Surface mount solid tantalum capacitors are a well-established technology and have been broadly employed in medical devices for decades. There are many reasons to choose tantalum, including their inherent reliability, self-healing capabilities, absence of any known wear-out mechanism, and their ability to pack the highest capacitance values into the smallest case sizes. This article describes the differences between ceramic and tantalum technologies, as well as the different design approaches for commercial and medical-grade capacitors.
Ceramic vs. Tantalum
When engineers are designing a medical device, it’s common to have to choose between tantalum and ceramic capacitors.
Multilayer ceramic capacitors (MLCCs) have smaller plate/surface areas and significantly thicker layers than tantalum, but the dielectric materials have a much higher permittivity (i.e., dielectric constant). MLCCs can be manufactured in case sizes that are not practical for tantalum capacitors, 01005 and 0201 for example, but, in larger case sizes, tantalum can offer similar capacitance values to some MLCC technologies.
MLCCs are broken into two classes. Class 1, referred to as NP0 or C0G, are temperature-compensating capacitors and are composed of dielectric materials like titanium dioxide modified with additives of other elements, such as zinc, zirconium, niobium, etc., which are necessary to achieve certain desired linear characteristics. The have predictable temperature coefficients and do not have an aging characteristic. They also offer the most stable capacitance with respect to applied voltage, temperature, and, to some extent, frequency. Class 1 ceramic capacitors have the lowest volumetric efficiency among ceramic capacitors, which results from the relatively low permittivity of their dielectric materials. As such, Class 1 capacitors have lower capacitance values than tantalums or Class 2 ceramics.
Class 2 ceramic capacitors have a dielectric that exhibits high permittivity and, as a result, have better volumetric efficiency than Class 1 capacitors, but lower accuracy and stability. Based on the chemistry of barium titanate, Class 2 capacitors can provide a wide range of capacitance; with the individual cap value depending on the applied voltage. The ceramic dielectric is also characterized by a nonlinear change of capacitance over the temperature range, meaning that Class 2 capacitors age over time.
Tantalum capacitors, on the other hand, do not experience similar aging characteristics and have no known wear-out mechanism. They also achieve high capacitance values by having a large surface area and a very thin dielectric layer. Their internal structure is made up of millions of tantalum particles that are fused together to form a sponge-like structure. If someone were to flatten out this spongy structure, the total surface area would be more than 200 times greater than the footprint of the capacitor body, as shown in Figure 1. Also, in addition to having no known wear-out mechanism, the average reliability of tantalum capacitors actually improves over time (see Figure 2).
In the medical market, MLCCs are typically considered the best choice for applications with capacitance ranges below 1 μF, and tantalum capacitors are normally chosen for applications with capacitance values at or above 10 μF. In the 1–10 μF range, choices are based on relative size, application requirements for capacitance stability over temperature and voltage, and rated voltage capability.
The Importance of Design
There are many differences between commercial and medical components, starting with the way they are designed, and tantalum capacitors are no different.
The tantalum powder used to make commercial capacitors has a super fine particle size, which translates into higher volumetric efficiency or a higher CV/gram capability, with CV referring to the capacitors’ volume of capacitance and voltage. The smaller particle size enables the significant downsizing of existing ratings and a general extension of the capacitance available at a given voltage level. The drawback of these higher CV/gram powders is related to the reduced strength of the connection between the particles due to smaller necks and increased difficulty in depositing the counter electrode material due to small internal pore structures. Finer powders also have a higher oxygen content, which is an impurity that can lead to higher direct current leakage (DCL).
The formation ratio is another way of saying dielectric thickness, and it is based on the voltage that tantalum pellets are formed at rather than the rated voltage of the final component. Commercial components typically have formation ratios that do not exceed 2:1, meaning that a 50-V component will be formed at 100 V. The use of high CV powders restricts the thickness of the dielectric layer; since smaller particles leave less room for dielectric growth, they can only be used to make lower voltage parts. This also means that the 50 percent derating recommendation is even more important for commercial-grade components since the applied voltage is a higher percentage of the formation voltage. Medical components, on the other hand, will typically approach a 4:1 formation ratio, resulting in a thicker dielectric that provides greater reliability in terms of surge current and electric field handling capabilities.
One critical aspect of any medical tantalum capacitor — and, really, all components intended to be used in FDA Class II or III medical devices — is the control of the design and any changes made to it. There is a profound difference in the overall philosophy of components built for consumer electronics compared with those supplied against medical specifications. Commercial component manufacturers follow the general rule that if a design change does not alter the form, fit, or function of the component, then no change notification is required, but medical component manufacturers and their customers have a completely different opinion.
In general, a commercial facility may change the materials used to build parts at-will in order to achieve cost savings or utilize a readily available material supply, but not all designs are created equal in terms of inherent reliability. Commercial products do not have the lot-to-lot control of medical designs or tantalum powders. Medical components have the strictest change control requirements in the industry, and generally require customer approval for changes in design, raw materials, supplier moves, and several other aspects in order to satisfy FDA requirements for Class II and Class III devices.
Qualification and Process Differences
Once the tantalum components are designed, they go through a rigorous internal qualification process. Requirements from medical customers force manufacturers to have comprehensive change control management systems in place so that any changes are made only after thorough analysis, qualification, and approval from all required parties. In an environment where customers require consistent performance over time, it is necessary to have strict process controls. To supplement the control aspects, it is common practice to have detailed standard operating procedures and robust training programs in place, because these help ensure that operators perform the operations correctly and consistently.
There are also extensive manufacturing execution system (MES) controls in key areas to reduce the opportunity for human error and ensure that the correct process flow and conditions are followed. These systems use barcode scanners and machine lockouts to keep track of the products in process, prevent processes from being skipped or duplicated, and monitor both equipment conditions and personnel during the processing. The use of statistical process control (SPC) can also be implemented throughout medical component production to manage key processes with tighter limits than commercial products would receive. There are many important differences in the manufacturing processes for medical tantalum and commercial tantalum components (see Table 1). Some suppliers even use tighter statistical controls designed to eliminate outliers and maverick lots, which are not common procedures for commercial components.
The most important electrical parameter for a tantalum capacitor is the DCL, which directly affects the battery life of a device and can thus be crucial in medical devices like implantable cardiac pacemakers and implantable cardioverter defibrillators.
DCL is a strong indicator of the quality of the dielectric and the overall reliability of the component. So, conservative designs, change control, maverick lot control, proprietary testing, and tantalum powder selection all contribute to a lower DCL. Medical-grade tantalum capacitors typically have maximum DC leakage levels that are 25–50 percent of the levels specified for commercial capacitors. This lower leakage translates into improved reliability and longer battery life for medical devices.
There are many reasons to employ tantalum capacitors in medical device designs. The latest technologies, including new process control technologies, the use of conservative design rules, robust change control systems, and the introduction of improved testing methodology, have enabled continuous DCL improvements in tantalum capacitors.
In fact, these technological improvements have already achieved DCL levels that were not considered possible just a few short years ago and, since DCL is the most important electrical parameter for tantalum capacitors, have also resulted in medicalgrade tantalum capacitors with notably improved reliability and extended operating lives — both of which are critical design concerns in the medical market.
This article was written by Lizzie Geismar, Senior Product Manager for AVX. For more information, Click Here .