Primary surface measurement reported
Equilibrium surface tension (aqueous/air) and oil–water interfacial tension (aqueous/WO-15) were reported as a function of NCC or surfactant concentration (Figures 8–9).
Client Citation Analysis
Primary surface measurement reported
Equilibrium surface tension (aqueous/air) and oil–water interfacial tension (aqueous/WO-15) were reported as a function of NCC or surfactant concentration (Figures 8–9).
Dropometer attribution in the paper
The Dropometer is cited as a “smartphone-based pendant drop tensiometer (Droplet Lab, Markham, ON, Canada)” using droplet-profile fitting with the Young–Laplace equation via axisymmetric drop shape analysis (ADSA).
How the surface-tension / interfacial-tension data were used in the study
The surface- and interfacial-tension trends are used to contrast NCC’s weak surface activity with the strong tension reductions observed for molecular surfactants and to discuss how interfacial tension behavior relates to concentration-dependent catastrophic phase inversion behavior.
Replication / reliability statement
Each measurement was repeated 10 times per solution, and average values were reported.
Primary surface / interfacial measurement
Equilibrium surface tension of aqueous phases containing NCC or surfactants, and oil–water interfacial tension (WO-15 oil vs aqueous phase) as a function of stabilizer concentration (Figures 8–9).
Supporting measurements
Catastrophic phase inversion behavior was tracked using conductivity measurements during incremental aqueous-phase addition under rotor–stator mixing, and emulsion coalescence stability was evaluated by monitoring separated aqueous-phase volume over time in graduated cylinders.
Surface tension / interfacial tension
smartphone-based pendant drop tensiometer (Droplet Lab, Markham, ON, Canada)
High-shear dispersion/emulsification
variable-speed Gifford–Wood (Hudson, NY, USA) rotor–stator homogenizer (Model 1-L) equipped with an open slotted stator head
Emulsion stability (phase separation volume)
500 mL graduated cylinders
The Dropometer (cited as a smartphone-based pendant drop tensiometer from Droplet Lab) was used for pendant-drop measurements in which a pendant droplet of the aqueous phase (containing NCC or surfactants) was formed at a stainless-steel needle tip and analyzed by fitting the droplet profile with the Young–Laplace equation using axisymmetric drop shape analysis (ADSA) to obtain surface tension (droplet suspended in air) and interfacial tension (droplet dispensed into a sealed quartz cuvette containing WO-15 oil).
These surface- and interfacial-tension curves are used in the paper to compare stabilizer surface activity and to discuss a low-concentration regime of sharp interfacial-tension reduction versus a higher-concentration regime where the interface is described as saturated in relation to catastrophic phase inversion behavior.
If you need surface and interfacial tension curves for surfactants or particle stabilized emulsions, we can show you a pendant drop measurement on your formulations.
NCC shows weak equilibrium surface-tension reduction
For NCC, equilibrium surface tension remains relatively high across the tested range, decreasing from 63.4 mN/m at 0.1 wt% to 62.3 mN/m at 2.0 wt% (Figure 8), and the paper characterizes this as negligible surface-tension reduction.
All four surfactants reduce surface tension with concentration
SDS and OTAC show large decreases in equilibrium surface tension from 0.1 wt% to 1.0 wt% (SDS: 44.23 → 33.74 mN/m; OTAC: 54.80 → 38.22 mN/m), while Amphosol and Alfonic reach 31.25 mN/m and 26.11 mN/m at 0.5 wt%, respectively (Figure 8).
Surfactants produce dramatic oil–water interfacial-tension reductions
The interfacial tension data show large reductions for surfactant systems, including SDS decreasing from 57.2 mN/m (no surfactant) to 4.61 mN/m at 0.1 wt% and 3.02 mN/m at 1 wt%, and Amphosol reaching 2.81 mN/m at 0.2 wt% while remaining below 3.4 mN/m across the tested range (Figure 9).
NCC shows limited, non-monotonic IFT behavior
For NCC, the paper reports an initial decrease in interfacial tension at 0.1–0.2 wt% reaching a minimum of 38.3 mN/m, followed by a plateau or slight increase at higher concentrations (Figure 9).
IFT concentration behavior is explicitly linked to phase inversion interpretation
The paper describes a sharp reduction in interfacial tension at lower surfactant concentrations before stabilizing near a minimum and states that this rapid initial drop coincides with a delay in catastrophic phase inversion; it also describes a higher-concentration threshold where the interface becomes saturated and further IFT decreases are minimal, while inversion occurs at lower aqueous phase fractions as excess surfactant accumulates in the bulk or micellar phase.
What it shows
Shows equilibrium surface tension (aqueous/air) for NCC and surfactants (SDS, OTAC, Amphosol, Alfonic) as a function of concentration.
What it shows
Shows oil–water interfacial tension (WO-15 oil / aqueous phase) for NCC and surfactants as a function of concentration, including sharp surfactant-driven reductions and a non-monotonic NCC trend.
Within the study’s focus on catastrophic phase inversion and emulsion stability, the Dropometer-derived pendant-drop surface and interfacial tension data provide the interfacial context used to compare particle-stabilized and surfactant-stabilized systems.
The paper reads these tension–concentration relationships alongside its conductivity based inversion data: NCC shows weak surface activity yet delays catastrophic phase inversion, which the authors attribute to interfacial jamming, while surfactant-stabilized systems show concentration-dependent inversion behavior associated with interfacial saturation and bulk/micellar accumulation at higher surfactant concentrations.
Pendant-drop surface + oil–water interfacial tension in one workflow
The paper reports pendant-drop measurements for surface tension (droplet suspended in air) and interfacial tension (droplet dispensed into WO-15 oil in a sealed quartz cuvette) using Young–Laplace fitting with ADSA.
Replicated measurements with reported averaging
Each surface/interfacial tension measurement was repeated 10 times per solution and reported as an average at 22 ± 1 °C.
Clear differentiator between NCC and surfactants
The reported surface-tension data keep NCC near ~63 mN/m across concentration while surfactants substantially reduce surface tension (Figure 8), supporting a direct contrast in surface activity.
Interfacial-tension regimes used to interpret inversion behavior
The paper links a sharp low-concentration IFT decrease and a higher-concentration near-minimum/plateau (interface saturation) to the observed concentration-dependent catastrophic phase inversion behavior (Figure 9).
Comparative interfacial activity across surfactants
The interfacial tension results differentiate very low IFT values for SDS and Amphosol from more moderate reductions for Alfonic and a more gradual decrease for OTAC (Figure 9).
The authors used a smartphone-based pendant drop tensiometer (Droplet Lab, Markham, ON, Canada) to measure the surface tension of aqueous phases containing NCC or surfactants, and their interfacial tension against WO-15 oil. Drop profiles were fitted with the Young–Laplace equation using axisymmetric drop shape analysis, each measurement was repeated 10 times per solution, and all measurements were made at 22 ± 1 °C.
NCC barely lowered surface tension, from 63.4 mN/m at 0.1 wt% to 62.3 mN/m at 2.0 wt%, and its interfacial tension reached a minimum of 38.3 mN/m at 0.1 to 0.2 wt%. The surfactants lowered interfacial tension much further, for example SDS from 57.2 mN/m without surfactant to 3.02 mN/m at 1 wt%, and Amphosol to 2.81 mN/m at 0.2 wt%.
The authors note that the sharp drop in interfacial tension at low surfactant concentrations coincides with a delay in catastrophic phase inversion. At higher concentrations they describe the interface as saturated, with excess surfactant accumulating in the bulk or micellar phase and inversion occurring at lower aqueous phase fractions. Phase inversion itself was tracked by conductivity during aqueous phase addition, not by the tensiometer.