with mass loss from the cell being recorded as a function of time in a thermogravimetric (TG)-
type apparatus result in high precision [41-43 ]. Effusion cell design is decisive for the accurate
measurements by using this technique. The cell must be very light, less than 1 g (determined
by the capacity of a microbalance) and it must allow the use of samples of a few tens of milligrams.
The effusion cell is suspended on the arm of a commercial recording electro-balance.
At least 5 mass% of compound under study should be vaporized before measurements are
begun. This ensures removal of any volatile impurities possibly present in the sample.
Another measuring principle is based on a homogeneously heated Knudsen cell with
the effusion orifice in line with a cooled target, which is connected to a sensitive microbalance
[44]. The whole system is mounted within a special high vacuum chamber with a
built in cold trap. The Knudsen cell is heated to a pre-selectable test temperature within the
temperature range of 223-573 K, controlled by a microprocessor programmer. When an
orifice is opened, the impulse of the molecular beam with a defined geometry is detected
by the microbalance as a weight loss step. The vapour immediately condenses on the
cooled target producing a weight increase with time, proportional to the effusion rate. In
the pressure range 10-3-1 Pa, there are two possibilities for the calculation ofvapour pressure
values: (a) the impulse method gives the vapour pressure without knowing the molar
mass of the vapour; (b) the combination with the effusion rate value for the vapour pressure
allows the determination of the molar mass. This molar mass determination is precise
enough to discriminate between monomeric, dimeric and trimeric vapours. Below 10 -3 Pa,
the impulse values (weight signals) are too small for an exact calculation [44].
2.3.5 Non-Isothermal Knudsen Effusion Method in TG-Type Apparatus
The non-isothermal Knudsen effusion method was developed in response to the need to
scan over a wide range of temperatures in a modest time when investigating compounds
with extremely low volatility [41 ]. As the sample cell is in a high vacuum and must receive
heat purely by radiation, heat transfer is a key concern. A long time is generally required to
reach thermal equilibrium in an isothermal experiment. Achieving a new steady-state temperature
typically requires hours. The non-isothermal method is a straightforward modification
of the usual Knudsen effusion technique [41 ], generally requiting few changes in
equipment and only a limited change in procedures described above (see Section 2.3.4). The
advantage of the non-isothermal technique is that it does not require the cell to reach a
steady-state temperature. A temperature ramp is imposed by increasing temperature of the
heating block of the TG, but the temperature difference between the block and capsule is no
longer of concern. The capsule will rise in temperature at a rate that is in some way related
to the rise in block temperature, but it need not track it perfectly. Using the effusion cell of
a much lower mass, 0.15 g, its temperature is tracked very well.
2.3.6 Mass-Loss Knudsen Technique with a Quartz Crystal Microbalance
The conventional Knudsen effusion method uses a balance to measure the rate of mass loss
through the orifice. A piezoelectric quartz crystal can be used as a weighing device if the