl)pyrrolidine can be isolated by vacuum distillation without
significant decomposition, the intra-/inter-molecular reactions proceed preferably in polar
media. Low solubility of 1-(3-chloropropyl)pyrrolidine in H2O appears to facilitate the
intramolecular cyclization, resulting in high yield and high purity. The utilization of low
solubility reactants in polar solvents for selective intramolecular reactions is applicable to
other intramolecular reactions which require highly diluted conditions.
Earlier attempts of AP Cl synthesis from 1-(3-chloropropyl)pyrrolidine failed, result-
ing in formation of large amounts of intractable byproduct.12 Cyclization of 1-(3-chloro-
propyl)pyrrolidine (13.8 g) in acetonitrile (850 ml) at 60–65ꢀC for 16 h has been
reported by Jin et al.13 However, our protocol (1-(3-chloropropyl)pyrrolidine (116.24 g)
in boiling H2O (400 mL) for 30 min) is green, faster, inexpensive, and scalable. Nishino
et al. disclosed a two-step preparation of AP BF4 starting from azetidine, 1-bromo-4-fluo-
robutane, and trifluoroborane ethylamine complex,7 however, the yield of 1-(4-fluorobu-
tyl)azetidine is low and the reagents are significantly more expensive compared to
pyrrolidine, 1-bromo-3-chloropropane, and aqueous tetrafluoroboric acid and thus unsuit-
able for the large scale synthesis.
For the AP salts, we measured the glass transition temperatures (DTg), melting points
(DTm), and related enthalpy (DHg and DHm, respectively) and entropy changes (DSg and
DSm, respectively) using differential scanning calorimetry (DSC). The data are summa-
rized in Table 1.
Both AP BF4 and AP NTf2 exhibit significantly higher total phase enthalpy change
(DHtotal D DHgCDHm) as compared to the reported value for the loosely-packed SBP
BF4 crystal (8.9 kJ/mol). Moreover, their DHtotal values are closer to that of 3,30-spiro-
bioxazolidinium tetrafluoroborate (SBO BF4, 24.6 kJ/mol) which is polar and forms a
tightly packed crystal (Figure 1).14,15 This closeness of DHtotal values of AP salts to SBO

4-Azoniaspiro[3,4]octane Salts
325
Table 1.
Thermodynamic Parameters of AP Salts.
AP BF4
AP NTf2
AP PF6
MW (gmol¡1
)
199.00
42.0
8.1
392.34
24.3
14.4
257.16
128.4
2.6
Tg (ꢀC)
DHg (kJmol¡1
)
DSg (JK¡1mol¡1
Tm (ꢀC)
)
23.7
206.3
15.3
30.3
23.4
44.5
6.0
111.4
10.5
25.5
DHm (kJmol¡1
)
DSm (JK¡1mol¡1
DHtotal (kJmol¡1
)
)
24.9
BF4 is attributed to smaller ion size and close packing of AP salts. Interestingly, AP PF6
started decomposing before melting and exhibited stable crystal behavior despite the ther-
mal instability of PF6 anion.
The crystallographic parameters and calculated thermodynamic parameters (UPOT
(potential energy) and DlattHꢀ (lattice enthalpy)) consistent with volume-based thermody-
namics (VBT) theory15 are summarized in Table 2.
Both crystals of AP BF4 and SBP BF4 belong to R3c space group. The smaller molecular
unit volume (Vm) of AP BF4 (0.238 nm3 vs. 0.261 nm3 for SBP BF4) leads to stronger charge-
charge interactions in the closer-packed AP BF4 crystal that results in a difference of
11.6 kJmol¡1 in lattice formation energy (DlattHꢀ) per VBT theory. This also explains the dif-
ference in the total phase change enthalpy of 14.5 kJmol¡1 between AP BF4 and SBP BF4.
Figure 2 shows the thermogravimetric analysis (TGA) spectra of SBP BF4, AP BF4,
AP NTf2, and AP PF6.
Among the AP salts, the onset temperatures for weight loss is lowest in case of AP
BF4, followed by AP NTf2 and AP PF6. While the highest onset temperature in the case
Table 2.
Crystallographic and VBT Parameters of AP, SBP and SBO Salts at 100 K.
13
13
Salt
AP BF4
AP NTf2
SBP BF4
SBO BF4
Crystal system
Space group
Rhombohedral Orthorhombic Rhombohedral Orthorhombic
R3c
19.9273(7)
19.9273(7)
12.4592(5)
90
P212121
8.7228(10)
13.5149(16)
38.143(4)
90
R3c
20.791(3)
20.791(3)
12.561(4)
90
Pna21
12.0719(11)
6.4498(6)
11.6510(11)
90
ꢀ
a (A)
ꢀ
b (A)
ꢀ
c (A)
a
b
90
120
90
90
90
120
90
90
g
V
Z
ꢀ
cell (A3)
4284.67
18
238
4496.56
12
375
4702.25
18
261
907.16(15)
4
227
ꢀ
Vm (A3)
Vm (nm3)
0.238
482.3
487.3
0.375
429.2
434.2
0.261
470.8
475.7
0.227
488.5
482.8
U
D
POT (kJmol¡1
lattHꢀ (kJmol¡1
)
)

326
Rane-Fondacaro et al.
100
80
60
40
20
0
––––––– SBP BF4
– – – –
–– –– –
––– – –
AP BF4
AP NTf2
AP PF6
0
200
400
Temperature (°C)
Figure 2. TGA analysis of AP and SBP salts.
of AP PF6 is presumably due to its stable crystal structure, there is no correlation between
the melting points (or glass transition temperatures) and onset temperatures for weight
loss in AP BF4 and AP NTf2 salts. Comparison of temperatures at which maximum
weight loss is observed (Tmwl, the temperature at the maximum of the first derivative of
the weight loss) for SBP BF4 and AP BF4 shows that the weight loss occurs at signifi-
cantly higher temperature in SBP BF4 (at 388ꢀC) as compared to AP BF4 (at 348ꢀC). The
lower weight loss temperature in AP BF4 is attributed to greater instability of the four-
membered ring of the AP cation.
We performed quantum chemical calculations to evaluate the strain in the 4- and 5-mem-
bered rings. The energies of spiro-structures (AP and SBP) were compared with ring-opened
structures by b-elimination (allyl-P and homoallyl-P, respectively, Figure 3), and the energy
differences between the closed and open structures are summarized in Table 3.
The DEopen-close of AP cation is negative due to the release of strained 4-membered
ring while the DEopen-close of the SBP cation is positive due to replacement of one sigma
bond with one double bond. The difference in ring strain between 4- and 5-membered
rings is calculated to be 14.1 kcalmol¡1. It has been reported the ring strain energies for
cyclobutane and cyclopentane are 26.5 and 6.2 kcal.mol¡1, respectively.16 Our estimate
is slightly smaller than this difference (20.3 kcal.mol¡1), probably due to presence of NC
in the SBP and AP cation.
Figure 3. Spiro- and ring-opened and refluxed structures of AP and SBP cation.

4-Azoniaspiro[3,4]octane Salts
327
Table 3.
Calculated Energy Differences of Spiro- and One-ring-opened and Relaxed AP
and SBP Ions.
DEopen-close
(kJmol¡1
)
(kcalmol¡1
)
AP
SBP
-16.2
42.8
59.0
-3.9
10.2
14.1
DESBP-AP
The room temperature electrical conductivity of various salts (AP BF4, AP PF6, and
SBP BF4) dissolved in multiple solvents in molarities ranging from 0.5 to 4.0 M in steps
of 0.5 is summarized in Figure 4.
Comparison of the conductivity of AP BF4 and AP PF6 in acetonitrile (AN) shows
significantly higher conductivity for AP BF4 /AN (44 to 70 mS/cm) than for AP PF6 /AN
(37 to 60 mS/cm). Moreover, AP BF4 has greater solubility in AN (up to 4 M) than AP
PF6 (up to 2.5 M) which is an important consideration from the electrolyte perspective.
The conductivity of SBP BF4 /AN ranged from 44 mS/cm to 65 mS/cm (at various molar-
ities) and was lower than that of AP BF4 /AN; nevertheless both showed conductivity
maxima in the same concentration range (between 2 and 2.5 M) and high solubility up to
4 M. The conductivity maxima for AP PF6 /AN was observed between 1.5 M and 2.0 M.
The conductivity of AP BF4 in propylene carbonate (PC, a common solvent used in elec-
trochemical capacitors) was significantly poor as compared to all other solvent systems
investigated.
The dissolution of AP BF4 in two low polarity and viscous solvents, namely,
dimethyl carbonate (DMC) and dichloromethane (DCM) at salt concentrations ꢀ2.5 M in
DMC and ꢀ2.0 M in DCM, resulted in the formation of homogeneous solutions; how-
ever, the conductivity was poor. This low conductivity was attributed to the clustering
behavior of ions in less polar solvents and was confirmed by the broad peaks observed in
80
70
60
50
40
30
20
10
0
0
1
2
3
4
Concentration (M)
AP BF4/AN
AP BF4/DCM
AP PF6/AN
AP BF4/EC
AP BF4/PC
SBP BF4/AN
AP BF4/DMC
SBP BF4/EC
Figure 4. Conductivity of solutions of AP and SBP salts at 25 oC.

328
Rane-Fondacaro et al.
the NMR spectrum of AP BF4 (chloroform-d). In ethylene carbonate (EC), both AP BF4
and SBP BF4 at ꢀ1.5 M concentration exhibited nearly identical conductivity and they
both formed eutectic mixtures.
In conclusion, we successfully synthesized high purity AP salts (> 99.9%) for elec-
trochemical energy storage applications in good yields using inexpensive and scalable
synthesis protocols. Intramolecular cyclization of 1-(3-chloropropyl)pyrrolidine under
non-homogeneous conditions exclusively produced AP Cl that eliminated prolonged and
expensive purification protocols for high purity AP salts. The cyclization under homoge-
neous conditions produced large amounts of byproducts, which were difficult to remove
by recrystallization due to the low solubility. This non-homogeneous intramolecular reac-
tion is applicable for other intramolecular reactions, which need high dilution of reactants
for preventing competitive intermolecular reactions.
The AP salts were less stable than their SBP counterparts, but showed higher conduc-
tivity, especially in acetonitrile. The stability was closely dependent on the anion type.
We investigated thermomechanical properties using volume based thermodynamic
modeling, TGA and DSC; and we studied the thermochemical properties of AP salts and
compared their properties with SBP BF4 and SBO BF4.
Experimental Section
1-(3-Chloropropyl)pyrrolidine was prepared by the reaction of pyrrolidine and 1-
bromo-3-chloropropane according to the literature procedure.17 1-Bromo-3-chloro-
propane and pyrrolidine were obtained from Acros Organics (NJ, USA). 50% HBF4
was purchased from GFS Chemicals (Powell, OH, USA). Aluminum oxide (acti-
vated, acidic, Blockmann Grade I, 58 angstroms) was purchased from Alfa Aesar
(Ward Hill, MA, USA). Lithium bis(trifluoromethanesulfonyl)imide (LiNTf2) was
purchased from TCI America (Portland, OR, USA). Lithium hexafluorophosphate
(LiPF6) was purchased from BASF Battery Materials (Iselin, NJ, USA). Nylon mem-
brane, Whatman 7402-004, 0.2 mm, was purchased from GE Healthcare Life Scien-
ces (Pittsburgh, PA, USA). Elemental analyses were performed by M-H-W
1
Laboratories, Phoenix, AZ. H (400.1 MHz), 13C (100.6 MHz), 19F (376.5 MHz) and
31P (162.0 MHz) NMR were measured on Bruker Avance 400 spectrometer. NMR
spectra of salts were measured in CD3OD (Cambridge Isotope Laboratories, Inc.,
Tewksbury, MA, USA) unless otherwise specified, and the solvent peaks of 1H
(3.31 ppm) and 13C NMR (49.0 ppm) and external CFCl3 and H3PO4 peaks of 19F
NMR and 31P NMR were used as references for the chemical shifts. Conductivity
was measured using a Radiometer Analytical CDM230 Conductivity Meter at 25ꢀC.
TGA data were recorded on a TA Instruments TGA 2050 thermogravimetric ana-
lyzer under dynamic nitrogen flow (total flow rate 100 cm3/min), using platinum
pans and a ramp rate of 10ꢀC/min. DSC data were recorded on a TA Instruments
DSC 2920 differential scanning calorimeter at 1 atm under dynamic nitrogen, using
hermetically sealed aluminum pans, with a ramp rate of 10ꢀC/min. Unit cell parame-
ters indexing of AP BF4 and AP NTf2 were performed on a Bruker D8 VENTURE
X-ray diffractometer with PHOTON 100 CMOS detector equipped with a Mo-target
ꢀ
X-ray tube (λ D 0.71073A) at T D 100(§2) K. Quantum chemical calculations were
carried out using the Firefly QC package (version 8.1.1, build number 9295)18 at
restricted B3LYP density function, using 6–311 basis set with s diffuse and p polari-
zation functions for hydrogen and sp diffuse and d polarization functions for the
other elements (similar to RB3LYP/6-311CCG(d,p) level).

4-Azoniaspiro[3,4]octane Salts
329
Synthesis of 1-azetidyl-10-spiropyrrolidinium (4-azoniaspiro[3.4]octane)
tetrafluoroborate (AP BF4)
Two methods were examined, namely non-homogeneous and homogeneous, for the syn-
thesis of AP Cl as described in a patent.11
A (non-homogenous method): 1-(3-Chloropropyl)pyrrolidine (116.24 g, 0.80 mol)
was added dropwise to boiling H2O (400 mL) for 15 min and further refluxed for addi-
tional 15 min. The resulting homogenous and pale yellow solution was washed with three
portions of CH2Cl2 (50, 30, 30 mL) and then the volatiles were removed by evaporation
yielding a yellow oil of crude AP Cl. Assuming 100% conversion by NMR analysis, 50%
aqueous HBF4 (138.39 g, 0.80 mol) and EtOH (150 mL) were added and the volatile
materials were removed by evaporation. Addition of EtOH (150 mL) and evaporation
were repeated two more times and the residue was recrystallized from EtOH (1.0L) to
give the crude product. The crude product was dissolved in CH2Cl2 (1.5 L) and applied
to acidic aluminum oxide (150 mL) column and fractions containing the product were
collected. The column was further eluted with CH2Cl2 (500 mL) and the combined eluent
was filtered using a Nylon membrane (Whatman 7402-0004, 0.2 mm). The volatiles were
removed by evaporation and the residue was recrystallized from EtOH (1.0 L) to give the
product (143.3 g, 92% yield). The recrystallization was repeated and the purified product
1
ꢀ
was dried in vacuo at 70 C for 12 h. H NMR (methanol-d4) d D 4.33 (t, 4 H, J D
8.2 Hz), 3.62 (m, 4 H), 2.60 (quintet-t, 2 H, J D 8.2, 2.2 Hz), 2.09 (tt, 4 H, J D 7.1,
1.5 Hz); 19F NMR d D -154.52, -154.57 (2 s, 19:81 (10B:11B)); 13C NMR d D 64.1 (br),
22.1, 15.3.
Anal. Calcd for C7H14BF4N: C, 42.25; H, 7.09; N, 7.04; Found: C, 41.99; H, 7.00; N, 6.75.
B (homogenous method): 1-(3-Chloropropyl)pyrrolidine (182.1 g, 1.23 mol) was dis-
solved in 3.75 L of EtOH and added to boiling EtOH (250 mL) over 2 h and further refluxed
for 1 h. The solvent was removed by evaporation and the residue was dissolved in H2O
(400 mL). The aqueous phase was washed with three portions of CH2Cl2 (150, 100, 100 mL).
The volatiles were removed by evaporation and then dried under vacuum at ambient tempera-
ture overnight while stirring continuously, resulting in a yellow oil of crude AP Cl. The
CH2Cl2 layer was dried over anhydrous MgSO4. The drying reagent was removed by filtration
and solvent was removed by evaporation to recover unreacted 1-(3-chloropropyl)pyrrolidine
(8.4 g, 5%, therefore the maximum yield of AP Cl was 95%, 1.17 mol). The crude AP Cl thus
obtained was mixed with MeOH (250 mL) which partially dissolved the salt. The Cl salt was
converted to the BF4 salt by addition of 50% aqueous HBF4 (206.3 g, 1.17 mol) and the vola-
tile materials in the resulting cloudy solution were removed by evaporation. EtOH (200 mL)
was added and evaporated for azeotropic removal of aqueous HCl. This azeotropic removal of
water and HCl was repeated two more times and the residue was recrystallized from EtOH
(1 L) (insoluble materials were removed by filtration while the solution was hot). The resulting
crystalline solid (216.3 g) was dissolved in CH2Cl2 at a concentration less than 50 g/L and
applied to an acidic alumina (»200 mL) column. The product was eluted with CH2Cl2, frac-
tions containing AP BF4 collected, and the solvent was removed by evaporation. It may be nec-
essary to repeat the column chromatography to obtain sufficiently pure fractions. There needs
to be less than 1% of byproduct peak at 2.2 ppm, compared to product peak at 2.1 ppm, in 1H
NMR for the subsequent efficient removal by recrystallization. The product (148.5 g) was
repeatedly recrystallized from EtOH (1.6 L) until the byproduct was undetectable in the con-
centrated filtrate to give pure product after the fourth recrystallization. The peak integration
ratio of 2.2 ppm to 2.1 ppm was less than 0.05% as per 1H NMR which resulted in 140.5 g,
57% yield of 99.95% pure salt.

330
Rane-Fondacaro et al.
Synthesis of 1-azetidyl-10-spiropyrrolidinium (4-azoniaspiro[3.4]octane)
bis(triflluoromethanesulfonyl)imide (AP NTf2)
AP BF4 (47.53 g, 0.238 mol) and LiNTf2 (68.57 g, 0.238 mol) were mixed with H2O
(100 mL) and CH2Cl2 (50 mL) for 10 min. This resulted in the separation of the organic
phase, which was washed with 5 portions of H2O and subsequently filtered using a Nylon
membrane (Whatman 7402-0004, 0.2 mm). The volatiles were removed by evaporation
and the residue was repeatedly recrystallized from EtOH or isopropyl alcohol to give the
product (84.8 g, 91% yield). The product turned gummy after drying in vacuo at 70ꢀC for
12 h due to the low glass transition temperature.1H NMR (methanol-d4) d D 4.33 (t, 4 H,
J D 8.2 Hz), 3.62 (m, 4 H), 2.61 (quintet-t, 2 H, J D 8.2, 2.2 Hz), 2.09 (m, 4 H); 19F
NMR d D -81.10; 13C NMR d D 121.10 (q, J D 320.5 Hz), 64.1 (br), 22.0, 15.2.
Anal. Calcd for C7H14F6N2O4S2: C, 27.55; H, 3.60; N, 7.14; Found: C, 27.62; H,
3.54; N, 6.97.
Synthesis of 1-azetidyl-10-spiropyrrolidinium (4-azoniaspiro[3.4]octane)
hexafluorophosphate (AP PF6)
Crude AP Cl (prepared on 0.121 mol scale) was dissolved in H2O (50 mL, the total
weight with AP Cl was about 60 g), and LiPF6 (20.22 g, 0.133 mol) dissolved in H2O
(50 mL) was added dropwise to the AP Cl solution at ambient temperature. This resulted
in immediate formation of a white precipitate, which was collected by filtration. After
brief drying under vacuum at ambient temperature, the wet product was dissolved in hot
H2O and filtered by Nylon membrane (Whatman 7402-004, 0.2 mm) and subsequently
recrystallized. The product was further recrystallized from H2O and dried in vacuo at
70ꢀC for 12 h to give the product (24.8 g, 80% yield).1H NMR (methanol-d4) d D 4.33 (t,
4 H, J D 8.2 Hz), 3.62 (m, 4 H), 2.60 (quintet-t, 2 H, J D 8.2, 2.2 Hz), 2.09 (m, 4 H); 19
F
NMR d D -74.93 (d, J D 708.3 Hz); 13C NMR d D 64.1 (br), 22.1, 15.3; 31P NMR d D
144.57 (septet, J D 707.6 Hz).
Anal. Calcd for C7H14F6NP: C, 32.69; H, 5.49; N, 5.45; Found: C, 32.40; H, 5.22; N,
5.27.
Acknowledgments
We acknowledge Prof. John T. Welch and Dr. Wei Zheng of the Department of Chemis-
try, University at Albany, for crystallographic measurements and helpful discussions.
Purchasing of the diffractometer was supported by an NSF award (award number:
1337594).
References
1. X. Yu, D. Ruan, C. Wu, J. Wang, and Z. Shi, J. Power Sources, 265, 309 (2014).
2. K. Chiba, T. Ueda, Y. Yamaguchi, Y. Oki, F. Shimodate, and K. Naoi, J. Electrochem. Soc.,
158, A872 (2011).
3. K. Chiba, T. Ueda, Y. Yamaguchi, Y. Oki, F. Saiki, and K. Naoi, J. Electrochem. Soc., 158,
A1320 (2011).
4. N. M. Jackson and M. Payne, ECS Transactions, 16, 139 (2008).
5. K. Chiba, T. Ueda, and H. Yamamoto, Electrochemistry, 75, 664 (2007).

4-Azoniaspiro[3,4]octane Salts
331
6. K. Chiba, T. Ueda, and H. Yamamoto, Electrochemistry, 75, 668 (2007).
7. S. Nishino, H. Shima, Y. Yoneda, and H. Oda, JP 2009023946, 2009.
8. Y. Watanuki, M. Nakano, and S. Tanabe, JP 2006024785, 2006.
9. Y. Watanuki and S. Tanabe, JP 2005347176, 2005.
10. K. Chiba and K. Naito, JP 2004269805, 2004.
11. E. A. Arnott, J. Crosby, M. C. Evans, J. G. Ford, M. F. Jones, K. W. Leslie, I. M. McFarlane,
and G. J. Sependa, WO 2008053221, 2008.
12. F. F. Blicke and E. B. Hotelling, J. Am. Chem. Soc., 76, 5099 (1954).
13. G. Jin, F. Jin, K. Wang, B. Zheng, Y. Fu, Z. Jin, and J. Liu, CN 106117217A, 2016.
14. S. Higashiya, A. S. Filatov, C. C. Wells, M. V. Rane-Fondacaro, and P. Haldar, J. Mol. Struct.,
984, 300 (2010).
15. S. Higashiya, T. S. Devarajan, M. V. Rane-Fondacaro, C. Dangler, J. Snyder, and P. Haldar,
Helv. Chim. Acta, 92, 1600 (2009).
16. E. V. Anslyn and D. A. Dougherty, “Modern Physical Organic Chemistry”, p. 101, University
Science, Sausalito, CA, 2006.
17. H. L. Yale, J. Am. Chem. Soc., 77, 2270 (1955).
18. A. A. Granovsky, Firefly version 8, http://classic.chem.msu.su/gran/firefly/index.html (accessed
on 2/3/2017).