METHOD FOR MEASURING PARYLENE THICKNESS USING

QUARTZ CRYSTAL MICROBALANCE

Henna Heinilä, Maunu Mäntylä and Pekka Heino

Institute of Electronics, Tampere University of Technology, P.O. Box 692, FI-33101 Tampere, Finland

Keywords: Parylene, Quartz Crystal Microbalance, Biomedical Coating.

Abstract: At present, the exact final thickness of parylene coating is difficult to specify in the beginning of the coating

process since the parylene thickness is a function of many components. The elements that control the

thickness are substrate surface area in a vacuum chamber, program parameters, and amount of dimer charge.

This paper describes a method for measuring parylene coating thickness using quartz crystal microbalance.

The thickness is measured by an oscillation frequency change of quartz crystal as parylene deposites on the

quartz crystal plate. These results can be used for specifying the parylene thickness real-time during the

coating process.

1 INTRODUCTION

Biomedical implants have many strict requirements

as they are being implanted for a long time under the

skin. Implantable medical devices need to be coated

hermetically before implantation. The coating

material has many strict requirements. One very

important, biological aspect is that the implant must

be biocompatible. To reduce inflammation, all of the

components of an implant should be nontoxic to

cells (Wolgemuth 2002). Since most of the materials

used in the device's electronics are not

biocompatible, the encapsulation of the device with

nontoxic materials is needed to prevent elusion into

the body. The human body is a very hostile

environment for any foreign materials. Therefore,

the implant must be biostable. This electrical

characteristic means that the device operation must

be protected from the living tissue (Wolgemuth

2002). Otherwise the body tries to destroy or isolate

the device. These two fundamental aspects, the

protection of the device against the biological

environment and the protection of living tissue

against device’s materials must be ensured before

the device is implanted in a human being

(Wolgemuth 2002). In addition, the long-term

stability of the device and the coating needs to be

high and meet the specifications.

Materials, which are used as coating materials

for medical applications and qualify above-

mentioned requirements, include many metals, metal

alloys, ceramics, polymers, and polymer composites

(Ratner et al. 1996). Small and rigid implantable

devices, like pacemakers and drug pumps, could be

coated with a metal case unlike the devices that must

be extremely small scale, like in MEMS

applications, or devices that must conform to the

tissue movements. These latter mentioned devices

like neural prosthesis should be made from flexible

substrate with flexible coating.

Many types of polymers are widely used for

medical purposes. Polymers like epoxies, silicones,

polyurethanes, and parylenes, have many desirable

properties, such as ease of tailoring and processing,

low cost, and an excellent corrosion resistance

(Ratner et al. 1996). Parylene conformal coating is

ideal for the medical applications because of its

many unique properties. Vacuum deposited parylene

is applied in a chamber by means of gas phase

polymerization (Noordegraaf & Hull 1997).

Compared to liquid coating processes, vacuum

deposited parylene coatings exhibit uniform

coverage of medical implants and electronics

components without the presence of pinholes or

pooling. During the coating process, all of the

exposed substrates in evacuated vacuum chamber

are coated and the coating grows as a conformal film

simultaneously on all surfaces and parylene

penetrates into the pinholes (Noordegraaf & Hull

1997). During the parylene coating process, no

impurities are generated and hence parylene coatings

222

Heinilä H., Mäntylä M. and Heino P. (2008).

METHOD FOR MEASURING PARYLENE THICKNESS USING QUARTZ CRYSTAL MICROBALANCE.

In Proceedings of the First International Conference on Biomedical Electronics and Devices, pages 222-226

DOI: 10.5220/0001056102220226

 SciTePress

are, like silicone, one of the highest purity coatings

on the market (Licari 2003). Parylene is flexible

coating material just like silicone, but compared to

silicone, parylene has better moisture and chemical

resistance and also extremely thin coatings are tight

enough for insulation and implants (Stieglitz et al.

2002). In addition, parylene has excellent adhesion

to most surfaces (Licari 2003, p. 157).

2 BACKGROUND

2.1 Parylene

Parylene (poly-para-xylylenes) is a universal term

for members of a unique polymer class (Licari

2003). By using dimer of di-para-xylylene, parylene

has the ability to be deposited by vacuum deposition

onto exposed surfaces at room temperature (Licari

2003). Varying the process parameters of deposition

can control the thickness of parylene and the

thickness may vary from 0,025 µm to several tens of

micrometers. According to Licari (2003), the final

thickness of coating can be controlled to ±10 % of

desired thickness. Nevertheless, our laboratory

results have proven that the thickness variance can

be even greater. The parylene coating is inert and

conformal and hence provides dielectric and

environmental isolation. Parylene coating is used in

many applications like aerospace, automotive and

military industry, and also in medical applications.

Nowadays, the four most frequently used

commercially available parylene variations are

parylene N, C, D, and HT. The two first have the

longest history of use and are most commonly used

in medical coating applications. This paper

concentrates on coating with the polymer parylene

C. (Specialty Coating System 2007)

Parylene C can provide extremely thin, uniform,

and pinhole-free coating. It has low electrical

dissipation factor, high dielectric and mechanical

strength, and good chemical, electrical, and

biological stability. It also has significantly lower

moisture, chemical, and caustic gas permeability

than parylene N. Above-mentioned reasons make

parylene C very compatible for medical implants. In

addition, parylene C is not cytotoxic and it is proven

to be compatible with body tissue and blood. (Yang

1998)

2.2 Parylene Coating Process

The parylene coating process can be divided into

three stages. The first stage is vaporization, the

second is pyrolysis, and the third stage is deposit. In

the beginning of the coating process, the raw

material, dimer that is white powder, is vaporized

under vacuum (1.0 mbar) and heated to a dimeric

gas at approximately 150 ˚C. During the second

stage, pyrolysis, the gas is pyrolized to cleave the

dimer to its monomeric form under vacuum (0.5

mbar) to approximately 680 ˚C. In deposition stage

the monomer reaches the room temperature

deposition chamber. The monomer gas

simultaneously absorbs and polymerizes on the

substrate as a transparent parylene film. The

substrate temperature never rises more than couple

of degrees above the room temperature. (Specialty

Coating System 2007; Pang et al. 2005, p. 4)

The surface area of substrates in deposition chamber

can vary a lot. The substrates to be coated are

positioned in a stand that spins in a vacuum

chamber. When a small amount of dimer is used, it

does not matter where the substrates are located in a

stand. Also the stand area might vary a lot. The

stand might for example have several levels and the

grid on each level might be tight. The coating

thickness is mainly a function of substrate surface

area in chamber and amount of dimer charge.

Program parameters have also minor importance.

Even 1 μm thick coating is discovered to be tight

enough for implants (Stieglitz et al. 2002). Thus it is

very important to be able to measure the thickness of

the coating accurately. The process is controlled by

the deposition process parameters. The process

parameters for our measurements are found from

known coating process recipes that are used also in

Para Tech Coating, Inc. in Sweden (Para Tech

Coating, Inc. 2006). Recipes determine the amount

of dimer, process temperatures and times, and

approximate final parylene thickness. It has been

proven that when different recipes are used for same

amount of dimer, the final thickness might be

different. Therefore, in addition to the amount of the

dimer, also the process parameters affect the final

thickness of parylene.

After the coating process, the achieved thickness

could be measured by releasing a sample parylene

film from the top of a preparat glass that has been in

vacuum chamber, and then measuring the thickness

of film. This does not give very accurate results,

since the thin film is charged electrically and it

might be creased. Also, after releasign the film from

the top of a preparat glass, the film surface might

already contain some impurities from the air that

affect the result. Moreover, as the parylene thickness

can depend on the location in the camber, the

thickness of the film on the preparate glass can be

METHOD FOR MEASURING PARYLENE THICKNESS USING QUARTZ CRYSTAL MICROBALANCE

223

different from the thickness of the sample coating.

Since the final thickness is hard to predict before the

coating process or to measure accurately after the

process, a real-time thickness monitoring system

would be useful in many cases.

2.3 Quartz Crystal

A crystal oscillator is an electronic circuit that

creates an electric signal with a certain frequency by

using the mechanical resonance of a vibrating quartz

crystal. The crystal is made of piezoelectric material

and it is placed between a pair of electrodes. When

these two electrodes are connected to an alternating

electric field, the quartz crystal starts to oscillate at

its resonance frequency due to the piezoelectric

effect.

A quartz crystal microbalance (QCM)

measurement technique is based on the oscillation at

a precise frequency. When any type of mass is added

on the surface of the crystal, the resonance

frequency of the quartz crystal decreases. According

to the Sauerbrey equation,

qq qq

Av v

ρρ

=−Δ = Δ = Δ−

(1)

the change in mass, Δm, is proportional to the

change in frequency, Δf. Here f

is the initial

resonant frequency of the crystal, f the resonant

frequency of the coated crystal, A is the effective

area of the crystal (between electrodes), ρ

is the

density of quartz, and v

is the shear velocity in

quartz. The change of mass is written in terms of

parylene thickness, Δx and density, ρ

. When very

accurate measurements of very small mass changes

need to be performed, the QCM technology is very

appealing and it has already been studied as a system

for measuring film thickness during deposition of

different materials. (Eggins 2002; Gulati, Auras, &

Rubino 2006) In addition to rigid deposits, the QCM

has been widely used for its respond to changes in a

liquid's viscoelastic properties. Some targets of using

QCM have been for example humidity sensor (Ito et

al. 2003), bacterial spores detector (Lee et al. 2005),

and proteins detector.

3 MEASUREMENTS

3.1 Preliminary Study

The starting point for the real-time thickness

measurements method was the conclusion that if

parylene penetrates inside the crystal oscillator and

covers the quartz crystal (Fig. 1), the mass of the

quartz crystal must increase. Based on the QCM

technology, the mass change of the crystal should be

directly proportional to its frequency change.

Moreover, since the mass change is proportional to

crystal area and parylene thickness, the frequency

change is proportional to parylene thickness, as

given in eq. (1). By increasing the parylene

thickness, the frequency of crystal oscillators placed

in vacuum chamber should linearly decrease.

Figure 1: A quartz crystal covered with parylene.

In order to enable parylene penetration to the surface

of the quartz crystal, small holes must be drilled to

the metal case of the crystal oscillator. As a

preliminary study for the measurements, the number

of holes on the sides of the oscillator required to

achieve the largest frequency change, was

determined. It seemed that three 1 mm holes

generated larger resonance frequency change than

one or two holes, even if all of the crystal oscillators

were in the same run of parylene equipment in

vacuum chamber. At the same time making four

holes on both sides gave the same result as three

holes so there was not need to increase the hole

number to more than three. Thus it seems that full

covering of the quartz resonator was not obtained

with only one or two holes in the resonator case.

These results bear evidence that parylene can not

penetrate through all pinholes easily.

3.2 Measurement Set-up

For the measurements, 21 crystal oscillators were

used. The resonance frequency of these crystal

oscillators was 3.579 MHz. Three 1 mm holes were

drilled on both sides of the crystals' metal case. Each

crystal was numbered with consecutive numbers.

After drilling the holes, the resonance frequency of

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices

224

each oscillator was measured with a network

analyzer.

The measurement arrangement was carried out in

the following order. After measuring the initial

frequency, all of the crystals in metal case were

coated with parylene. Model 3000 Labtop, Parylene

Deposition System, Para Tech Coating (Aliso Viejo,

USA) equipment was used for coating. The

resonators were coated using 1.7 g of dimer and

process parameters for 2 μm coating. Therefore the

estimated thickness of the parylene film was 1.7–2.0

μm. The same procedure was repeated 11 times for

all crystals. Each time there was an estimated 1.7–

2.0 μm growth on the preceeding coating. The

frequency of each crystal was measured with

network analyzer after each coating run, and results

were documented. The frequency decreased due to

the mass increment on the quartz crystals.

To be able to analyse the results, an accurate film

thickness after each parylene coating must be

known. These measurements were done visually, by

using microscope Olympus BX60M. For reference

data, silicon samples were coated in the same

processes with the crystal resonators. Eleven silicon

chips (approximately 1.0 cm x 1.0 cm), in addition

to crystals, were placed into the vacuum chamber in

the beginning of the first coating. One silicon chip

was taken away from the chamber after every

coating and marked with consecutive numbering.

Meaning chip having number i should have

approximately i × 1.7–2.0 μm parylene film on the

silicon. To be able to define the accurate thicknesses

of parylene coats, each chip was placed in a mould

and covered with epoxy. The cross-sectional

samples were then mechanically prepared for

microscopic thickness examination by grinding and

polishing.

Figure 2: The frequency change of crystals as a function of

parylene thickness. The vertical lines indicate the average

value ± standard deviation of 21 crystal samples.

4 RESULTS

After 11 coating runs, measurement results proved

that there is a certain interrelation between the

resonance frequency and the thickness of parylene.

The measurement results are illustrated in Fig. 2.

The frequency change of crystals increases linearly

as parylene thickness increases. The slope of this

line is 8.0 kHz/μm.

The coefficient appearing in the Sauerbrey

equation, (1), is 3.64 kHz/μm. As compared to this

value, the frequency change in Fig. 2 is quite a lot

larger. There are several possibilities that can

explain this discrepancy.

On each coating run, parylene was added on the

last coat of parylene. It is possible that these layers

or the interfaces had collected some impurity that

affected the results; though impurities were not seen

in visual microscope examination. Another source of

discrepancy might be in the coating process

parameters. When different recipes for same amount

of dimer are used, the final thicknesses are unequal.

Therefore, it seems that the process parameters

affect the final density of parylene, and/or the

parylene coats different locations in the chamber

with different thicknesses. Furthermore, the parylene

density after each coating is not measured, and

hence it is unreliable to use the literature value for

parylene density in Sauerbrey equation. Finally,

when the crystal is coated, parylene covers the entire

quartz crystal area including the sides of the crystal

and the electrodes connected to the quartz crystal.

Hence it might prevent the vibration of the quartz

crystal and measurement results in frequency change

are not equal to frequencies measured with

Sauerbrey equation.

In the future, these measurement results are

usable basis for developing the parylene thickness

measurement set-up, though additional experimental

results will be needed for accurate reference data.

Idea is to control the deposition thickness of

parylene during the deposition process by

monitoring the resonance frequency change of

crystal oscillator in vacuum chamber. The real-time

measurement set-up would be composed of the

crystal oscillator placed in vacuum chamber, the

network analyser, and the measuring cables for

generating the connection between the crystal

oscillator's electrodes and the network analyser. The

possibility to produce sealed hole, for example to the

observation window of vacuum chamber to allow

measuring cables to pass through, should be studied.

In real-time measurement, the network analyser

measures the resonance frequency change and as the

METHOD FOR MEASURING PARYLENE THICKNESS USING QUARTZ CRYSTAL MICROBALANCE

225

earlier defined frequency change for target parylene

thickness is reached, the coating run could be cut

off.

5 SUMMARY

We have presented a method to measure the

thickness of parylene coating, especially for medical

electronic devices. The method is based on

frequency change of coated quartz crystals. We

proved that the frequency change is proportional to

the parylene thickness, and determined the factor

relating the thickness and frequency change. The

applicability of this factor to different parylene

coating processes was discussed. The method is

applicable also for real-time measurements enabling

the measurement of parylene thickness during the

growth process. In real-time measurements, the

growth process could be stopped after the target

thickness has been reached.

REFERENCES

Eggins, B. R. 2002, 'Chemical Sensors and Biosensors',

John Wiley & Sons. 291 p.

Gulati, N., Auras, R., Rubino, M. 2006, 'Determination of

barrier properties of poly(lactide) polymers using a

quartz crystal microbalance', ANTEC 2006 Platics,

North Carolina, 2006, p. 1530–1534.

Ito, H., Kakuma, S., Ohba, R., and Noda, K., 2003,

'Development of a Humidity Sensor using Quartz

Crystal Microbalance', SICE Annual Conference,

Fukui, Japan, August 4.-6., 2003, pp. 1175–1178.

Licari J., 2003, Coating Materials for Electronic

Applications: polymers, process, reability, testing,

Noyes/William Andrew, New York, pp. 154–168

Noordegraaf J. and Hull, H. 1997, 'C-shield parylene

allows major weight saving for EM shielding of

microelectronics', 1st IEEE International Symposium

on Polymeric Electronics Packaging, Norrkoping,

Sweden, October 26.-30., 1997, pp. 189–196.

Pang, C., Cham, J. G., Nenadic, Z., Musallam, S., Tai, Y-

C., Burdick, J. W., and Andersen, R. A. 2005, 'A New

Multi-Site Probe Array with Monolithically

Intergrated Parylene Flexible Caple for Neural

Prostheses', 27

IEEE Annual Conference on

Engineering in Medicine and Biology, Shanghai, 2005, 4p.

Para Tech Coating, Inc. 2006. Retrieved September 7,

2007, from http://www.parylene.com/index.html.

Ratner, B. D., Hoffman, A. S., Schoen, F. J., and Lemons,

J. E., 1996, 'Biomaterials science: an introduction to

materials in medicine', Academic press, San Diego,

484 p.

Lee, S.-H., Stubbs, D. D., Cairney, J., and Hunt, W., D,

2005, 'Rapid Detectioon of Bacterial Spores Using a

Quartz Crystal Microbalance (QCM) Immunoassay',

IEEE Sensors Journal, vol. 5, no. 4, pp. 737–743.

Specialty Coating System 2007, Retrieved March 27,

2007, from http://www.scscoatings.com/.

Stieglitz, T., Kammer S., Koch K. P., Wien S., Robitzki A.

2002, 'Encapsulation of Flexible Biomedical

Microimplants with Parylene C', International

Functional Electrical Stimulation Society

(IFESS):231-3.

Wolgemuth, L. 2002, 'The Surface Modification

Properties of Parylene for Medical Applications',

Medical Device Manufacturing & Technology, pp. 1–4.

Yang, G. R., Ganguli, S., Karcz, J., Gill, W. N., and Lu T.

M. 1998, 'High deposition rate parylene films'. Journal

of crystal growth, vol. 183, pp. 385–390.

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices

226