Research Study about NanoDiamonds

Nanodiamond Promotes Surfactant-Mediated Triglyceride Removal
from a Hydrophobic Surface at or below Room Temperature
Xianjin Cui,† Xianping Liu,‡ Andrew S. Tatton,§ Steven P. Brown,§ Haitao Ye,⊥ and Andrew Marsh*,†
⊥School of Engineering and Applied Science, Aston University, Birmingham, B4 7ET United Kingdom
†
Department of Chemistry, ‡

School of Engineering, and §

Department of Physics, University of Warwick, Coventry, CV4 7AL United

Kingdom
*S Supporting Information
ABSTRACT: We demonstrate that ca. 5 nm nanodiamond
particles dramatically improve triglyceride lipid removal from a
hydrophobic surface at room temperature using either anionic
or nonionic surfactants. We prepare nanodiamond−surfactant
colloids, measure their stability by dynamic light scattering and
use quartz crystal microbalance−dissipation, a technique
sensitive to surface mass, in order to compare their ability to
remove surface−bound model triglyceride lipid with ionic and
nonionic aqueous surfactants at 15−25 °C. Oxidized, reduced,

ω-alkylcarboxylic acid, and ω-alkylamidoamine surface-modi-
fied adducts are prepared, and then characterized by

techniques including 13C cross-polarization (CP) magic-angle spinning (MAS) NMR. Clear improvement in removal of
triglyceride was observed in the presence of nanodiamond, even at 15 °C, both with nanodiamond−surfactant colloids, and by
prior nanoparticle deposition on interfacial lipid, showing that nanodiamonds are playing a crucial role in the enhancement of the
detergency process, providing unique leads in the development of new approaches to low-temperature cleaning.
KEYWORDS: nanodiamond, tristearin, lipid removal, detergency, surfactant, zeta potential, quartz crystal microbalance (QCM),
QCM with dissipation monitoring (QCM-D)
■ INTRODUCTION
The removal of complex crystallized lipid in laundry soil,
(“sebum”: usually modeled by a triglyceride such as tristearin),
is time- and energy-intensive requiring high temperature, even
with modern detergents.1−6 The UK consumer performs an
average of five washes at 90 °C and 83 at 60 °C each year, at up
to 2.7 and 1.6 kWh per wash, respectively, in order to remove
this soil.7 The major cost is in heating water, hence consumer
and industry desire for low-temperature cleaning where energy
consumption might be reduced to 0.1−0.2 kWh for 20 °C wash
cycles. Methods typically used for cleaning include mechanical
abrasion, enzymes8 to break down polymeric protein soils and
hydrolyze triglycerides,5,9 together with surfactants at higher
temperature.10−13 Although enzymes are key ingredients in
modern detergents, those currently available are ineffective at
hydrolyzing crystallized lipid at room temperature, requiring
both effective adsorption and high mobility on the lipid surface
to facilitate the key catalytic activity at the water−lipid
interface.9,14 Glucose-derived nonionic surfactants including
decyl β-D-glucopyranoside,15−18 or surfactant mixtures19 have
been found to be more efficient than many amphiphiles at
removing tristearin from model surfaces.15,16 Optimum
detergency of nonionic surfactants, including these, occurs
above the phase inversion temperature, which can be reduced
by adding a small amount of lipophilic amphiphile,1,20,21 or by
making the surfactants more hydrophobic. This approach has

also been explored to enhance cleaning at lower temper-
atures,19,22 although the effects of nanoparticulates have not

been studied. We speculated that detonation diamond

nanoparticles (‘nanodiamond’),23,24 known to possess hydro-
phobic features, may aid the removal of lipid from surfaces at

room temperature, plausibly by aiding the detergency of
surfactants, in addition to altering the lipid surface upon
deposition, and other mechanisms. Nanodiamonds possess a
surprisingly large surface area that can adsorb lipid and/or
other small molecules,25,26 and offer opportunities for further

functionalization27 in order to tune surface properties. Nano-
particles’ (or carbon nanotubes’)

28 tendency to self-aggregate in
air and aqueous solution, in order to minimize surface energy
brings a barrier to their use, usually mitigated by surface
modification29−31 or surfactant additives32−34 that stabilize
colloidal nanoparticle suspensions. In this work, we initially
explored a range of anionic (sodium dodecylbenzenesulfonate,

SDBS), nonionic (decyl β-D-glucopyranoside, G1C10; decyl β-D-
maltopyranoside, G2C10; polyoxyethylene(12) nonylphenyl

ether, NFE10; and polyethyleneglycol dodecyl ether, Brij35)
and zwitterionic (N,N-dimethyldodecylamine N-oxide, DDAO)
surfactants with untreated nanodiamonds. We then also
Received: March 30, 2012
Accepted: June 6, 2012

Research Article
www.acsami.org

modified the nanoparticles’ surface chemistry to further
understand mechanisms for the effects we observed, using 13C cross-polarization (CP) magic-angle spinning (MAS)
nuclear magnetic resonance NMR to confirm structure.
Dynamic light scattering (DLS) was applied to characterize
the stability of nanodiamond colloids or suspensions, and
quartz crystal microbalance with dissipation monitoring
(QCM-D)35−38 used in order to study the lipid removal
process.11,15
■ EXPERIMENTAL SECTION
Tristearin lipid and cetyltrimethylammonium bromide (CTAB) were
purchased from Tokyo Chemical Industry (TCI) Ltd. Diamond
nanopowder 97% (ND97) and all other reagents were purchased from
Sigma-Aldrich and used without further purification. Zeta potential of
particles in solution were measured by dynamic light scattering (DLS)
on a Malvern Zetasizer at 25 °C, in Milli-Q water with a neutral pH
value from 7.0 to 7.2. QCM data were recorded on a Q-Sense E4 from
Q-Sense AB, Sweden at a 25 °C and processed using Q-Tools
software.9,37,39 Raman spectra were recorded on a Renishaw inVia
Raman Microscope with a CCD detector and a laser of 514.5 nm.
Infrared (IR) Spectra were recorded on a PerkinElmer Spectrum 100
FT-IR spectrometer, using attenuated total reflectance. Scanning
Electron Microscopy (SEM) images were obtained on JEOL 4500
FIB/SEM, and thermogravimetric analysis (TGA) was carried out on a
Mettler Toledo DSC1−400 under N2 with a heating rate of 10 °C/
minute. 13C CP MAS experiments were recorded using a Bruker
Avance II+ spectrometer operating at a 1

H Larmor frequency of 599.4
MHz (B0 = 14.1 T). Experiments described in panels a and c in Figure
6 were recorded using a Bruker 3.2 mm HX probe, whereas the
experiment in (b) was recorded using a Bruker 2.5 mm triple
resonance probe, operating in double resonance mode. An MAS
frequency of 20 kHz was used for all spectra presented. A 1
H pulse of
duration 2.5 μs was used to excite initial transverse magnetization. The
number of coadded transients recorded for each spectrum was (a)
14436, (b) 10980 and (c) 2000, with recycle delays of (a) 3 s, (b) 2 s,
and (c) 3 s. 13C magnetization was created using a cross-polarization
ramp40 of magnitude 80% to 100%, with a contact time of duration (a)
3 ms, (b) 2 ms, and (c) 3 ms. SPINAL-6441 heteronuclear decoupling
was applied during acquisition at a 1

H nutation frequency of 100 kHz,
for (a) 10 ms, (b) 12 ms, and (c) 10 ms. All resonances are referenced
to the carboxylic acid group in L-alanine at 177.8 ppm, which
corresponds to a primary reference to tetramethylsilane (TMS) at 0
ppm.
Preparation of 1. Commercial ND97 was heated in air in a tube
furnace at over 400 °C for 1−5 h, with a heating rate of 10 °C/minute.
Preparation of 2. Commercial ND97 was treated with hydrogen
plasma in a custom-made molybdenum container within an ASTeX

5010 (Seki Technotron Corp., 2.45 GHz, 1.5 kW) MPCVD system.
Hydrogen plasma was achieved by heating the 200 S.C.C.M (standard
cubic centimeter per minute) diluted hydrogen flow under pressure of
50 Torr. The ND97 powder was kept in the resultant plasma for 10
min and then cooled in H2 atmosphere.
Preparation of ω-Alkylcarboxyl-Functionalized Diamond
Nanoparticles 3,. Route A. A suspension of 2 (100 mg) in toluene
(50 mL) was sonicated for 1 h; 10-undecenoic acid (3 mL) was then
added and the resultant mixture sonicated for a further hour. The
reaction was then irradiated with UV light (254 nm) for 6 h with
vigorous stirring. The product 3 was collected by centrifugation as a
powder, washed with ethanol 5 times, and dried at 60 °C under
vacuum.
Route B. A suspension of 2 (100 mg) in triethylamine (50 mL) was
sonicated for 1 h then 10-undecenoic acid (3 mL) was added
dropwise. The resultant mixture was sonicated for a further hour, then
irradiated with UV light (254 nm) for 6 h with vigorous stirring. The
product 3 was obtained as a solid by centrifugation and washed with
ethanol, 3 M HCl and water.
Preparation of ω-Alkylamine-Functionalized Diamond
Nanoparticles 4. This was carried out in toluene using the same
method as for 3, with 1-(4-methylpiperazin-1-yl)undec-10-en-1-one in
place of 10-undecenoic acid.

Preparation of QCM Sensor and Lipid Removal Experi-
ments. All QCM sensors used in this paper are gold coated quartz

crystals from Q-Sense, with a fundamental frequency of 5 MHz. The
gold QCM sensor exhibited a water contact angle of 62° and in order
to obtain a hydrophobic surface suitable for tristearin deposition,
sensors were coated with a self-assembled octadecylmercaptan
(CH3(CH2)17SH) monolayer as follows. Crystals were immersed in
CHCl3 for 12 h to remove organic contaminants then placed under
UV light (254 nm) for 20 min, and immersed in a mixture of
H2O:NH3·H2O:H2O2 (5:1:1) at 75 °C for 30 min (CAUTION!
Reacts violently with organic material. Prepare, handle and dispose of
small quantities only). After rinsing with Milli-Q water the sensors

were immersed in a solution of CH3(CH2)17SH (1 mmol/L in n-
hexane) for 20 h, rinsed with n-hexane, and finally dried under a gentle

flow of N2 gas.
QCM-D Experiments. The lipid film was obtained by dropping
and slowly evaporating tristearin solution (10 μL 1 g/L) in chloroform
and hexane (1:9) on the thiol-coated QCM-D sensors, causing a
frequency decrease of around 1500 Hz. The lipid-coated sensors were
then exposed to pure water and QCM-D recording begun when a flat
and stable baseline was obtained. After a few minutes the solution was
changed to the nanodiamond suspension, depositing nanoparticles on
the lipid layer. The buffer was then switched to surfactant solution
when the adsorption of nanoparticles reached saturation. Finally, the

surfactant solution was changed back to pure water to clean the QCM-
D system. Key experiments were carried out in at least duplicate with

Table 1. Lipid Removal by Anionic Surfactant SDBS with and without Prior Deposition of Nanoparticles, T = 25 °Ca
entry nano- diamond [SDBS] (mmol L−1

) zeta potential (mV) t (min) R1 (Hz) R2 (Hz)
1 none present 1 n.d. n.d. −30 No removal
2 ND97 1 ca. 15c 5 −300 10
3 none present 20 n.d. 7 −30 19
4 ND97 20 ca. 15c 20 −220 60
5 none present 40 n.d. 2 350 ± 90 520 ± 30
6 ND97 40 ca. 15c 120 650 ± 110 860 ± 140
7 3 40 26.1 110b 580 ± 125 730 ± 80b
8 2 40 41.5 170 540 ± 40 670 ± 70
9 4 40 48.1 600b 370 590b
10 2 and 1 40 <41.5 65 300 570
11 4 and 1 40 <48.1 204 660 ± 10 940 ± 160
a
For meaning of R1 and R2, please refer to main text. Compound key: 1 = oxidized nanodiamond, 2 = hydrogenated nanodiamond, 3 = ω-tert-amine
functionalized nanodiamond, 4 = ω-carboxylic acid functionalized nanodiamond. Error bars represent standard deviation of duplicate measurements.
t is the time taken to reach equilibrium. n.d. = value not determined. b

The system did not reach equilibrium. c

Approximate value because the colloid

was unstable over time.
ACS Applied Materials & Interfaces Research Article

B dx.doi.org/10.1021/am300560z | ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

mean and standard deviations displayed as error bars in Table 1 and
Figures 3 and 4.
■ RESULTS AND DISCUSSION

Nanodiamond and Colloid Characterization. Trans-
mission electron microscopy imaging of commercial nano-
diamond powder of 97% purity (ND97) revealed aggregates

several hundred nanometres across with primary particle size
5−6 nm (see the Supporting Information, Figure S1),
consistent with the mean size calculated from X-ray powder
diffraction using the Scherrer eq (4 nm, see the Supporting
Information, Figure S2). The primary particles were seen to be
encapsulated in amorphous carbon, lighter in contrast than
crystallized regions (see the Supporting Information, Figure
S4), with a broad peak at 1600 cm−1 in the Raman spectrum
((see the Supporting Information, Figure S2), confirming the
presence of what is generally referred to as “sp2 carbon”.
30
The zeta potential of ND97 was measured in water (pH 7.0
to 7.2) to be ca. 15 mV, consistent with observed poor colloidal
stability. A system with zeta potential outside +30 to −30 mV is
usually considered stable,42 and ionic surfactants such as SDBS,
SDS and CTAB were found to impart high magnitude zeta
potential, even at low concentration (see the Supporting
Information, Figures S7 and S8). Nonetheless, zeta potential
alone does not estimate colloid stability well, especially in
solutions of high ionic strength, since many factors
contribute.42 Hence, we also measured average particle size
over time to confirm the stability of nanodiamond suspensions
(see the Supporting Information, Figure S9), implying that

commercial ND97 is most stable in 2 mmol/L hexadecyl-
trimethylammonium bromide (CTAB) solution, then 10

mmol/L sodium dodecylbenzenesulfonate (SDBS) solution.
Obvious aggregation was seen for ND97 in 40 mmol/L SDBS
solution, since the average size of particles kept increasing over
a 14-h period, despite a zeta potential of −50 mV.
The adsorption of nonionic or zwitterionic surfactants onto
particles will reduce the charge perceived at the shear plane,42
but the nanoparticles remain stabilized due to the surfactants’
kosmotropicity.43,44 For example, ND97 was found to be very
stable in 8 g/L polyethylene glycol (Mn = 1000, PEG-1000)
aqueous solution, and reasonably stable in 8 g/L polyethylene
glycol dodecyl ether (Brij35) solution ((see the Supporting
Information, Figure S10), principally due to the energy
required to dehydrate the hydrophilic PEG as any two
nanodiamond-surfactant assemblies approach. Interestingly,
low stability was seen in a solution of zwitterionic surfactant
N,N-dimethyldodecylamine N-oxide (DDAO) above its critical
micellar concentration, possibly due to the relatively small,

albeit potent hydrophilic moiety in DDAO compared to PEG-
1000, or that amphiphile’s preference to exist as small micellar

assemblies.
In the absence of surfactant, nanodiamond particles did not
remove tristearin from surfaces, rather they were adsorbed45 on
a lipid-coated QCM sensor exposed to a suspension of ND97
in water (Figure 1, region (ii)). However, when this
nanodiamond-coated surface was exposed to SDBS, the
removal of tristearin was significantly enhanced relative to
SDBS alone. To quantify this effect we report two frequency
changes (a directly observed quantity, rather than mass change,
a derived quantity) for each surface treatment, where R1
represents the baseline−to−equilibrium desorption of surface
bound materials including lipid and R2 the maximum
deflection11 (as previously reported), presenting an average of

two side-by-side experiments with standard deviation (Table 1).
Data for individual runs are shown in Figure 2 indicating in this
case that ND97-SDBS colloid gives a 23% improvement over
SDBS solution (40 mM) alone.

The nonionic surfactants decyl β-D-glucopyranoside (G1C10)
and decyl β-D-maltopyranoside (G2C10) have been reported11,17
as effective at removing tristearin at 50 °C,16 but we observed
them to be less so at room temperature. Surfactant G1C10 was
still found to remove lipid (Figure 2, R1 = 440 Hz), increasing
to R1 = 690 Hz in the presence of G1C10/ND97 colloid, an
enhancement of 57%. In contrast, G2C10 was ineffective at
room temperature (Figure 3 and the Supporting Information).
If, as has been done previously,46,47 we assume the Sauerbrey
equation applies here, the removal of tristearin by G1C10 in the
Figure 1. Frequency and dissipation changes to third overtone
measured by QCM-D after exposure of the tristearin coated sensor to:
region (i) 0−7 min, water; region (ii) 7−52 min, ND97 suspension;
region (iii) 52−175 min, SDBS solution. [ND97] = 0.1 g/L, [SDBS] =
40 mmol/L, T = 25 °C; R1 = 654 Hz, R2 = 1054 Hz.

Figure 2. Frequency change to third overtone measured by QCM-D
when the tristearin coated sensor is exposed to: region (i) 0−7 min,
water is used for both sensors to obtain a stable baseline; region (ii)
7−17 min, (a) G1C10/ND97 colloid R1 = 690 Hz, R2 = 740 Hz, (b)
SDBS/ND97 colloid, R1 = 540, R2 = 730 Hz, (c) SDBS solution, R1 =
440, R2 = 490 Hz, and (d) G1C10 solution R1 = 440 Hz, R2 = 440 Hz
were used to remove tristearin lipid; region (iii) 17−26 min, SDBS
solution was used for sensors b and c, whereas G1C10 solution was
used for sensors a and d to clean the QCM system. [SDBS] = 40
mmol/L, [G1C10] = 3.1 mmol/L, [ND97] = 0.1 g/L, T = 25 °C.
ACS Applied Materials & Interfaces Research Article

C dx.doi.org/10.1021/am300560z | ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

presence of ND97 can be estimated to be 43%, since
evaporating a solution of tristearin (10 μL, 1 g/L) in a mixture
of CHCl3 and n-C6H14 (1: 9) on the sensor resulted in around
a 1500 Hz decrease in frequency. The observed changes in
dissipation also confirmed the removal of lipid from the surface
(see the Supporting Information). The improvements in
detergency induced by ND97 particles were also seen in the
presence of nonionic surfactants Brij35, and for zwitterionic
surfactant DDAO, albeit to a much lesser extent (Figure 3).
The efficiency of nonionic surfactant NFE10 at tristearin
removal was not improved in the presence of ND97, although
the detergency process proceeded more rapidly (see the
Supporting Information, Figures S21−23).
About the same amount of lipid removal was observed for
ND97/SDBS, or ND97/G1C10 colloids compared with
exposure to ND97 suspension in water followed by surfactant
solution (SDBS or G1C10). Hence interfacial nanodiamond is
confirmed as the key player in lipid removal.
This nanodiamond effect is even more significant at lower
temperature (Figure 4). As expected, very limited lipid removal
(12 Hz) was observed for 40 mmol/L SDBS solution at 15 °C.
Dramatic improvement to a value of 161 Hz was seen at the
same temperature in the presence of ND97. Nonionic
surfactant G1C10 was ineffective at lipid removal when the
temperature was decreased to 15 °C, even in the presence of
ND97 (see Figure 4 and the Supporting Information).

Nanoparticle Enhanced Cleaning Mechanisms. Nano-
diamond has a known ability to adsorb small molecules,25,26,32

although that alone cannot explain our observations, because
nanodiamond by itself is ineffective at removing tristearin from
a surface. The beneficial effect of adding a hydrophobic surface
to nonionic surfactants1,20 may be significant in bringing about
the observed effects and this could be further tested by
measuring the phase inversion temperature of surfactant−lipid
mixtures in the presence and absence of nanodiamond or other

nanoparticulates. Roughening of an otherwise relatively hydro-
phobic surface by deposition of the nanoscale objects,

effectively lowering the interfacial energy and ability of
surfactant to penetrate the lipid−water interface is a third
way in which cleaning might be assisted. Finally, the acidic sites

or other impurities on the nanodiamond surface could serve to
catalyze tristearin hydrolysis, introducing fatty acids that are
know to dramatically enhance solubilization rates of triacyl
glycerol esters.48 Any of these mechanisms might be susceptible
to alteration in nanodiamond surface chemistry or electrostatic
charge. To further investigate these effects and how they might
modulate classic mechanisms for removal of lipid from
surfaces,6 we prepared functionalized nanodiamonds by
oxidative, reductive, and subsequent photochemical routes
(Scheme 1).

Modification of Nanodiamond. Thermal oxidation is an
efficient way to remove some of the ‘sp2

’ surface carbon, to give
polar or ionizable functional groups including carboxylic
acids.30 After heating the nanodiamond powder in air at
400−500 °C to give oxidized product (1), Raman spectroscopy
and transmission electron microscopy (TEM) (see the

Figure 3. Summary of lipid removal (R1) obtained by surfactants in
the presence or absence of ND97 at 25 °C. SDBS/ND97 represents
the colloid of ND97 stabilized by SDBS. ND97+SDBS represents
exposure to ND97 suspension in water, followed by SDBS solution.
[SDBS] = [DDAO] = 40 mmol/L, [G1C10] = 3.1 mmol/L, [Brij35] =
8 g/L, [NFE10] = 1.5 wt %, [ND97] = 0.1 g/L, [G2C10] = 2.1 mmol/L.
Error bars represent standard deviation of duplicate measurements.

Figure 4. Summary of lipid removal (R1) obtained by surfactants
SDBS or G1C10 in the presence of ND97 at 25 and 15 °C. [SDBS] =
40 mmol/L, [G1C10] = 3.1 mmol/L, [ND97] = 0.1 g/L. Error bars
represent the standard deviation of duplicate measurements.

Scheme 1. Preparation of Derivatized Diamond
Nanoparticles 1−4 from Commercial Nanodiamond Powder
(97% purity, Nd97). Reagents and conditions: (i) 485 °C,
air, 1-5 h; (ii) H2 plasma, 10 min; (iii) 10-undecenoic acid,
PhMe (or Et3N), hυ, λ = 254 nm, or 1-(4-methylpiperazin-1-
yl)undec-10-en-1-one, PhMe, hυ, λ = 254 nm
ACS Applied Materials & Interfaces Research Article

D dx.doi.org/10.1021/am300560z | ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Supporting Information, Figures S3 and S4) clearly revealed
that so-called ‘sp2

’ or amorphous carbon had been selectively
removed, with the intensity of the peak at 1330 cm−1 relative to
that at 1600 cm−1 increasing with time of treatment. Infrared
spectra revealed a new peak at 1800 cm−1 (see the Supporting
Information, Figure S11), some 40 cm−1 higher than that of
carboxyl groups reported previously.29,30
To more fully understand this peak, oxidized nanodiamond
sample was immersed in aqueous 3 M KOH, whereupon IR
spectroscopy of the dried sample showed that the peak had
moved to longer wavelength, 1775 cm−1

. Upon treatment with
aqueous 3 M HCl the absorption returned to 1800 cm−1
,
indicative of reprotonation of a carboxylate, rather than
hydrolysis of a putative anhydride. The reason for the higher
wavenumber observed herein for the carboxyl absorbance
remains uncertain, but the vibration frequency of surface
carbonyls on nanodiamond has been reported to be very
sensitive to their local environment, as well as temperature.49
Reduction of powder ND97 with hydrogen plasma at an

optimized 900 W was used to prepare H-derivatized nano-
diamond (2). Infrared spectroscopy (see the Supporting

Information, Figure S5) confirmed new hydrocarbon bonds
on the surface of the particles with peaks at 1394, 1331, and
840 cm−1 characterized as alkyl C−H bending with those at
2877 and 2940 cm−1 attributed to alkyl C−H stretching. TEM
revealed regions of high crystallinity as expected (Figure 5),

together with areas of amorphous carbon. This reduced
material was used as a versatile platform from which to prepare

nanodiamond bearing alkyl ω-carboxyl and ω-amine function-
ality (3 and 4, respectively; Scheme 1), through a

straightforward photoinitiated reaction.50−52 Characteristic
alkyl chain C−H stretching vibrations at 2852 and 2933 cm−1
and other diagnostic peaks such as a new amide carbonyl CO
stretch at 1725 cm−1 and N−H bending at 1566 cm−1
confirmed the expected functionality. Solid-state CP MAS
NMR spectra (Figure 6) demonstrated disappearance of the
alkene signals and other features consistent with the expected
photoinitiated reaction. Alkyl carbons resonating at 45, 33, and
30 ppm,53−61 are seen, but no peaks for alkene carbons are
observed between 110 and 150 ppm. The signal of carbonyl
group at 177 ppm is clearly seen for amido-amine 4 but is
weaker in the case of ω-acid 3, possibly because of difference in
dynamics.
These features alone cannot be considered unambiguous
proof of covalent attachment, but thermogravimetric analysis

(see the Supporting Information, Figure S13) additionally
showed that there was a clear weight loss at temperatures of up
to 250 and 300 °C for sample 4 and sample 3 respectively,
confirming that the alkyl chains are in fact covalently attached
to the particles. By comparison, only a slight decrease in mass
for the starting material 2 was evident below 100 °C, consistent
with the loss of surface associated solvent or water, with no
other signal visible up to 700 °C.
The zeta potential of commercial nanodiamond ND97 (15
mV) increased to 41.5 mV after treatment with H2 plasma.
Reduction in zeta potential from 41.5 to 26.1 mV was observed
after the photoinitiated reaction with ω-alkenylcarboxylate
-(CH2)10-COOH, whereas a slight increase to a zeta potential

of 48.1 mV was seen after reaction with 1-(4-methylpiperazin-
1-yl)undec-10-en-1-one. In contrast, the oxidized nanodiamond

samples 1 have a negative zeta potential, of up to −46.5 mV,
offering further evidence of carboxylate surface functionality in
that material. The zeta potential of oxidized samples also
correlates positively with the intensity of carbonyl functional
group infrared stretch (Figure 7).
Average aggregate size as a function of time indicated the
stability of nanodiamond suspensions (see the Supporting
Information, Figure S12), showing that the order is: 1 > 4 >
ND97 ≈ 3 > 2 and that neither 2 or 3 are very stable in water,
exhibiting aggregation and sedimentation. The oxidized powder
1 was found to be extremely stable with a small aggregate size
in water, plausibly because of their negatively charged and
hydrophilic surface.
Results for lipid removal by SDBS in the presence of
nanodiamonds exhibiting different zeta potentials clearly
showed a trend that time to reach equilibrium t increases
with the zeta potential of particles (Table 1). For example, in
the presence of ND97 with a zeta potential of 15 mV, it took
around 120 min to achieve maximum removal, compared to
more than 600 min in the presence of amine functionalized
nanoparticles 4, with a zeta potential of 48.1 mV (Figure 8,
left). Thus the rate of lipid removal with anionic surfactant
SDBS in the presence of nanoparticles follows the order: ND97
> 3 > 2 > 4, proportional to the strength of electrostatic

Figure 5. TEM image of 2 obtained from ND97 by hydrogen plasma
treatment. Circled area indicates amorphous carbon.

Figure 6. 13C NMR CP MAS spectra of diamond nanoparticles
samples (1
H Larmor frequency 600 MHz, 20 kHz MAS): (a)
alkylamine functionalized diamond nanoparticles 4; (b) alkylcarboxyl
functionalized diamond nanoparticles 3, and (c) reference spectrum of
starting materials 2.

ACS Applied Materials & Interfaces Research Article

E dx.doi.org/10.1021/am300560z | ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

interactions between anionic surfactant and cationic particles.
Furthermore, the presence of a phenyl ring in SDBS, also seen
in polyoxoethylene(100) nonylphenyl ether, known to be
beneficial in solubilizing carbon nanotubes,62 may provide
additional π-surface interactions that aid both the observed
nanodiamond assemblies and lipid removal.
Role of Nanodiamond Particles. Unlike oily soils, solid
and crystalline lipids are reported to be removed by direct
solubilization, rather than emulsification, or other mechanisms.6
When surfactant is adsorbed to the lipid surface, repulsive

interactions between hydrated, and especially charged head-
groups relative to micellar nanodiamond assemblies in

suspension provide an energetic driving force for this lipid
solubilization process. Introduction of oleic acid to triolein is
known to cause rapid and large increase in solubilization
rates,48 and the nanodiamond assemblies might similarly

introduce negatively charged headgroups by multiple mecha-
nisms. Headgroup repulsion (whether charge or hydration

induced) would be exacerbated at sites of negative curvature

where the nanodiamond aggregates meet the lipid interface and
hence the roughening of the surface provided by the 5 nm
nanodiamonds also seems a key driver in the enhanced lipid
removal process (Scheme 2).
The solubilization process is thus directly linked to the
presence and size of the particles and would act to enhance the
recognized close packing of surfactant molecules on a lipid
surface that decreases the energy required for lipid removal.3,63
This seems a plausible way of aiding a phase inversion process
(interfacial water-in-oil to bulk oil-in-water) at the lower
temperature observed, although not one that has been
described previously. From the present data we cannot
distinguish whether the nanodiamonds substantially remain
on the surface (Scheme 2c), or are removed along with the
lipid-surfactant aggregates (Scheme 2d).
To further investigate the effect of interfacial charge on lipid
removal, we exposed tristearin lipid-coated QCM sensors to
positively charged diamond nanoparticle suspensions (2 or 4,
0.1 g/L), and then 1 colloid (0.1 g/L) followed by SDBS
surfactant (40 mmol/L). The negatively charged particles 1
were expected to neutralize the positive zeta potential of the
initially deposited particles (2 or 4), and weaken the

electrostatic attractions between the surfactant and nano-
particles. A colloidal suspension of nanodiamond 1 was

successfully deposited on the lipid layer with nanoparticles 4,
reflected by a decrease in frequency after exposure to 1 (Figure
8, right). As expected, there was an initial adsorption after
swapping the buffer from 1 to 40 mmol/L SDBS immediately
followed by desorption of lipid. However, compared to the lipid
removal shown in Table 1, the removal process was now
observed to be quicker in the presence of 4 and 1 combined,
with a significant decrease in time t from 600 to 200 min.
Similar effects were seen for particles 2 (Table 1). These
phenomena further confirmed that the rate of lipid removal is
largely determined by the interactions between particles and
surfactant molecules.64 They also imply that it is possible to
control the removal process by tuning the charge environment
on the particle surface.
■ CONCLUSIONS
Nanodiamond and nanodiamond−surfactant aggregates have
been demonstrated for the first time to enhance the removal of
tristearin, a model lipid, in the presence of both anionic and

Figure 7. IR spectra with aqueous zeta potential of samples obtained
from ND97 by thermal treatment in air as follows: (a) at 495 °C for 3
h; (b) at 485 °C for 5 h; (c) at 485 °C for 3 h; (d) at 485 °C for 1 h;
(e) at 425 °C for 5 h; and (f) reference spectrum of ND97. There is a
strong correlation between zeta potential and the intensity of the
carbonyl stretch.

Figure 8. Frequency and dissipation changes to thirrd overtone of the tristearin-coated sensors measured by QCM-D. Left: region (i) 0−7 min,
water; region (ii) 7−120 min, amine-functionalized nanodiamond particles 4; region (iii) 120−720 min, SDBS solution. In the presence of
nanoparticles 4 with zeta potential of 48.1 mV, R1 = 364 Hz, R2 = 583 Hz in 600 min. Right: region (i) 0−7 min, water; region (ii) 7−14 min,
nanodiamond suspension 4; region (iii) 14−21 min, colloid 1 was used to neutralize the positive charged surface of nanoparticles, and region (iv)
21−225 min, SDBS was used for lipid removal. In the presence of oppositely charge nanoparticles, R1 = 670 Hz, R2 = 785 Hz after exposure to SDBS
solution for 204 min. Note that equilibrium has not been fully reached in either case; [4] =[1] = 0.1 g/L, [SDBS] = 40 mmol/L, T = 25 °C.
ACS Applied Materials & Interfaces Research Article

F dx.doi.org/10.1021/am300560z | ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

nonionic surfactant. Most significantly, this effect is observed
between 15 and 25 °C and the surfactants used are found in
typical laundry detergent preparations. Acute, long-term
toxicity and other health effects of nanoparticulates are under
scrutiny however.65,66 Together with other commercial and
environmental considerations, this may limit their use for this
new application at this time. Nanoparticles appear to promote
lipid removal by roughening the surface to enhance surfactant
adsorption from solution and providing sites of negative
curvature at the interface that we suggest serve to lower the
energy of phase inversion and perhaps formation of an

intermediate phase.48,67 There is also dependence on nano-
diamond surface functionality whereby strong electrostatic

interactions between nanoparticles and surfactant encourage
adsorption, but hinder the subsequent release of lipid-surfactant
aggregates into the solution phase. Lipid removal thus
correlates with nanodiamond surface chemistry and zeta
potential, providing new tools with which to explore alternative
structures that bring similar advantage to this important
process.
■ ASSOCIATED CONTENT
*S Supporting Information
Structures of surfactants and lipid tristearin used. Detailed
methods and images for TEM and XRD pattern analysis of
untreated nanodiamond powder ND97; curves of the Z-average
size of ND97 particles in nonionic surfactant solution, zeta
potential of ND97 particles in nonionic solutions, Methods for
QCM-D, and QCM-D data for adsorption of nanodiamond
suspensions and colloids on tristearin surfaces, including
frequency and dissipation changes for lipid removal by SDBS
with different concentrations and at different temperatures (15
and 25 °C). Lipid removal by 40 mmol/L SDBS solution in the
presence of nanoparticles with different zeta potential, lipid
removal by nonionic surfactant G1C10 (3.1 mmol/L and

cationic surfactants (2 mmol/L) in the presence or absence of
nanoparticles, lipid removal by other surfactants in the presence
or absence of nanoparticles. This information is available free of
charge via the Internet at http://pubs.acs.org/.
■ AUTHOR INFORMATION
Corresponding Author
*E-mail: a.marsh@warwick.ac.uk.
Notes
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
We are grateful to EPSRC and P&G plc for funding (Grant
Reference EP/H034269/1) and thank Drs David York and
Alan Brooker (P&G plc) for helpful discussions. We thank Dr
Richard Walton, Department of Chemistry, University of
Warwick for assistance in collecting and analyzing the X-ray
powder diffraction patterns, Mr Steve York, Department of
Physics, for obtaining the TEM images and Mr Su Shi, School
of Engineering and Applied Science, Aston University, for his
assistance in CVD treatment of nanodiamond powder and Miss
Dorota A. Dobrzanska for the preparation of 1-(4-methyl-4-

oxopiperazin-1-yl)undec-10-en-1-one. The QCM-D, powder X-
ray diffraction, Raman microscope and light scattering equip-
ment used in this research were obtained through Birmingham

Science City: Innovative Uses for Advanced Materials in the
Modern World with support from Advantage West Midlands

(AWM) and part funded by the European Regional Develop-
ment Fund (ERDF). EPSRC are thanked for funding solid-state

NMR infrastructure and for CASE PhD funding with
GlaxoSmithKline for AST.
■ REFERENCES
(1) Raney, K. H.; Benton, W. J.; Miller, C. A. J. Colloid Interface Sci.
1987, 117, 282.

Scheme 2. Nanodiamond Plays a Key Role in Surfactant-Mediated Removal of Lipid. (a) Adsorption of Nanodiamond−
Surfactant Complex May Promote Insertion of Surfactant in Lipid Layer, Interfacial Phase Inversion and Tristearin Lift-off, (b)
Lipid−Surfactant Vesicle Budding, (c, d) Solubilization with or without Removal of Nanodiamond from the Surface
ACS Applied Materials & Interfaces Research Article

G dx.doi.org/10.1021/am300560z | ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

(2) Raney, K. H.; Miller, C. A. J. Colloid Interface Sci. 1987, 119, 539.
(3) Backstrom, K.; Lindman, B.; Engstrom, S. Langmuir 1988, 4, 872.
(4) Sonesson, A. W.; Callisen, T. H.; Elofsson, U. M.; Brismar, H. J.
Surfactants Deterg. 2007, 10, 211.
(5) Thirunavukarasu, K.; Edwinoliver, N. G.; Anbarasan, S. D.;
Gowthaman, M. K.; Iefuji, H.; Kamini, N. R. Process Biochem. 2008, 43,
701.
(6) Miller, C. A.; Raney, K. H. Colloid Surf., A 1993, 74, 169.
(7) Market Transformation Programme Briefing Note BNW05;
DEFRA, Ed. 2008.
(8) Aehle, W. Enzymes in Industry: production and applications, 3rd
ed.; Wiley-VCH: Weinheim, Germany, 2007.
(9) Snabe, T.; Petersen, S. B. Chem. Phys. Lipids 2003, 125, 69.
(10) Mori, F.; Lim, J. C.; Raney, O. G.; Elsik, C. M.; Miller, C. A.
Colloids Surf. 1989, 40, 323.
(11) Weerawardena, A.; Drummond, C. J.; Caruso, F.; McCormick,
M. Langmuir 1998, 14, 575.
(12) Robb, I. D.; Stevenson, P. S. Langmuir 2000, 16, 7939.
(13) Wells, D.; Fong, C.; Krodkiewska, F.; Drummond, C. J. J. Phys.
Chem. B 2006, 110, 5112.
(14) Sonesson, A. W.; Brismar, H.; Callisen, T. H.; Elofsson, U. M.
Langmuir 2007, 23, 2706.
(15) Weerawardena, A.; Drummond, C. J.; Caruso, F.; McCormick,
M. Colloid Surf., A 1999, 146, 185.
(16) Hussain, Y.; Krim, J.; Grant, C. Colloid Surf., A 2005, 262, 81.
(17) Boyd, B. J.; Drummond, C. J.; Krodkiewska, I.; Weerawardena,
A.; Furlong, D. N.; Grieser, F. Langmuir 2001, 17, 6100.
(18) Rojas, O. J.; Stubenrauch, C.; Schulze-Schlarmann, J.; Claesson,
P. M. Langmuir 2005, 21, 11836.
(19) Tongcumpou, C.; Acosta, E. J.; Quencer, L. B.; Joseph, A. F.;
Scamehorn, J. F.; Sabatini, D. A.; Chavadej, S.; Yanumet, N. J.
Surfactants Deterg. 2003, 6, 205.
(20) Beaudoin, S. P.; Grant, C. S.; Carbonell, R. G. Ind. Eng. Chem.
Res. 1995, 34, 3307.
(21) Nickel, D.; vonRybinski, W.; Kutschmann, E. M.; Stubenrauch,
C.; Findenegg, G. H. Fett-Lipid 1996, 98, 363.
(22) Tanthakit, P.; Nakrachata-Amorn, A.; Scamehorn, J.; Sabatini,
D.; Tongcumpou, C.; Chavadej, S. J. Surfactants Deterg. 2009, 12, 173.
(23) Mochalin, V. N.; Shenderova, O.; Ho, D.; Gogotsi, Y. Nat.
Nanotechnol. 2012, 7, 11.
(24) Krueger, A.; Lang, D. Adv. Funct. Mater. 2012, 22, 890.
(25) Huang, H.; Pierstorff, E.; Osawa, E.; Ho, D. Nano Lett. 2007, 7,
3305.
(26) Puzyr, A. P.; Baron, A. V.; Purtov, K. V.; Bortnikov, E. V.;
Skobelev, N. N.; Moginaya, O. A.; Bondar, V. S. Diam. Relat. Mater.
2007, 16, 2124.
(27) Ferrari, A. C.; Robertson, J. Philos. Trans. R. Soc., A 2004, 362,
2477.
(28) Premkumar, T.; Mezzenga, R.; Geckeler, K. E. Small 2012,
10.1002/smll.201101786.
(29) Gibson, N.; Shenderova, O.; Luo, T. J. M.; Moseenkov, S.;
Bondar, V.; Puzyr, A.; Purtov, K.; Fitzgerald, Z.; Brenner, D. W. Diam.
Relat. Mater. 2009, 18, 620.
(30) Osswald, S.; Yushin, G.; Mochalin, V.; Kucheyev, S. O.; Gogotsi,
Y. J. Am. Chem. Soc. 2006, 128, 11635.
(31) Takimoto, T.; Chano, T.; Shimizu, S.; Okabe, H.; Ito, M.;
Morita, M.; Kimura, T.; Inubushi, T.; Komatsu, N. Chem. Mater. 2010,
22, 3462.
(32) Chen, M.; Pierstorff, E. D.; Lam, R.; Li, S.-Y.; Huang, H.;
Osawa, E.; Ho, D. ACS Nano 2009, 3, 2016−2022.
(33) Liu, R. H.; Kay, B. K.; Jiang, S. Y.; Chen, S. F. MRS Bull. 2009,
34, 432.
(34) Rouhana, L. L.; Jaber, J. A.; Schlenoff, J. B. Langmuir 2007, 23,
12799.
(35) Rodahl, M.; Hö
ö
k, F.; Fredriksson, C.; A. Keller, C.; Krozer, A.;
Brzezinski, P.; Voinova, M.; Kasemo, B. Faraday Discuss. 1997, 107,
229.
(36) Hö
ö
k, F.; Rodahl, M.; Kasemo, B.; Brzezinski, P. Proc. Natl.
Acad. Sci. U.S.A. 1998, 95, 12271.

(37) Ohlsson, G.; Tigerstrom, A.; Hö
ö
k, F.; Kasemo, B. Soft Matter

2011, 7, 10749.
(38) Reviakine, I.; Johannsmann, D.; Richter, R. P. Anal. Chem. 2011,
83, 8838.
(39) Voinova, M. V.; Rodahl, M.; Jonson, M.; Kasemo, B. Phys. Scr.
1999, 59, 391.
(40) Metz, G.; Wu, X. L.; Smith, S. O. J. Magn. Res., Ser. A 1994, 110,
219.
(41) Khitrin, A.; Fung, B. M. J. Chem. Phys. 2000, 112, 2392.
(42) Rosen, M. J. Surfactants and Interfacial Phenomena; Wiley: New
York, 2004.
(43) Kane, R. S.; Deschatelets, P.; Whitesides, G. M. Langmuir 2003,
19, 2388.
(44) de Gennes, P. G. Adv. Colloid Interface Sci. 1987, 27, 189.
(45) Caruso, F.; Serizawa, T.; Furlong, D. N.; Okahata, Y. Langmuir
1995, 11, 1546.
(46) Weerawardena, A.; Boyd, B. J.; Drummond, C. J.; Furlong, D. N.
Colloid Surf., A 2000, 169, 317.
(47) Naderi, A.; Claesson, P. M. Langmuir 2006, 22, 7639.
(48) Chen, B.-H.; Miller, C. A.; Garrett, P. R. Langmuir 1998, 14, 31.
(49) Chu, C. D.; Perevedentseva, E.; Yeh, V.; Cai, S. J.; Tu, J. S.;
Cheng, C. L. Diam. Relat. Mat. 2009, 18, 76.
(50) Buriak, J. M. Chem. Commun. 1999, 1051.
(51) Sun, Q.-Y.; de Smet, L. C. P. M.; van Lagen, B.; Giesbers, M.;
Thune, P. C.; van Engelenburg, J.; de Wolf, F. A.; Zuilhof, H.;
Sudholter, E. J. R. J. Am. Chem. Soc. 2005, 127, 2514.
(52) Steenackers, M.; Lud, S. Q.; Niedermeier, M.; Bruno, P.; Gruen,
D. M.; Feulner, P.; Stutzmann, M.; Garrido, J. A.; Jordan, R. J. Am.
Chem. Soc. 2007, 129, 15655.
(53) Donnet, J.-B.; Fousson, E.; Delmotte, L.; Samirant, M.; Baras,
C.; Wang, T. K.; Eckhardt, A. C. R. Acad. Sci., Ser. IIc: Chim. 2000, 3,
831.
(54) Dubois, M.; Guerin, K.; Petit, E.; Batisse, N.; Hamwi, A.; ́
Komatsu, N.; Giraudet, J.; Pirotte, P.; Masin, F. J. Phys. Chem. C 2009,
113, 10371.
(55) Shames, A. I.; Panich, A. M.; Kempiski, W.; Alexenskii, A. E.;
Baidakova, M. V.; Dideikin, A. T.; Osipov, V. Y.; Siklitski, V. I.; Osawa,
E.; Ozawa, M.; Vul, A. Y. J. Phys. Chem. Solids 2002, 63, 1993.
(56) Panich, A. M. Diam. Relat. Mat. 2007, 16, 2044.
(57) Alam, T. M. Mater. Chem. Phys. 2004, 85, 310.
(58) Cunningham, G.; Panich, A. M.; Shames, A. I.; Petrov, I.;
Shenderova, O. Diam. Relat. Mat. 2008, 17, 650.
(59) Fang, X.; Mao, J.; Levin, E. M.; Schmidt-Rohr, K. J. Am. Chem.
Soc. 2009, 131, 1426.
(60) Levin, E. M.; Fang, X. W.; Bud’ko, S. L.; Straszheim, W. E.;
McCallum, R. W.; Schmidt-Rohr, K. Phys. Rev. B 2008, 77.
(61) Schmidt-Rohr, K.; Rawal, A.; Fang, X. W. J. Chem. Phys. 2007,
126.
(62) Lee, Y.; Geckeler, K. E. Adv. Mater. 2010, 22, 4076.
(63) Malmsten, M.; Lindman, B. Langmuir 1989, 5, 1105.
(64) Weiss, J.; McClements, D. J. Langmuir 2000, 16, 5879.
(65) Krueger, A. J. Mater. Chem. 2011, 21, 12571.
(66) Mohan, N.; Chen, C. S.; Hsieh, H. H.; Wu, Y. C.; Chang, H. C.
Nano Lett. 2010, 10, 3692.
(67) Buchanan, M.; Egelhaaf, S. U.; Cates, M. E. Langmuir 2000, 16,
3718.

ACS Applied Materials & Interfaces Research Article

H dx.doi.org/10.1021/am300560z | ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Study about NanoDiamonds

Read more