QSense sensors are optically polished and have a surface roughness of <3 nm, RMS. All other coatings are added on top of the Au layer, which means that the surface roughness of other materials is equal or slightly larger than 3 nm RMS.
The sensitivity distribution becomes narrower for higher overtones. The distribution is Gaussian and approaches zero at the edges in an asymptotic way. It is therefore difficult to discuss active area and sensitivity dependence on radius/surface of the sensor. In addition, all calculations in QSoft and QTools assume that the complete surface is covered or at least homogeneously covered and any calculations should therefore be done over the whole surface.
The sensor can be coated with almost any material, as long as it can be applied as a thin (nm range), homogenous layer firmly attached to the underlying surface. The layer thickness can vary between nanometers and micrometers, depending on the viscoelastic properties of the applied material. QSense offers pre-coated sensors with e.g., gold, Ti, SiO2, AlO3, stainless steel, hydroxyapatite, polystyrene and biotin. Several other materials are also available, e.g., most metals, metal oxides or spin-coated polymers.
The layer is rigid enough when the overtones overlap. If they start to deviate, it is up to the user to determine if the mass estimate with the Sauerbrey relation will be a sufficient approximation. This can be done by comparing the Sauerbrey masses calculated using the lowest and highest harmonic, respectively.
Maximum possible thickness very much depends on the viscoelastic properties of the polymer in question. With a very viscous polymer such as silicone it can be hard to oscillate the sensor already at a thickness of 1 micron. However, for a stiffer polymer it might work with a several microns thick adlayer. For example, up to 10 microns thick polystyrene adlayers have been used with a still functioning sensor. But generally—the thinner, the better!
The reason why one sensor has higher absolute dissipation than others is most likely not due to roughness. The absolute D can be a little higher than normal due to how various resonance modes couple in the crystal, which in turn depends on crystallographic defects and small variations polishing that give small geometrical effects. All sensors have many vibrational modes that you see as peaks on the plot when QSoft sweeps the frequency as it is finding the main resonance peak. If one of these other modes are close to the main mode, this drains energy from the main mode and hence a results in higher dissipation. However, since you are measuring dissipation shifts, this normally does not influence your measurement. However, if the coupling is too large, you will see this as an increase in noise and drift since the other modes are not as stable as the main mode.
If a small notch is introduced to the sensor edge during handling, how will it influence the performance?
This depends on the size and location of the notch. First, it must be small enough so the seal at the O-ring remains intact (approximately 1.6 mm/1 mm thickness for standard and other modules respectively). Secondly, it should not affect the electrodes. We cannot guarantee a good performance if this is the case but it might still be possible to use it with good results. You are recommended to ensure that the resonance peaks can be found with normal dissipation values, in which case the sensor can be used without problem.
It is good, but not necessary, to check the resonances in air to make sure that they look good. One can routinely check the resonances in air before introducing the fluid in the modules.
Yes. Ethanol can be used to clean polystyrene.
The lower limit for hydroxyapatite is between pH 5 and 5.5. Below this, HA starts to alter and dissolve. Basic solutions should be no problem.
Cleaning hydroxyapatite sensors (QSX 327) with SDS may leave residues on the surface.
SDS in an anionic detergent and should not adsorb to negatively charged surfaces such as SiO2. However, one thing to consider is that SDS hydrolyzes with time, forming a hydrophobic product (a phase change in the SDS solution) that might act as a contaminant on the surface. So, to be on the safe side, fresh SDS solutions should be used if the surface cleaning is performed in situ.
UV/ozone treatment after SDS will remove these residues.
Heating will influence the sensors, but the extent depends on temperature, exposure time and type of sensor.
At 573 °C the quartz undergoes a transition from a to b phase, which alters its piezoelectric properties. Both the quartz and the Au coating withstand temperatures below this point, however the adhesion layer (Cr) is known to migrate in Au, and the migration rate is affected both by temperature and time. We have verified that 500 °C for 5 minutes is okay for the QSX301. However, we cannot guarantee the performance of any additional coating outside the standard working temperature range of the E-Series (15-65 °C).
Significant Cr migration has been seen at annealing temperatures of 250 °C and 450 °C for periods of 0.5-22 h. So either lower temperature for longer times or high temperature for shorter time would be recommended.
For coated sensors (i.e. all but QSX 301), an additional adhesion layer and the coating material has to be taken into account, which makes it even more complex.
The resonance frequency of the sensor is a function of the motion of the sound wave in the sensor and out in the layer on top of the sensor (attached molecules and liquid medium). If the layer on top of the sensor surface (for example a layer of adsorbed molecules) is very thin and rigid it will move in phase with the sensor and the influence of this layer on the motion of the sensor will be negligible. In this case all overtones will give the same response and the Sauerbrey equation is valid. If an attached layer is soft it will not move entirely in phase with the sensor and this will influence and dampen the motion of the sensor. The amount of dampening is a function of the viscoelastic properties of the attached layer. Also, since each overtone makes the sensor move in a different way, the viscoelasticity influences each overtone in a different way. This is the reason of the spreading of the overtones in the case of a soft layer.
There should be no double peaks at any harmonic. By double peaks one means two peaks that are of the similar height and the width of the peaks are partially overlapping. If one peak is a lot higher than the other, the measurement will work just fine anyway. In order to analyze data with QTools, at least two harmonics are needed. So, not using one of the overtones will be okay.
It is common to coat the QCM-D sensors ex situ by immersing into different coating solutions outside the measurement chamber. One can then get a rough mass estimation by first acquiring a baseline of the uncoated sensor in air and then stitching the file in QSoft (open both files in QSoft and go to ‘File’ › ‘Stitch data file’) with the baseline from the coated sensor in air.
Please note that the error can be quite large since placing the sensor differently onto the O-ring each time might give different frequency changes. It is still possible to get a rough idea of the adsorbed mass, but it is important to be careful and repeat the experiment with different sensors and calculate a mean ± SD. After immersion it is also important to remove residues from the counter gold electrode before remounting into the QCM-D flow module.
The standard sensor has a diameter of 14 mm. Because the sensor is most sensitive in the center, it is important to ensure that the same reaction is taking place over the entire surface in order to compare measurements, i.e., that your sample forms a homogeneous film.
The notch depicts the x-axis of quartz crystal, which is the direction of oscillation. The x-axis is perpendicular to the flat part of the notch.
The sensor oscillates in thickness shear mode, which means that the movement is in-plane. Ideally, this would be the only motion, but it is not possible to totally eliminate out-of-plane oscillation (flexure mode)*. In liquid, this mode causes pressure waves to occur that can disturb the normal oscillation. The fundamental 5 MHz oscillation is most perceptible to this disturbance.
*If you fix a rubber band between two points and then pinch it in the middle and move the center towards one side you will see how the band gets thinner on one side and buckles on the other. The same thing happens with the sensor (but on a much, much smaller scale) when it oscillates since the oscillation amplitude is larger in the center of the sensor than at the edges.