Solid state NMR of SiO2 nanotube coated ammonium tartrate crystal

Article abstract of DOI:10.1016/S0038-1098(99)00046-0

Ammonium tartrate crystal and SiO₂ nanotube coated ammonium tartrate crystal were studied by ¹³C CP/MAS NMR, and the structure of two samples were verified using the ¹³C NMR spectrum. The spin-lattice relaxation times for the carbons in the rotating frame, T_1ρ, have been measured as a function of temperature. All relaxation times of the carbons in the two materials undergo slow motions, i.e. motions on the slow side of the T_1ρ minimum. From these relaxation times, we determine the activation energy for the ammonium tartrate crystal and SiO₂ nanotube coated ammonium tartrate crystal, respectively. The activation energies for the SiO₂ nanotube coated ammonium tartrate crystal were found to be generally higher than those of ammonium tartrate crystal. We think that the higher activation energy for the hydrocarbon in the SiO₂ nanotube coated ammonium tartrate crystal is because of the bonding between the oxygen in the SiO₂ nanotube and the hydroxyl group of the ammonium tartrate crystal.

Full text of DOI:10.1016/S0038-1098(99)00046-0

SSC 4606
PERGAMON
Solid State Communications 110 (1999) 333–338
Solid state NMR of SiO2 nanotube coated ammonium tartrate
crystal
a,
b
c
*
A.R. Lim , G.T. Schueneman , B.M. Novak
aDepartment of Physics, Jeonju University, Jeonju 560-759, South Korea
bDepartment of Polymer Science and Engineering and the Materials Research, Science and Engineering Center, University of Massachusetts,
Amherst, MA 01003, USA
cDepartment of Chemistry, North Carolina State University, Raleigh, NC 27695, USA
Received 4 September 1998; received in revised form 2 December 1998; accepted 30 December 1998 by A. Pinczuk
Abstract
Ammonium tartrate crystal and SiO2 nanotube coated ammonium tartrate crystal were studied by 13C CP/MAS NMR, and the
structure of two samples were verified using the 13C NMR spectrum. The spin–lattice relaxation times for the carbons in the
rotating frame, T1r , have been measured as a function of temperature. All relaxation times of the carbons in the two materials
undergo slow motions, i.e. motions on the slow side of the T1r minimum. From these relaxation times, we determine the
activation energy for the ammonium tartrate crystal and SiO2 nanotube coated ammonium tartrate crystal, respectively. The
activation energies for the SiO2 nanotube coated ammonium tartrate crystal were found to be generally higher than those of
ammonium tartrate crystal. We think that the higher activation energy for the hydrocarbon in the SiO2 nanotube coated
ammonium tartrate crystal is because of the bonding between the oxygen in the SiO2 nanotube and the hydroxyl group of
the ammonium tartrate crystal. ᭧ 1999 Elsevier Science Ltd. All rights reserved.
Keywords: A. Nanostructures; A. Polymers; B. Chemical synthesis; E. Nuclear resonances
1. Introduction
200–300 mm in length. The SiO2 nanotube coated
ammonium tartrate crystal was also obtained using
the sol–gel method at room temperature by the
Novak group [13]. The mechanism of fiber formation
has been recently elucidated. It was discovered that
upon addition of ammonium hydroxide TTA forms an
ammonium tartrate (ATT) salt which crystallizes in
the form of long prisms with rectangular cross-
sections. Simultaneous with crystal formation, SiO2
rapidly condenses and coats the crystal surfaces.
Washing the products with water dissolves the ammo-
nium tartrate crystals out from within the SiO2 coating
leaving a hollow tube. Washing the products with
ethanol leaves the ammonium tartrate crystals in
place yielding a SiO2 coated ATT crystal. Condensing
Inorganic hollow carbon [1–6], boron nitride [7–
9], and silica tubes [10,11] were prepared under high
temperature reaction conditions. Nakamura and
Matsui reported on the synthesis of SiO2 nanotubes,
which were synthesized in a solution of dl-tartaric
acid (TTA), water, absolute ethanol, and tetraethyl
orthosilicate (TEOS) via the addition of ammonium
hydroxide [12]. They had square shapes with an outer
diameter of 0.8–1.0 mm, inner diameters of 0.02–
0.8 mm, shell thicknesses of 0.015–0.3 mm, and
* Corresponding author. Fax: ϩ 82-652-220-2362.
E-mail address: aeranlim@hanmail.net (A.R. Lim)
0038-1098/99/$ - see front matter ᭧ 1999 Elsevier Science Ltd. All rights reserved.
PII: S0038-1098(99)00046-0

334
A.R. Lim et al. / Solid State Communications 110 (1999) 333–338
Fig. 1. The hollow SiO2 nanotubes by scanning electron microscopy.
SiO2 that does not encounter a crystal forms spherical
particles as is commonly obtained when using a strong
base as sol–gel catalyst [14]. The ammonium tartrate
crystals have a monoclinic structure with cell dimen-
Spectroscopic grade water and reagent grade ammo-
nium hydroxide (30% NH3) were obtained from
Fisher Scientific. All reagents were used without
further purification. Synthesis of the nanotubes was
carried out in a 50 ml Erlenmeyer flask by dissolving
30 mg of dl-TTA in a solution of 7.5 absolute ethanol
and 0.09 ml of water. When a clear room temperature
solution was obtained, 1.18 ml of TEOS was added
and the solution was allowed to stand for 30 min.
During this time the solution remained clear and
there was no noticeable change in viscosity. After
30 min, 3 ml of ammonium hydroxide were added
to the solution. After this addition, a cloudy white
layer formed on the top of the solution. The solution
became clear upon mixing and then increasingly
cloudy and viscous until it was completely opaque
and non-flowing. This state was achieved 15–
20 min after the addition of ammonium hydroxide.
The reaction scheme is show in Fig. 3. The precipitate
was washed with ethanol, collected via vacuum filtra-
tion on a 0.45 mm filter, and dried in an oven at 50ЊC
for 24 h.
˚
˚
˚
sions of a ˆ 7.083 A, b ˆ 6.128 A, c ˆ 8.808 A and
b ˆ 92.42Њ at room temperature [15]. The crystals are
known to have piezoelectric properties. Fig. 1 shows
the hollow SiO2 nanotubes. The rectangular inner
shapes of these tubes are remarkably uniform. Fig. 2
shows the SiO2 nanotube coated ammonium tartrate
crystal, and shows the end of one of these fibers and
the X-ray spectra collected from the shaft and the tip
of fiber. The end appears to have fractured off expos-
ing the interior. X-ray spectra show that the tip is
composed primarily of carbon and the shaft is silicon
and oxygen.
In this article, we confirmed the structure of the
ammonium tartrate crystal and the SiO2 nanotube
coated ammonium tartrate crystal (ATT) by 13C CP/
MAS NMR. The quantitative distribution of the
various groups in the SiO2 nanotube were also
obtained from the 29Si CP/MAS NMR spectra. In
addition, the 13C spin–lattice relaxation times in the
rotating frame, T1r , were measured as a function of
temperature for two samples. From these results, the
activation energy for each carbon in the ATT and SiO2
nanotube coated ATT were obtained.
2.2. Solid state NMR spectroscopy
The solid state NMR experiments were performed
using Brucker 200 NMR and Varian 300 NMR spec-
trometer. Cross-polarization, magic-angle spinning
(CP/MAS) 29Si and 13C NMR experiments were
performed at the Larmor frequencies of 39.73 and
75.46 MHz, respectively. The samples were measured
in powder form using solid state NMR spectroscopy.
The samples were placed in a CP/MAS probe of 7 mm
and the magic-angle spinning rate was 5 kHz to mini-
mize spinning sideband overlap. In order to measure
2. Experimental
2.1. Nanotube synthesis
Tetraethyl orthosilicate (TEOS) and dl-tartaric
acid (TTA) were obtained from Aldrich Chemical.

A.R. Lim et al. / Solid State Communications 110 (1999) 333–338
335
Fig. 2. Scanning electron microscopy picture of SiO2 nanotube coated ATT crystal and X-ray spectra from shaft (left) and (tip).
the 13C spin–lattice relaxation time, the p/2 pulse
time was 5 ms, corresponding to a spin-locking field
strength of 50 kHz. 13C T1r was determined by vary-
ing a 13C spin-locking pulse after a 0.8 ms CP prepara-
tion period. The decay of the 13C magnetization in the
spin-locking field was recorded for spin-locking times
of up to 30 ms.
cycles, Q3 the chain branching sites and Q4 the three-
dimensionally cross-linked groups [16,17]. The solid
state 29Si NMR spectrum is shown in Fig. 5. The total
range of 29Si chemical shifts is usually observed in the
29Si NMR spectra in the range of about Ϫ60 to about
Ϫ160 ppm, i.e. about 100 ppm [18]. Within this
range, two dominant resonances are observed at the
chemical shifts Ϫ111 and Ϫ102 ppm. Based on a
model compound, the resonances are assigned to the
Q3 and Q4 units, respectively. The Q2 unit at chemical
shift Ϫ92 ppm has a relatively small intensity. The
relative concentrations of the Qn structural units can
be obtained directly from the integrated peak areas.
The resultant percentages of Q2, Q3, and Q4 branching
points (i.e. cross-links) are 4.0, 40.3, and 55.7%,
respectively.
3. Results and discussion
The solid state structures and dynamics of the ATT
and SiO2 nanotube coated ATT were examined using
solid state NMR. Fig. 4 shows the solid state 13C CP/
MAS NMR spectrum for the ATT. The spectrum
consisted of four signals at chemical shifts of d ˆ
180.20, 178.58, 76.71 and 74.45 ppm at room
temperature. The splitting of the 180.20 and
178.58 ppm (76.71 and 74.45 ppm) can be explained
by the effect of isomerization of the samples. Peaks of
1 and 2 are assigned with the carbon and hydrocarbon,
respectively, as shown in Fig. 4. Spinning sidebands
are marked with asterisks. The 13C CP/MAS NMR
spectrum in the case of the SiO2 nanotube coated
ATT is very similar to the ATT.
The 13C spin–lattice relaxation time in the rotating
frame, T1r , for the ATT and SiO2 nanotube coated
ATT was taken in several temperature ranges with
variable spin-lock on the carbon channel following
cross-polarization. Normal cross-polarization uses a
time delay during which the proton rf field is turned
The structure of silicate anions commonly uses a Qn
notation. In this notation, Q represents a silicon atom
bonded to four oxygen atoms forming a tetrahedron.
The superscript n indicates the connectivity, i. e. the
number of oxygen atoms attached to the Si. Thus, Q0
denotes the monomeric orthosilicate anion Si, Q1 the
end-group chains, Q2 the middle groups in chains or
Fig. 3. Reaction scheme for nanotube synthesis.

336
A.R. Lim et al. / Solid State Communications 110 (1999) 333–338
Fig. 5. Solid state 29Si CP/MAS NMR spectrum of ATT at room
temperature.
Fig. 4. Solid state 13C CP/MAS NMR spectrum of ATT at room
temperature.
motion conditions vtc q v1tc q 1) [21,22]. On the
slow side of the T1r minimum a decrease in T1r
implies a smaller value of tc, and therefore, an
increased mobility at higher temperatures.
off. This delay is introduced after the carbons are
pulsed into the rotating frame. The time interval
allows the carbon magnetization to decay by dipolar
coupling [19]. The T1r relaxation process follows this
exponential function [20]:
For the spin–lattice relaxation time in the rotating
frame, the experimental value of T1r can be expressed
in terms of a correlation time tC for the molecular
motions by the BPP (Bloembergen–Purcell–Pound)
function [23].
Further insight into the relative molecular motions
can be gained by examining the activation energies
associated with the various carbons. The theoretical
temperature dependence of the rate of rotational
motions is a simple Arrhenius expression [24]
Mꢀt† ˆ M0expꢀϪt=T1r†;
ꢀ1†
where M0 is the initial height of the free induction
decay when the radiofrequency field is switched off.
Thus, monitoring of the decay of magnetization inten-
sities, M(t), permits an analysis of the T1r spin–lattice
relaxation behaviour and determination of the T1r
relaxation times.
tC ˆ t0expꢀEa=RT†
ꢀ2†
where Ea is activation energy for the molecular
motions and R is the molar gas constant. Thus, a
plot of the natural logarithm of the relaxation time
as a function of the inverse temperature is linear
with a slope that is proportional to the activation
energy for motion. Fig. 7(a) and (b) shows represen-
tative plots of the correlation time as a function of the
inverse temperature for carbon and hydrocarbon in the
ATT and SiO2 nanotube coated ATT. The activation
energies for all the carbons are derived from the slope
of the straight line passed through the tC data. The
activation energies for all carbons of the SiO2 nano-
tube coated ATT are longer than those measured for
the ATT. In carbons, the activation energy for ATT
and SiO2 nanotube coated ATT is 2.87 and 5.91 kJ
molϪ1, respectively, while hydrocarbons show activa-
tion energy in ATT and SiO2 nanotube coated ATT of
Fig. 6(a) and (b) shows T1r for each carbon in both
the ATT and SiO2 nanotube coated ATT as a function
of temperature. The 13C spin–lattice relaxation time
in the rotating frame decreases with increasing
temperature. The T1r corresponding to two carbons
in the SiO2 nanotube coated ATT shows long relaxa-
tion times relative to the ATT. The relaxation times of
carbon similarly decrease at increasing temperatures
for both the ATT and SiO2 nanotube coated ATT.
However, the greatest difference between the ATT
and SiO2 nanotube coated ATT samples appears in
the alcohol side chains: the relaxation times for the
hydrocarbons here differ substantially, with the SiO2
nanotube coated ATT showing the larger value.
The relaxations of all carbons undergo motions on
the slow side of the T1r minimum (i.e., under slow

A.R. Lim et al. / Solid State Communications 110 (1999) 333–338
337
Fig. 6. (a) Temperature dependence of the 13C spin–lattice relaxa-
tion time, T1r , for carbon in the ATT and SiO2 nanotube coated
ATT. (b) Temperature dependence of the 13C spin–lattice relaxation
time, T1r , for hydrocarbon in the ATT and SiO2 nanotube coated
ATT.
Fig. 7. Arrhenius plot of the natural logarithm of the correlation
times as a function of the inverse temperature. (a) carbon, and (b)
hydrocarbon in the ATT and SiO2 nanotube coated ATT.
4. Conclusions
2.42 and 11.68 kJ molϪ1, respectively. Hence, the
molecular motions of the ATT are restricted by
interpenetration of the SiO2 nanotube. It is clear that
strong interactions, be they covalent C–O–Si linkages
or hydrogen bonding, do exist. This is best evidenced
by comparing the activation energies of the hydrocar-
bon immediately adjacent to the hydroxyl group in the
ATT and SiO2 nanotube coated ATT. The activation
energy for hydrocarbon in the SiO2 nanotube coated
ATT is nearly five times larger than in ATT alone,
which indicates that substantially restricted motion
occurs.
Activation energy is a quantitative measure of
rigidity, and the activation energies for each carbon
were obtained from the correlation times as a function
of the temperature. Molecular motions in the SiO2
nanotube coated ATT have higher activation energies
than in ATT alone. Higher activation energies indi-
cate a higher rigidity for the polymer chains in the
SiO2 nanotube coated ATT sample. It is worth noting
that the activation energy of the hydrocarbon in SiO2
nanotube coated ATT is distinctly greater than that in
ATT. This higher activation energy is associated with

338
A.R. Lim et al. / Solid State Communications 110 (1999) 333–338
[7] E.J.M. Hamilton, S.E. Dolan, C.M. Mann, H.O. Mc Donald,
S.J. Shore, Science 260 (1993) 659.
[8] P. Gleize, S. Herreyre, P. Gadelle, M. Caillet, J. Mater. Sci. 29
(1994) 1571.
strong bonding between hydroxyl group in ATT and
the oxygen atoms attached to the Si for Q3 or Q4 type
in SiO2 nanotube.
[9] P. Gleize, S. Herreyre, P. Gadelle, M. Mermoux, M.C. Chey-
net, L.J. Abello, J. Mater. Sci. Lett. 13 (1994) 1413.
[10] T. Nemetschek, U.Z. Hofmann, Naturforsch 8b (1953) 410.
[11] T. Nemetschek, U.Z. Hofmann, Naturforsch 9b (1954) 166.
[12] H. Nakamura, Y. Matsui, J. Am. Chem. Soc. 117 (1995) 2651.
[13] G.T. Schueneman, B.M. Novak, A.I. Lesser., Chem. Materi-
als, 1998. (To be submitted).
[14] W. Stober, A. Fink, E. Bohn, J. Colloid Interface Sci. 26
(1968) 62.
[15] V.S. Yadava, V.M. Padmanabhan, Acta Cryst. B 29 (1973)
493.
[16] G.E. Maciel, D.W. Sindorf, Macromolecules 102 (1980) 7606.
[17] D.W. Sindorf, G.E. Maciel, J. Phys. Chem. 86 (1982) 5208.
[18] G. Engelhardt, D. Michel, High-Resolution Solid State NMR
of Silicates and Zeolites, Wiley, New York, 1987.
[19] L. Laupretre, L. Monnerie, J. Virlet, Macromolecules 17
(1996) 1397.
[20] A.R. Lim, K.S. Hong, J.H. Chang, Solid State Commun. 100
(1996) 731.
[21] D. Shaw, Fourier Transform NMR Spectroscopy, Elsevier
Scientific, New York, 1976.
Acknowledgements
One of the authors (A.R.L.) was supported by the
Korea Science and Engineering Foundation through
the RCDAMP at Pusan National University (1997–
2000). B.M.N. acknowledges the Materials Research
Science and Engineering Center at the University of
Massachusetts, the Office of Naval Research, and the
FAA for financial support.
References
[1] R.J. Bacon, J. Appl. Phys. 31 (1960) 283.
[2] R.T.K. Baker, M.A. Barber, P.S. Harris, F.S. Feates, R.J.
Waite, J. Catal. 26 (1972) 51.
[3] R.T.K. Baker, Cabon 27 (1989) 315.
[4] S. Motojima, M. Kawaguchi, K. Nozaki, H. Iwanaga, Carbon
29 (1991) 379.
[5] S. Iijima, Nature 354 (1991) 56.
[6] S. Amelinckx, X.B. Zhang, D. Beraerts, X.F. Zhang, V.
Ivanov, J.B. Nagy, Science 265 (1994) 635.
[22] P. Laszzlo, NMR of Newly Accessible Nuclei, Academic
Press, New York, 1993.
[23] A.R. Lim, B.M. Novak, Solid State Commun. 109 (1999) 29.
[24] G.P. Jones, Phys. Rev. 148 (1966) 332.