Potentiometric sensors (ion-selective electrodes or ISEs) are well known and have been used for many years for important applications in clinical measurements (e.g. blood electrolyte analysis) and environmental monitoring (e.g. measurement of phosphates, nitrates and chlorides in rivers and lakes). Perhaps the best-known sensor of this type is the glass electrode, which is normally used for pH measurements. In a typical ISE, the sensing element is a membrane, which selectively interacts with the target species in the sample. This interaction leads to the generation of a membrane potential, which can be measured, with a voltmeter. Changes in the observed potential can thus be related to variations in the amount of target species in the sample. 

In a conventional ISE the sensor is filled with a so-called internal filling solution, which is required to stabilise the internal side of the membrane and maintain electrical contact with an electrode through which the signal is conveyed to the meter (figure 1).

The presence of this filling solution is a real problem in terms of automation of the manufacturing process, and compatibility with rapidly developing planar technology fabrication processes used in the microelectronics industry. It also restricts the design and therefore mode of operation of the device as it is difficult to minimise and must normally be used in a upright manner to prevent air bubbles becoming trapped at the internal membrane surface.

 PLANAR DESIGNS UNDER DEVELOPMENT AT BEST CENTRE

At the BEST Centre, we are currently working on a solid-state, plannar version of these sensors which combines screen printing technology and expertise from the University of Ulster, with hydrogel formulations developed at QUB, and PVC ion-selective membrane and ISE fabrication expertise from DCU. The device currently under evaluation is represented schematically in Figure 2. The electrodes are screen printed onto a flexible plastic backing and one is covered with a salt-doped adhesive hydrogel, which replaces the internal filling solution. The PVC sensing membrane is attached to the hydrogel layer and the whole device sealed with a covering film which has two contact holes for the sensing membrane and the uncovered reference electrode.

APPLICATIONS 

The new sensors have been found to perform very well, and we are now identifying potential applications. One, which we have already started to investigate, is to develop a new, simple diagnostic test for Cystic Fibrosis. Cystic Fibrosis (CF) is the commonest lethal hereditary disease amongst Caucasians, and there is a particularly high incidence of the disease in Ireland and Britain. For example, in the UK, around 2 million people are carriers of the defective CF Gene which gives rise to the condition, and 3 people die from its effect every week. The disease affects the glands, which secrete body fluids, damaging many organs including the lungs, the pancreas and the digestive tract and the reproductive system. Symptoms of CG in new born infants include a blocked bowel at birth, chest infections and poor growth. 

EXISTING DIAGNOSIS

In CF, abnormal exocrine secretions contain too little water relative to protein and electrolyte concentrations, leading to a raised sodium (Na) level of about 60 mM in sweat compared to about 20mM normally. The detection of the abnormal Na level in sweat forms the basis of the ‘Gold Standard’ Pilocarpine iontophoresis test, which is universally used as diagnostic confirmation of CF in infants. The test was devised by Gibson and Cooke in 1959 and later further developed by Tocci and Mckey. While some further work involving debate on whether the chloride or sodium ion is the best marker has been published, there has been no development of the technique, which involves collection of at least 100 mg of sweat (this is difficult in infants, and the technique of pilocarpine iontophoresis is used to stimulate sweat generation), followed by laboratory analysis by ion-selective electrode and / or atomic absorbance spectroscopy. The existing analytical technique therefore suffers from the following disadvantages:

• Difficult sampling
• Relatively large sample volumes required
• Centralised testing needed leading to delays in diagnosis
• Skilled and experienced personnel required

In addition, even experienced personnel are disappointed with the levels of false diagnosis generated by both false positive and negative results with the existing test. Clearly there is a need for an improved method of CF diagnosis.

ADVANTAGES OF THE PROPOSED DESIGN

By covering the outer surface of the novel device with a bioadesive coating, it should be possible to stick the sensor directly to the skin of the patient.

1. No sampling required as the sweat in direct contact with the sensor.
2. Compatible with cheap, portable instrumentation – no need to send samples to central facility. Diagnosis can be performed on the spot.
3. Improved precision and accuracy arising from better sensor design, simple calibration, and small sample volumes required.
4. Sensor patch can be fabricated using planar design compatible with cheap, mass production technologies such as screen-printing. The high degree of reproducibility offered by planar fabrication technology makes the sensor performance more predictable, and hence calibration is simpler.
5. Cheap fabrication enables a disposable sensor regime to be employed. No possibility of cross-infection while in use, or for expensive and chemically/physically hostile sterilisation procedures.

Preliminary results (Figure 3) suggest that our design will be able to discriminate levels of sodium chloride typically found in CF positive patients (sodium levels are typically>60mM in CF positive patients and around 20-30 mM in normal patients). During the coming year, we intend to finalise our design and ideally commence a limited clinical trial for the device with the aid of clinical specialists from the CF unit in Our Lady’s Hospital for Sick Children, Crumlin, Dublin. This unit currently performs approximately 200 tests per year, giving a good comparative basis for determining the performance reliability and accuracy of the new system.

We anticipate a wide range of applications for this technology, which will enable a wide range of low-cost disposable, planar sensors to be fabricated by simply changing the sensing component in the PVC membrane. The stability and reproducibility of the responses obtained (Figure 3) suggest that it may be possible to develop methods, which are calibration free.