Micromachined Interfaces for Medical and Biochemical ...9144/FULLTEXT01.pdf · Micromachined...

Micromachined Interfaces for Medical and

Biochemical Applications

Patrick Griss

Microsystem Technology

Department of Signals, Sensors and Systems (S3)

Royal Institute of Technology (KTH)

TRTA-ILA-0201

ISSN 0281-2878

Submitted to the School of

Computer Science – Electrical Engineering – Engineering Physics (DEF),

Royal Institute of Technology (KTH), Stockholm, Sweden,

in partial fulfillment of the requirements for the degree of Doctor of Philosophy.

Stockholm 2002

Micromachined Interfaces for Medical and Biochemical Applications

ii

The front cover shows a micromachined array of sharp spikes manufactured by the author.

The spike array is used for the measurement of biopotentials in vivo, in particular for the

monitoring of the activity of the brain. The spikes reduce the high impedance characteristic

of the human outermost skin by painlessly penetrating into deeper layers.

Copyright 2002 by Patrick Griss

Printed by KTH Högskoletryckeriet, Stockholm 2002

Micromachined Interfaces for Medical and Biochemical Interfaces

iii

Abstract

The possibility to miniaturize medical devices by using micromachining techniques offers new

opportunities for maintaining health and curing of diseases. The main advantage of micromachined

devices is that they are minimally invasive when used during surgical, diagnostic or therapeutic

treatments. Trauma can be reduced which facilitates the healing process, increases comfort for the

patient, and reduces recovery time. Thus, the efficiency of medical treatments can be increased.

Microfluidic systems enable the miniaturization of analytical techniques in biochemistry. They

offer the possibility for chemical compounds occuring in organisms to be studied in parallel and in an

automated way at large scale. In addition, sensitivity and speed of an analysis can be considerably

increased. Microfluidic devices have thus become very important in the search for new therapies and

as diagnostic tools.

The purpose of this thesis is to explore micromachined devices that act as interfaces in medical and

biochemical applications. A central concept in all described interfaces is the use of tapered

projections in the micro scale (also referred to as microspikes, microneedles or microtips). The

fabrication of the structures is based on deep reactive ion etching technology that was adapted to each

individual application.

The first described interface is a skin applied electrode for electroencephalographic recording, i.e.

the measurement of the electrical activity of the brain. The electrode is a device, which electrically

interconnects the human body to measurement amplifiers.

The second described device deals with microfluidic transdermal interfacing, i.e. the transfer of

liquid through the outer skin layers. This includes the controlled injection of drugs into the skin.

The third discussed device is an electrospray ionization emitter tip, which allows the formation of

ions of unknown molecules, e.g. proteins or peptides. Once the molecules are ionized, they can be

introduced to a mass spectrometer for the detection of their mass. Thus, the emitter tip interfaces a

liquid sample solution to the mass spectrometer.


iv

"To fly as fast as thought, to anywhere that is," he said, "you must begin by knowing that you have

already arrived..."

(from Jonathan Livingston Seagull by Richard Bach)


v

Contents

1 List of Publications 1

2 General Introduction 52.1 Medical and Biochemical Microsystems 52.2 Outline of the Thesis 6

3 Technology Basis 73.1 Tapered Projections in the Micrometer Scale 73.2 Deep Reactive Ion Etching Bosch Process 93.3 Combination of Isotropic and Anisotropic Etching 1 0

4 Skin Anatomy and Physiology 1 34.1 Epidermis 1 34.2 Dermis 1 4

5 Micromachined EEG Electrode 1 55.1 Anesthesia in Surgical Procedures 1 55.2 The Origin of Biopotentials 1 65.3 Spiked EEG Electrode 1 7

6 Microfluidic Transdermal Liquid Delivery System 2 36.1 Microfluidic Transdermal Interface 2 46.2 Dosing & Actuation Unit 2 8

7 Micromachined Electrospray Emitter Tips for MassSpectrometry 3 37.1 Electrospray Ionization 3 37.2 Proteomics 3 47.3 Micromachined Electrospray Emitter Tips 3 5

8 Final Conclusions 4 1

9 Acknowledgements 4 3

1 0 Summary of the Appended Papers 4 5

1 1 Glossary 4 9

1 2 References 5 1

1 3 Paper Reprints 5 7

1

1 List of Publications

This thesis is based on following reviewed journal papers:

1 Micromachined Electrodes for Biopotential Measurements

P. Griss, P. Enoksson, H. K. Tolvanen-Laakso, P. Meriläinen, S. Ollmar, and G. Stemme,

Journal of Microelectromechanical Systems, vol. 10, pp. 10-16, 2001.

2 Micromachined Barbed Spikes for Mechanical Chip Attachment

P. Griss, P. Enoksson, and G. Stemme,

Sensors & Actuators: A Physical, vol. 95, pp. 94-99, 2002.

3 Expandable Microspheres for the Handling of Liquids

P. Griss, H. Andersson, and G. Stemme,

Lab-on-Chip, vol. 2, 2002.

4 Characterization of Micromachined Spiked Biopotential Electrodes

P. Griss, H. Tolvanen-Laakso, P. Meriläinen, and G. Stemme,

IEEE Transactions on Biomedical Engineering, Accepted for publication, 2002.

5 Side-Opened Out-of-Plane Microneedles for Microfluidic Transdermal Liquid Transfer

P. Griss and G. Stemme,

Submitted, 2002.

6 Development of Micromachined Hollow Tips for Protein Analysis Based on

Nanoelectrospray Ionization Mass Spectrometry

P. Griss, J. Melin, J. Sjödahl, J. Roeraade, and G. Stemme,

Submitted, 2002.

The contributions of the author to these papers are:

1 – 4 All fabrication, part of experiments, major part of writing.

5 All fabrication, all experiments, major part of writing.

6 Part of fabrication, part of experiments, major part of writing.


2

The work was also presented at following conferences:

1 Spiked Biopotential Electrodes

P. Griss, P. Enoksson, H. Tolvanen-Laakso, P. Meriläinen, S. Ollmar, and G. Stemme,

presented at The 13th Annual International Conference on Micro Electro Mechanical Systems,

Miyazaki, Japan, 2000.

2 Barbed Spike Arrays for Mechanical Chip Attachment

P. Griss, P. Enoksson, and G. Stemme,


Interlaken, Switzerland, 2001.

3 Novel, Side Opened Out-Of-Plane Microneedles for Microfluidic Transdermal Interfacing

P. Griss and G. Stemme,


Las Vegas, USA, 2002.

4 Liquid Handling Using Expandable Microspheres

P. Griss, H. Andersson, and G. Stemme,


Las Vegas, USA, 2002.

5 Characterization of Micromachined Hollow Tips for 2D Nanoelectrospray

Massspectrometry

J. Sjödahl, P. Griss, J. Melin, Å. Emmer, J. Roeraade, and G. Stemme,

presented at the 15th International Symposium on Microscale Separations and Analysis,

HPCE 2002, Stockholm, Sweden.


3

In addition, the author has contributed to the following related journal papers:

1 Expandable Microspheres - Surface Immobilization Techniques

H. Andersson, P. Griss, and G. Stemme,

Sensors and Actuators B: Chemical, Accepted for publication, 2002.

2 Hydrophobic Valves of Plasma Deposited Octafluorocyclobutane in DRIE Channels

H. Andersson, W. van der Wijngaart, P. Griss, F. Niklaus, and G. Stemme,

Sensors and Actuators B: Chemical, vol. 75, pp. 136-141, 2001.

3 Low-Temperature Wafer-Level Transfer Bonding

F. Niklaus, P. Enoksson, P. Griss, E. Kälvesten, and G. Stemme,

Journal of Microelectromechanical Systems, vol. 10, pp. 525-531 2001.

4 A Variable Optical Attenuator Based on Silicon Micromechanics

C. Marxer, P. Griss, and N. F. de Rooij,

IEEE Photonics Technology Letters, vol. 11, p.233-235, 1999.

5

2 General Introduction

Microelectronics has allowed people to process information at unprecedented speed, at an

affordable cost and with computers that are portable. Micromachining methods that allow the

realization of the basic elements of microelectronics on a scale, or size, of micrometers are the keys

for the introduction of processors and memories in our daily lives.

The use of micromachining methods to fabricate microscaled sensors and actuators instead of

integrated circuits some 30 years ago has now developed into the area of microsystem technology.

Microsystems and –devices can process mechanical, fluidic, electrical, optical, thermal and radiative

signals, often integrated, on one single chip [1]. This area is about to complete the penetration of

microelectronics into our society and further enables people to interact with their environment and

their own physiology in a way that was not possible before.

2.1 Medical and Biochemical Microsystems

The possibility to miniaturize medical devices by using micromachining technology offers a whole

range of opportunities for the maintenance of health and the cure of diseases. The main advantage of

micromachined devices is that they are minimally invasive when used during surgical, diagnostic or

therapeutic treatments. Trauma can be reduced which facilitates the healing process, increases

comfort for the patient, and reduces recovery time. Thus, the efficiency of medical treatment can be

increased. Miniaturization also allows light and portable devices to be fabricated, an important issue

for diagnostic systems. This is in line with modern healthcare’s demand for comfortable, cheap, home

treatment and point-of-care diagnosis.

Microfluidic systems enable the miniaturization of analytical techniques in biochemistry and are

often referred to as micro total analysis systems or Lab-on-Chip. They offer the possibility for

parallelization and automation of the study of chemical compounds occurring in organisms at large

scale. In addition, sensitivity and speed of an analysis can be considerably increased. Microfluidic

devices have thus become very important as diagnostic tools and in the search for new therapies.


6

2.2 Outline of the Thesis

The purpose of this thesis is to explore micromachined interfaces for medical or biochemical

applications. A central concept is the use of tapered projections in the micro scale (also referred to as

microspikes, microneedles or microtips). The fabrication of these structures is based on deep reactive

ion etch technology that was adapted to each individual application. The fundamentals of the chosen

fabrication technology are described in chapter 3.

The first described interface is a skin applied electrode for electroencephalographic recording, i.e.

the measurement of the electrical activity of the brain. The electrode is a device, which electrically

interconnects the human body to measurement amplifiers. The work around this electrode is the

foundation of this thesis and was performed in collaboration with Datex-Ohmeda. Datex-Ohmeda is

the largest division of Instrumentarium Corporation, one of the world’s leading medical technology

companies. Datex-Ohmeda represents the group's core business, anesthesia and critical care. Datex-

Ohmeda is headquartered in Helsinki, Finland, and has manufacturing operations in Finland, Sweden

and the United States [2, 3].

The second described device deals with microfluidic transdermal interfacing, i.e. the transfer of

liquid through the outer skin layers. This includes the controlled injection of drugs into the skin.

The third discussed device is an electrospray ionization emitter tip, which allows the formation of

ions of unknown molecules, e.g. proteins or peptides. Once the molecules are ionized, they can be

introduced to a mass spectrometer for the detection of their mass. Thus, the emitter tip interfaces a

liquid sample solution to the mass spectrometer.

Both the EEG electrode and the microfluidic transdermal interface deal with the electrical,

chemical and mechanical properties of the skin. Skin anatomy and physiology are briefly described in

chapter 4.

7

3 Technology Basis

The devices explored in this thesis rely on tapered projections in the micrometer scale. Tapered

microprojections are often referred to as microneedles, microspikes or microtips. They can be hollow

or solid depending on their function.

Tapered projections have been used throughout history to solve various engineering problems, and

since the invention of the gramophone, their use is in the sub-millimeter scale. In this chapter an

overview on the different uses of micromachined projections in scientific and commercial applications

is given. Subsequently, the fundamentals of the fabrication technology employed to manufacture the

devices of this thesis are described.

3.1 Tapered Projections in the Micrometer Scale

Recently, activities in micromachining microprojections for various applications have increased.

Two fundamentally different fabrication principles have been used as illustrated in Figure 3.1:

(i) out-of-plane fabrication, where the resulting projection is protruding perpendicularly from

the plane of the substrate material (e.g. a silicon wafer), or

(ii) in-plane fabrication, where the resulting projection is “lying” in the wafer plane.

The choice between these two methods depends strongly on the final application of the device. If it

is desired to have a two-dimensional (2D) array of microprojections where both extensions in the x

and y directions contain several (even hundreds) of single projections, it is probably more efficient to

use the out-of-plane manufacturing method. This is because the array is formed at the same time as

the needles. If “long” projections are required, for example in the millimeter range, in-plane

fabrication is preferable. In this case 2D-array building requires an additional fabrication step, which

is not on wafer level and therefore time consuming. The advantage of in-plane needles is that there is

the possibility to manufacture electrical, microfluidic, or optical functionality on the needle shaft.

A literature study reveals that there are many areas where individual or arrays of tapered

microprojections are of interest. Table 3.1 is a summary on the use of microneedles, microtips and

microspikes for scientific and commercial applications. The microprojections are classified

corresponding to the physical nature of their interaction with the environment, the type of interaction,

and the purpose of their use.


8

OP

TIC

AL

Sub

wav

elen

gth

emis

sion

Supe

r re

solu

tion

mic

rosc

opy

SNO

M3 [

4]

com

bine

d

SNO

M3 a

nd

AFM

1 [5]

-

Ele

ctri

c fi

eld

conc

entr

atio

n

Par

ticl

e ge

nera

tion

elec

tros

pray

em

itter

nozz

les

[6, P

aper

6]

prin

t-he

ad [

7]

fiel

d em

itter

s [8

, 9]

X-R

ay s

ourc

e [1

0]

elec

tros

pray

em

itter

need

les

[11,

12]

EL

EC

TR

ICA

L

Tun

nelin

g

curr

ent

Surf

ace

prof

iling

STM

2 [13

]

-

Surf

ace

cont

acti

ng

Surf

ace

prof

iling

AFM

1 [14

, 15,

16]

com

bine

d A

FM1

and

SNO

M3 [

5]

-

Tra

nsfe

r of

mat

ter

tran

sder

mal

liqui

d tr

ansf

er[1

7, 1

8, 1

9,Pa

per

5]

DN

A tr

ansf

erin

to ti

ssue

[20

,21

]

tran

sder

mal

liqui

d tr

ansf

er[2

3,24

,25,

26, 2

7]

mic

rodi

alys

is[2

8]

Tra

nsfe

r of

ener

gy

stim

ulat

ion/

re-

cord

ing

from

visu

al c

orte

x [2

9]

EE

G r

ecor

ding

[Pap

er 1

, Pap

er 4

]

deliv

ery

of c

hem

ical

s an

d re

cord

ing

ofel

ectr

ical

res

pons

e [2

2]

elec

tric

alre

cord

ing

in th

eC

NS4 [

30]

ther

mal

mar

king

and

mon

itori

ng o

fne

ural

tiss

ue [

31]

coch

lear

impl

ants

[32]

ME

CH

AN

ICA

L

Surf

ace

pene

trat

ion

Surf

ace

mod

ific

atio

n

“Mill

iped

e”da

tast

orag

e[33

]

-

Out

-of-

plan

e

fabr

icat

ion

In-p

lane

fabr

icat

ion

1 Ato

mic

For

ce M

icro

scop

e 2 Sc

anni

ng T

unne

ling

Mic

rosc

ope

3 Sc

anni

ng N

ear-

Fiel

d O

ptic

al M

icro

scop

y

4 Cen

tral

Ner

vous

Sys

tem

TA

BL

E 3

.1:

OV

ER

VIE

W O

F T

HE

USE

OF

MIC

RO

MA

CH

INE

D T

APE

RE

D P

RO

JEC

TIO

NS

IN S

CIE

NT

IFIC

OR

CO

MM

ER

CIA

L A

PPL

ICA

TIO

NS.

TH

E C

ITE

D E

XA

MPL

ES

AR

E C

LA

SSIF

IED

AC

CO

RD

ING

TO

TH

E P

HY

SIC

AL

NA

TU

RE

OF

TH

E I

NT

ER

AC

TIO

N W

ITH

TH

E E

NV

IRO

NM

EN

T, T

HE

TY

PE O

F IN

TE

RA

CT

ION

AN

D T

HE

PU

RPO

SE O

F T

HE

IR U

SE.

FO

R E

AC

H E

XA

MPL

E, A

T L

EA

ST

ON

E R

EFE

RE

NC

E I

S A

NN

OT

AT

ED

.

Phy

sica

l nat

ure

of th

e

inte

ract

ion

Typ

e of

Int

erac

tion

Pur

pose

of

use

App

licat

ion

exam

ples

Technology Basis

9

3.2 Deep Reactive Ion Etching Bosch Process

Design and function of microdevices is strongly related to their fabrication process. There is a vast

variety of microfabrication technologies. The most commonly used are derived from the fabrication

technologies developed to fabricate integrated circuits [1, 34]; the most powerful of these is

photolithography. However, there are complementary and alternative methods such as polymer

replication [35], soft lithography [36], polymer lamination [37], and high-aspect ratio photosensitive

resins [38]. Therefore, microdevices need not be made from only silicon. Yet, silicon has some

advantages compared to glass or polymers, for example. Aside from the possibility to integrate

circuitry directly onto the transducer’s substrate, the excellent mechanical and electrical properties are

often very important for the performance of the microdevice.

Masking Material

Substrate Material(e.g. Silicon)

Out-of-Plane In-Plane

z

x y

Figure 3.1: Schematic representation of the fabrication principles of micromachined tapered projections.

The development of anisotropic etch tools for silicon that allow etching of deep holes or high free

standing structures are of considerable interest for microsystem technology and in particular for this

thesis [39, 40, 41]. The anisotropic etching of high-aspect ratio structures in silicon is often referred to

as deep reactive ion etching (DRIE).

Commercially available anisotropic etchers include inductively coupled plasma (ICP) etchers.

They rely on a high-density plasma source and an alternating process of etching and protective

polymer deposition to achieve aspect ratios up to 30:1 [39]. This approach utilizes an etching cycle

using only SF6 gas and then switches to a passivation cycle using only C4F8, as schematically

represented in Figure 3.2. During the etching cycle, the passivating film is preferably removed from

the bottom of the trenches due to directional ion bombardment. The final result is an anisotropic etch

profile that exhibits scalloping. As it becomes clear from the drawings, the actual anisotropic etch

consists of a large number of isotropic etches. Controlling plasma parameters as well as gas

pressure/flow parameters, the etch characteristics and profile can be controlled [42].


10

C F84

SF6

SF6

SF6

Silicon

Silicon dioxide

C F84

Etch

Passivation

Passivation

Etch

Et Cetera ...

Figure 3.2: Everything You Always Wanted to Know About Bosch’s Process (But Never Dared to Ask).

By suppressing switching of etch and passivation gases, the equipment can be run with continuous

flow of SF6 or C4F8. Thus, using only SF6 it is possible to etch isotropically and using only C4F8 it is

possible to deposit Teflon like layers on three-dimensional structures. All structures presented in this

thesis were micromachined using a DRIE capable ICP from Surface Technology Systems.

3.3 Combination of Isotropic and Anisotropic Etching

The combination of isotropic and anisotropic etch steps is a fundamental part of the fabrication of

the devices presented in this thesis. It was observed that if a silicon dioxide mask is underetched and

subsequently anisotropically etched, the resulting cross-section of the structure corresponds to the

mask (Figure 3.3 a)-b)). It was further observed that vertical walls of DRIE etched high aspect ratio

silicon structures stay vertical during an isotropic plasma etch (Figure 3.3 b)-c)). Based on these

observations, a variety of tapered out-of-plane projections were manufactured for different

applications.

Technology Basis

11

Depending on the application, it is possible to shape the geometry of the projections. Overall

dimensions, sharpness of the tip, “smoothness” of the shape, and mechanical stability can be

controlled by controlling the length of the various etch steps, as summarized in Table 3.2.

Isotropic etch 1 Anisotropic etch 1 Isotropic etch 2

D

sβ

H

h

d

Anisotropic etch 2

Silicon

Silicon dioxide

a) b) c) d)

Figure 3.3: Combination of isotropic and anisotropic etch steps to achieve out-of-plane, tapered, silicon

projections. The shown dimensions in d) are controlled via process parameters. Table 3.2 is a summary of how

processing time influences the represented geometrical dimensions.

TABLE 3.2:

PROVIDED THAT THE ICP SETTINGS ARE SUCH THAT ANISOTROPIC ETCH RESULTS IN VERTICAL PROFILES, THE

LENGTH OF THE DIFFERENT ETCH STEPS INFLUENCE THE SHAPE AND DIMENSIONS OF THE RESULTING MICROSPIKE.

ISOTROPIC 1 AND ISOTROPIC 2 ARE THE ETCH TIMES OF THE FIRST AND SECOND ISOTROPIC ETCH, RESPECTIVELY.

ANISOTROPIC 1 AND ANISOTROPIC 2 ARE THE ETCH TIMES OF THE FIRST AND SECOND ANISOTROPIC,

RESPECTIVELY.

Etch timeGeometrical dimension

(refer to Figure 3.3) Isotropic 1 Anisotropic 1 Isotropic 2 Anisotropic 2

s ↑ ↓↓↓↓ N/I ↓↓↓↓ N/I

D ↑ N/I N/I ↓↓↓↓ N/I

d N/I N/I N/I N/I

H ↑ ↑↑↑↑ ↑↑↑↑ ↑↑↑↑ N/I

h ↑ ↑↑↑↑ ↑↑↑↑ ↑↑↑↑ ↑↑↑↑

β ↑ ↓↓↓↓ N/I ↑↑↑↑ N/I

↑... increasing value N/I...no influence ↓… decreasing value

13

4 Skin Anatomy and Physiology

In this chapter basic anatomy and physiology of the human skin is explained. Understanding of the

skin is of importance for two of the described devices, i.e. the spiked electrode and the microfluidic

transdermal interface.

The seemingly inert appearance of the skin hides its multifunctional properties. It contains the

body’s fluids and tissues and is a protective layer against physical, chemical and biological attacks. It

is a receptor of tactile, painful and thermal stimuli. It regulates heat loss from the body and plays an

important role in the control of immunity via dendritic cells [43, 44]. The skin has a layered structure

and can be subdivided into the Dermis and the Epidermis. Underneath the Dermis is the subcutaneous

tissue, which mainly consists of fat. Within the Dermis are varying numbers of hair follicles and

sweat glands.

4.1 Epidermis

The Epidermis can be divided into three important layers: the outer most Stratum Corneum (also

called the horny layer) composed of fibrous keratinized dead cells; a layer of Living Epidermis; and

the basal layer or Stratum Germinativum, as schematically represented in Figure 4.1.

The basal layer gives rise to living epidermal cells by division of the cells within the Stratum

Germinativum. The epidermal cells possess a high metabolic rate and gradually move upward to

transform into keratinized cells, which are flat, compact and dehydrated. The Epidermis is constantly

renewing itself approximately once every 30 days.

Located within the Stratum Germinativum and the Living Epidermis are dendritic cells, e.g.

melanocytes and Langerhans cells. They are the initiators and modulators of the immune response

[44].

The Stratum Corneum imparts the major barrier function to skin, which prevents loss of water and

invasion by microorganisms as well as a large number of chemical agents. At the same time the

Stratum Corneum provides a flexible, tough mechanical protection. Because of its dehydration it is

electrically isolating. There are no blood vessels in the Epidermis; nutrition diffuses outwards from

the rich blood supply of the Dermis, and there are no sensory organs [43, 45].


14

The Stratum Corneum is about 10-15 µm thick on the forearm for example, while it is twice as

thick on the back and hundreds of microns thick under the feet. The Living Epidermis is around

150 µm thick.

4.2 Dermis

Figure 4.1 shows that the Dermis is much thicker than the Epidermis and mainly consists of

collagen, which builds a supportive connective tissue matrix. It supplies the Epidermis with vital

nutrients and oxygen via a network of blood capillaries. It is the connective tissue of the Dermis that,

when properly prepared from animals by tanning, becomes the tough leather that has many uses [45].

The Dermis contains many sensory organs for touch, pressure, pain and temperature and is richly

supplied with specialized nerve endings, nerve fibers and pressure sensors.

The dermal thickness varies between 600 µm on the eyelids to several millimeters on the back,

palms, and soles.

Stratum Corneum

Langerhans Cell

Stratum Germinativum

Blood Capillaries

Blood Capillaries

Hair

Dermis

SubcutaneousTissue

Epidermis

Sweat Gland

Sensory Organs andNerve Endings

Melanocyte

Living Epidermis

Figure 4.1: Layered architecture of the skin along with organs embedded in the skin. Sharp projections that

extend into the Stratum Germinativum and not further can be expected to be painfree since there are no sensory

organs or nerve endings that could trigger any sensation of pain. Also, such projections cannot cause bleeding

since blood capillaries cannot be reached.

15

5 Micromachined EEG Electrode

In neurological and cardiological applications, the classical methods of recording biopotentials

using skin applied electrodes are well established. Biopotentials are produced by the electrochemical

activity of excitable cells, e.g. nerve cells in the brain. An excited nerve cell creates an electrical

current of ions (e.g. Cl-) which induces a measurable potential on the surface of the body, i.e. the skin

[46].

Examples of biopotential recordings include monitoring of heart activity (ECG) and diagnosis of

epilepsy via electroencephalography (EEG). Among these methods EEG is the most difficult to

record since signal levels are very low [47].

Towards the end of the 20th century, sophisticated processing of EEG data emerged. Researchers

started to better understand how to efficiently extract information from the very complex and fractal

EEG signals [48]. This information can be used in patient monitoring via EEG. One example is

determining the level of sedation of patients during surgery.

In this chapter a background on anesthesia monitoring is given. Subsequently, the design,

fabrication and testing of micromachined electrodes measuring EEG and acting as interfaces between

the patient and the monitoring equipment is shown.

5.1 Anesthesia in Surgical Procedures

Most surgical procedures cause pain and traumatic experiences and are stressful for the patient. To

spare the patient from pain and stress, anesthetists and monitoring equipment are needed to protect the

patient. This entails removing the sensation of pain by administering anesthetic agents and pain

medication in optimal combinations, often also producing unconsciousness. Anesthetists also give

medication to achieve a suitable degree of muscle relaxation to ensure optimal surgical conditions.

Often, the above-mentioned delivery of anesthetic agents disables spontaneous breathing. A ventilator

therefore maintains the transport of breathing gases to and from ambient air.

Until now, anesthetists have had no direct means of assessing a patient’s level of consciousness.

They have generally relied on recommended dosages and on indirect indicators of consciousness as

blood pressure or heart rate [2].


16

EEG Electrodes

Figure 5.1: Scene representing the Datex-Ohmeda S/5™ Anesthesia System, which integrates management of

the entire anesthesia process from pre-operative assessment to post-operative analysis. EEG electrodes are the

sensor elements allowing brain activity to be monitored for direct measurement of the level of unconsciousness.

The complete S/5™ system can include ventilation and monitoring of parameters like airway gases, cardiac

function (ECG), circulation, lung mechanics, etc. (Source: [2], with permission)

Recent studies by Datex-Ohmeda, Helsinki, Finland, have suggested that parameters corresponding

to the level of complexity of the EEG signal indicate the anesthetic effect, e.g. the level of hypnosis

[48]. This means that measuring EEG of a patient during surgery offers the possibility to directly

assess the effect of anesthetics, i.e. monitor the patient via EEG recordings as shown in Figure 5.1.

Roughly 75 million patients receive general anesthesia every year in main industrialized countries.

Monitoring these people with EEG results in several advantages for patients, hospitals and society:

(i) the risk assessment of surgical awareness (i.e. the unintentional return of consciousness

during surgery) can be improved

(ii) the patient’s quality of recovery is improved

(iii) the patient’s time in the operating room and the post anesthesia care unit is reduced

(iv) the patient wakes up faster

(v) drug use and costs can be reduced.

5.2 The Origin of Biopotentials

Biopotentials are produced by the electrochemical activity of excitable cells that are found in

nervous, muscular and glandular tissue. Active transport of Na+ out of the cell and of K+ into the cell

as well as different permeabilities of Na+, K+ and Cl- result in a negative net charge of the cell interior

and a negative membrane potential, as explained in Figure 5.2 [49]. When excited, for example by

depolarizing the cell membrane with an electrical impulse, this equilibrium is momentarily disturbed.

Micromachined EEG Electrode

17

Due to the voltage dependence of the membrane permeabilities, Na+ rushes into the cell, leaving the

cell further depolarized. This potential is referred to as action potential and is stopped at about

+20 mV since the increase of the Na+ permeability is also time dependent and relatively short [46].

Action potentials propagate along an excitable cell and therefore the cell acts like a signal carrier.

A solenoidal ionic current is induced in the interstitial liquid surrounding the cell, as shown in Figure

5.2. This current causes electrical potential differences in the human body fluids and is measurable on

the skin [47].

The brain is mainly made up of nervous tissue. Therefore, its activity is reflected on the skull as a

changing pattern of potential differences that can be recorded from specific locations on the skull, in

particular on the forehead. Such a recording is referred to as electroencephalograph (EEG).

ActiveRegion

Ionic Current

Direction of signal propagation

- - - - - -- - ++ + ++ + ++ + ++ - - - - -- - - - --

- - - - - -- - ++ + ++ + ++ + ++ - - - - -- - - - --

Net charge ofthe cell interior

Mem

bran

e po

tent

ial [

mV

]

ExciteableCell

Cell membrane

-80

0

+20

Figure 5.2: Signal propagation along the excitable cell.

The measurement of biopotentials on the skin or within tissue in vivo involves an electrode-

electrolyte interface [46, 50]. The measurement is an electrochemical process, which transforms an

ionic current into an electronic current. Only electronic currents can be used in analyzing electronics.

For traditional skin applied electrodes, an electrolytic gel is applied between the metal electrode and

the skin acting. Because of the gel they are referred to as “wet” electrodes.

5.3 Spiked EEG Electrode

Inappropriate electrode technology was one impediment to using EEG in routine patient

monitoring. As the signals to be measured are very low, conventional EEG electrodes require

preparation of the skin (cleansing with alcohol, abrasion of the outer skin layer) and/or the application

of electrolytic gel at the electrode application site. Both skin preparation and electrolytic gel are

intended to reduce the excellent electrical isolating characteristics of the Stratum Corneum (refer to

chapter 4). Often, conventional electrodes are not suited for clinical use because the application of the


18

electrode is time consuming and the signal quality depends on the skill of the nurse. Research with the

goal to eliminate skin preparation has resulted in dry (i.e. without use of electrolytic gel) ceramic

based NASICON electrodes [51], on chip amplified dry electrodes [52, 53], and the wet Zipprep™

electrode [54].

no electrolytic gel

electrolytic gel

Stratum Corneum

Epidermis

Dermis

spikes

wire to measurement electronics

5 - 10 m50 - 100 m

Figure 5.3: Artistic drawing comparing the concept of the spiked electrode (left) to the concept of conventional

electrodes (right).

In collaboration with Datex-Ohmeda Division, Instrumentarium Corp., Helsinki, Finland, a novel

type of EEG electrode was developed. An array of microscaled, sharp spikes penetrates the Epidermis

to circumvent the isolating characteristics of the Stratum Corneum. This novel principle of a skin

applicable biopotential electrode is schematically represented in Figure 5.3 [Paper 1].

5.3.1 Objectives

Low electrode-skin impedance is the main goal. The lower the impedance, the more effectively the

electrical activity of the brain is measured since the charges from the tissue are effectively transferred

into the analyzing electronics. Another aspect of low electrode-skin impedance is the minimization of

noise in EEG recordings since impedance mismatch of different electrodes is reduced [55].

Another demand on the spiked EEG electrode is to avoid any skin preparation and electrolytic gel

to achieve very easy use and fast, uncomplicated application on the skin.

The minimization of noise resulting from electrochemical processes in/on the skin is critical too,

both in time, i.e. fast stabilization after application on the skin, as well as in amplitude.

High patient acceptance can be reached by maximizing the comfort for the patient. The spiked

EEG electrode should therefore be more comfortable than other products, especially wet electrodes.

Last but not least, the electrode must not create a hazard to the patient. This includes physical as well

as chemical factors, e.g. the electrode should be highly resistant to mechanical stress and it should not

be toxic.

Refer to Table 5.1 for a summary of the objectives of the micromachined EEG electrode.

5.3.2 Prototype Devices

DRIE fabrication of silicon was chosen to fabricate spike arrays [Paper 1]. Figure 5.4 shows a

scanning electron microscope (SEM) of a chip containing an array of spikes 160 µm in length. These

types of spikes were pressed into human skin in vitro and Figure 5.5 shows the photograph of a section


19

of the penetration site of a spike. The figure shows that the Stratum Corneum is totally disrupted and

that the spike penetrates into the living Epidermis, slightly further extending into the Dermis.

Experience showed that spikes of up to 200 µm length were painfree and comfortable for the patient

when applied on the forehead. In some cases a somewhat tickling feeling was reported during the

application of the electrode on the forehead, which probably results from the edges of the chip rather

than the spikes itself. Touching the micromachined surface of a spiked electrode feels like touching

fine sandpaper.

TABLE 5.1:

LIST OF THE MAIN OBJECTIVES OF THE MICROMACHINED EEG ELECTRODE IN DECREASING ORDER OF PRIORITY.

1. Minimize skin-electrode impedance caused by the Stratum Corneum

2. Avoid skin preparation

3. Avoid electrolytic gel

4. Minimize electrical noise levels

5. Maximize comfort for the patient (pain free, minimize irritation, minimize tissue trauma)

6. Maximize the mechanical stability of the microneedle (array)

7. Disposability

Figure 5.4: SEM image of silver coated spikes. The

length of the spikes is 160 µm, the diameter is 40 µm.

Note the through hole in the center of the image

which is electrically interconnecting the front to the

back side of the chip.

Figure 5.5: Section of human skin previously

penetrated by microspikes, 160 µm in length in

vitro. The penetration depth is approximately

100 µm and the spike apex extends slightly into the

dermis. The dotted line represents the limit between

Dermis and Epidermis.


20

The developed fabrication process allows the spikes to be modeled. The spikes show a sharp tip

apex that favor easy penetration into the skin, a sturdy shaft and a rounded base, which decreases

stress concentration when the spike is loaded with shear forces.

The spikes are coated with a silver-silverchloride (Ag-AgCl) double layer to reduce polarizablity of

the electrode-electrolyte interface, where the electrode is the spike surface and the electrolyte the

tissue in the Living Epidermis [46, 50, 56, 57, Paper 1, Paper 4].

A suitable package consists of a thin, circular disk (e.g. a double sided printed circuit board) glued

to the back side of the spike array chip, as shown in Figure 5.6. A ring-shaped adhesive tape is fitted

onto the disk and firmly adheres the disk and the electrode chip to the skin. The spiked electrode is

connected to the analyzing electronics by a lead wire. As the wire diameter is larger than the length of

the spikes, it must be attached to the back side of the package to avoid any influence on the proper

penetration of the spikes into the skin. Dry use of the spiked electrode without any skin preparation

presupposes that the electrode chip is securely attached to the skin allowing the spikes to pierce the

Stratum Corneum and penetrate the Epidermis. Choosing an appropriate tape with very good adhesion

performance is therefore of importance.

Lead wire

PCB

Double sidedcopper pad

Double sidedmedical tape

Spike array chip

Conductingepoxy

Solder

a) b)

Figure 5.6: Packaging of spiked EEG electrodes. a) A drawing of the exploded view of the package. The

double-sided printed circuit board (PCB) features electrically interconnected copper pads on both sides. The chip

is glued to the lower pad using conductive epoxy; the lead wire is soldered to the upper pad. b) Two photographs

(from the front – and the back side) of the electrode assembly ready to be applied on the skin. The scale is 1 cm.

Pre-clinical tests were done at Datex-Ohmeda and at the Microsystem Technology group involving

volunteer test persons [Paper 4]. During these tests we observed that no skin preparation is needed

prior to the application of the spiked electrode on the skin and no electrolytic gel is required during the


21

use of the electrode. The electrode-skin impedance (ESI) is low and stable. Both ESI magnitude and

stability are in the range of commercially available high performance electrodes using electrolytic gel,

e.g. Zipprep™ or 3M Red Dot (see Table 5.2). The microscaled spikes do not create a sensation of

pain and especially during long-term measurements, the spiked electrode is more comfortable for the

patient than wet commercial electrodes. Due to the mechanical strength of the spikes, handling of the

spiked electrodes is convenient and does not require special care by the personnel.

Figure 5.7 shows a 5 second epoch of a typical EEG recorded with spiked electrodes. The shown

signal is representative when monitoring anesthesia via EEG. In the case of the shown EEG, the

subject is still conscious and thus unwanted signals, for example eye blinking or muscle activity, are

recorded too. Preliminary analysis of the calculated entropy of the recorded EEG does not show any

significant difference between the spiked and the wet control electrode. Further, visual examination of

both signals shows that they are similar enough to be used for entropy determination [48].

The spiked electrode project has won the national final of the Swedish Innovation Cup 2000,

organized by Skandia, a global savings company, and Dagens Industri, an industrial news paper. The

prize awards innovative research projects with commercial interest.

Time [s]

Spiked Electrode

Control Electrode

15

10

5

0

-5

-100 1 2 3 4 5

Figure 5.7: Comparison of the EEG recorded by spiked electrodes and a commercial control electrode

(ZipprepTM). The total epoch is 5 s long; the sampling rate is 100 Hz. The three distinct transients correspond to

eye blinks. As long as the patient is still awake, the shown EEG is representative during the monitoring of

anesthesia levels.

In an attempt to simplify the electrode packaging, a fabrication process was developed that allows

barbs to be created at the tip of the spikes, as shown in Figure 5.8 [Paper 2]. The barbs allow a spike

to penetrate the skin but offer resistance to detaching forces. Thus there is the possibility to attach a

spike array to the skin without using adhesive tape. Although the barbs significantly increase the


22

required detachment forces compared to spike arrays without barbs, the resulting attachment force

created by the barbs was not large enough to be used in clinical applications

TABLE 5.2:

COMPARISON OF THE MAXIMAL AND MINIMAL MEASURED TOTAL IMPEDANCE OF COMMERCIAL 3M RED DOTTM

AND SPIKED ELECTRODES AT 1000 HZ AND APPROXIMATELY 0.5 HZ. THE SPIKED ELECTRODES CONSIDERED IN

THIS TABLE ARE CHLORIDED AND HAVE A SIZE OF 3X3 MM OR LARGER.

Max kΩ Min kΩ

R at 1000Hz Dry Spiked 4.23 0.65

R at 1000Hz Wet 3M Red Dot 3.17 0.5

R at 0.6 Hz Dry Spiked 87 16

R at 0.5 Hz Wet 3M Red 177 14.8

.

Figure 5.8: (left) SEM image of an array of barbed spikes. (right) Photograph to demonstrate the attachment of a

chip to human skin by barbed spikes (the same type of barbs as shown in the left image). The attachment is such

that the skin is pulled upwards by the barbed array without the help of any other adhesion enhancer.

In this work we substantiate that spiked electrodes have the potential to be used as sensors

recording EEG for anesthesia monitoring. EEG monitoring methods have increasing impact on

patient treatment in hospitals and represent a considerable market potential.

Future work will include the transfer of the microneedle technology to industrial manufacturing

methods for high volume production. Based on initial tests, it appears that plastic injection is a very

plausible candidate as such a production method. However, silicon processing will always have an

important position in the fabrication process. It is from silicon master structures that the moulds are

electroplated, which is a well-proven technology known from the compact disk fabrication lines.

23

6 Microfluidic Transdermal Liquid Delivery

System

Biotechnology and modern pharmaceutical technology enable a new generation of macromolecular

compounds with great therapeutic promise, including recombinant protein drugs [58]. Oral delivery is

not a realistic option for these molecules because of their degradation in the stomach, intestine or liver.

Alternative delivery routes must be found.

If drug volume and subcutaneous tolerability are acceptable, compounds that cannot be delivered

orally will most often be administered by injection. The use of standard hypodermic needles in

conjunction with syringes raises concerns related to comfort for the patient and thus to their suitability

for home treatment and self-injection. Therefore, it seems that optimization of the delivery of drugs

via the transdermal route is very important for modern therapies.

Large molecules, such as proteins or DNA chains, can not be delivered through the skin without

penetration enhancement since the excellent barrier characteristics prevented transport in

therapeutically relevant rates [59]. In contrast to low molecular weight entities used in patches for

smoke cessation, for example.

In future, many modern therapies will be vaccinations. Vaccines currently under development

include peptide drugs that raise high-density lipoproteins (i.e. “good cholesterol”) as well as

immunotherapies against Alzheimer’s disease and cancer [60, 61]. Another trend is visible regarding

the delivery route. In 2001, out of 129 drug delivery candidate products under clinical evaluation in

the USA, 51 were transdermal or dermal systems [62].

In this chapter micromachined interfaces are explored, which interconnect the lower skin layers or

hypodermic tissue with the outer world via microchannels and thus efficiently enhances the transport

of large molecules into or from the tissue. The interface also contains a unit, which accurately doses

and actuates the liquid to be delivered.

The transdermal liquid delivery system is subdivided in two sub-systems:

(i) the microfluidic transdermal interface (MTI) which creates physical pathways for the

liquid through the Stratum Corneum using microneedles and

(ii) the dosing and actuation unit (DAU) which is pumping the appropriate amount of liquid

through the MTI at the appropriate moment.


24

Both sub-systems are sought to be disposable devices and it is very important that they can be

sterilized by standard methods.

6.1 Microfluidic Transdermal Interface

6.1.1 Objectives

An important objective of the MTI is reduction of pain for the patient compared to syringe

injections. Reducing pain in transdermal drug delivery increases patient acceptance and thus increases

market potential for a liquid transdermal therapy. In addition, the MTI should be minimally invasive

to reduce trauma caused in the skin during liquid injection. Reducing trauma reduces the risk for

infection and speeds up healing processes.

The concept of the MTI is based on the creation of physical pathways, i.e. channels, into the tissue

below the Stratum Corneum. Low flow resistance of these channels enables effective liquid transfer

through the Stratum Corneum. During the injection, the tissue must absorb the liquid injected through

the channels, which results in a flow resistance created by the tissue. Therefore, the larger the delivery

area of the microneedles, the lower the flow resistance created by the tissue.

Penetration of the microneedles into the skin is a prerequisite for the described transdermal liquid

delivery. Therefore it is important to aim for sharp microneedles that offer smooth penetration into the

skin.

Lastly, a high resistance to mechanical stress is critical. The microneedles should not break while

inserted into the skin to avoid harm to the patient.

Table 6.1 is a summary of the objectives of the MTI.

TABLE 6.1:

LIST OF THE MAIN OBJECTIVES OF THE MICROFLUIDIC TRANSDERMAL INTERFACE IN DECREASING ORDER OF

PRIORITY.

1.Minimize the resistance of the Stratum Corneum to the transdermal delivery of large

molecules

2. Significantly reduce pain compared to standard injection

3. Minimize tissue trauma caused by the delivery

4. Minimize the mechanical resistance of the skin to microneedle penetration

5. Maximize the mechanical stability of the microneedle (array)

6. Disposability

Microfluidic Transdermal Liquid Delivery System

25

6.1.2 Prototype Devices

DRIE technology presented in chapter 2 was used to manufacture prototypes of MTI chips as

shown in Figure 6.1 [Paper 5]. The microneedles are approximately 200 µm long and feature side

openings and a solid tip apex.

Figure 6.1: Side opened microneedles approximately 200 µm in length.

Let us consider hollow tip opened microneedles presented elsewhere and schematically shown in

Figure 6.2 a) [18]. As it is desirable to minimize the flow resistance through the device, the bore in

the needle should be as large as possible. This creates a problem because the larger the bore, the more

probable it is to actually clog the bore during penetration of the skin. In other words: instead of

inserting the needle smoothly into the skin, the bore literally stamps a piece out of the skin. To avoid

this problem, the diameter of the bore can be reduced, which then again increases flow resistance in

the device and reduces delivery area. Another disadvantage of this design is that the tip opening is

never larger than the cross-section of the bore, which means that the drug delivery area in the tissue is

limited by the bore cross-section.

Efforts were made to overcome the “cookie cutter effect” of tip opened microneedles by taking

standard hypodermic needles as examples, as shown in Figure 6.2 b) [17, 19]. This increases delivery

area and creates a sharp edge, which facilitates skin penetration. The needle is slicing the skin with

the protruding edge before the bore enters the Stratum Corneum.

The side-opened microneedle design shown in this thesis features a novel approach, as illustrated in

Figure 6.2 c). The delivery area is at the shaft of the microneedle instead of at the tip apex, thus

avoiding the “cookie cutter effect”, even when large bore diameters are used. A slender shape of the

microneedle in combination with the sharp tip apex and four blades cutting into the tissue offer low

resistance during skin penetration. The piercing characteristics of this needle design into aluminum

foil is very good, the aluminum is sliced in a well-defined manner, as depicted in Figure 6.3. The

author believes that the side-opened microneedles penetrate into the skin in a similar, well-defined

manner. Because of the out-of-plane configuration of the design, two-dimensional arrays of needles


26

are fabricated on wafer level. This is an advantage for mechanical stability: A needle array inserted

into the skin is much more resistant to mechanical stress than one single in-plane needle. The

microneedle array as shown in Figure 6.1 was inserted into the skin and removed from the skin

repeatedly without observing any damage.

Figure 6.5 illustrates that the developed process technology makes it possible to design a delivery

area, which is adapted to a specific application. The size of the delivery area is independent of the

bore diameter and it can be positioned along the needle shaft. The delivery area of the needles shown

in Figure 6.1 is twice as large as it would be for a tip opened needle having the same bore diameter.

Path of the injected liquid

a) b) c)

Tip opening Oblique opening Side opening

Figure 6.2: Schematic drawing of the vertical cross-section of different hollow microneedle concepts. (a)

Suffers from high clogging probability during penetration into the skin. (b) Shows a schematic drawing of a

microneedle with an oblique opening, much like standard hypodermic neeldes. (c) Is the side opened

microneedle concept: low clogging probability combined with low flow resistance and good penetration

capabilities.

Figure 6.3: SEM of a MTI piercing 10 µm thick aluminum foil. Note the well-defined manner the needle cuts

the aluminum.


27

200

300

400

500

600

700

800

900

18

20

22

24

26

28

30

32

2 3 4 5 6 7

Hypodermic Needle Diameter [µm]

Hypodermic Needle Gauge

Hyp

oder

mic

Nee

dle

Dia

met

er [

µm

]

Hypoderm

ic Needle G

auge

Chip Side Length [mm]

Figure 6.4: Comparison of the MTI chip to standard hypodermic needles. (left) SEM photograph of a 3×3 mm MTI

chip and two hypodermic needles of different diameters (0.36 mm and 0.9 mm). The semi-spherical “bump” at the

right corner of the MTI chip is conductive epoxy required for fixation of the chip to the sample holder. (right)

Relation between the side length of a MTI chip and the diameter of a hypodermic needle, both having the identical

total area of the bore cross-section.

sr t

Figure 6.5: The left schematic drawing illustrates some dimensions (e.g. r, s, and t) that the designer of the side-

opened microneedle is free to choose, in addition to the total length. The right SEM image shows that the size of

the delivery area can be pushed to the base of the needle.

It is interesting to compare the effective cross-sectional area of the liquid channel into the skin

created by standard hypodermic needles and the MTI, as shown in Figure 6.4. The diameter of a

hypodermic needle can be represented both with metric diameter and the needle gauge. A 3 × 3 mm

MTI chip as shown creates a channel into the skin which has approximately the same cross-sectional

area as a gauge1 27 needle for insulin delivery.

Vaccinations are well suited for micromachined needle arrays. In section 4.1 we have seen that

dendritic cells, which are initiators and mediators of the immune response, can be found in the Living

1 gauge 27 corresponds to a needle diameter of 360 µm


28

Epidermis and Stratum Germinativum. Although there are several possible sites for vaccine delivery,

including the muscle and mucosal surfaces, it seems that the optimal delivery of vaccines is via

injections into the Epidermis or the Dermis [66]. This kind of injection is difficult to administer with

standard hypodermic needles since the targeted skin thickness is shallow, often below the diameter of

the actual needle and therefore prone to human error.

There is a pronounced public interest in a painfree drug injection device replacing standard

injection needles/syringes [63, 64, 65]. La raison d’être for this interest can be found in the often

traumatic experiences and pain associated with hypodermic needles.

Future work includes the detailed investigation of the mechanisms of liquid injection into the skin

via side opened microneedle arrays. This study should not only include the performance of the actual

micromachined interface at a specific location on the body, but also the influence of the skin on

different locations on the body. If the intra- and/or transdermal liquid delivery proves to be

reproducible, a first step towards an applicable novel instrument will have been taken.

6.2 Dosing & Actuation Unit

6.2.1 Objectives

The dosing and actuation unit (DAU) relieves the patient from concerns about the dosing and

delivery of a drug over a certain period of time. The DAU should thus release the drug without human

interaction. There are two main objectives:

(i) measure the exact quantity of the therapeutic agent and

(ii) pump that quantity through the MTI into the tissue.

The pumping action is also referred to as actuation. In view of prolonged delivery schemes and

excellent portability it is desirable to have an actuator which spares energy consumption and which

has small dimensions. It is very important that the drug is not modified chemically or physically by

the materials used or the actuation principle.

TABLE 6.2:

LIST OF THE MAIN OBJECTIVES OF THE DOSING AND ACTUATION UNIT IN DECREASING ORDER OF PRIORITY.

1. Digital dosing and actuation of the drug to be delivered

2. No chemical modification of the drug

3. Minimize device complexity

4. Minimize dimensions and weight

5. Disposability


29

6.2.2 Expandable Microspheres

Expancel microspheres are produced by Expancel, Sundsvall, Sweden, in a wide variety of types.

They are small plastic particles that expand when heat is applied. The dramatic expansion of the

microspheres when heated is caused by a small amount of a liquid hydrocarbon, e.g. isobutane or

isopentane, encapsulated by a gas tight thermoplastic shell. The shell is composed of a copolymer of

some monomers, e.g. vinylidene chloride, acrylonitrile, and methylmethacrylate. Upon heating, the

shell softens and the liquid undergoes a phase change to gas. Thus, the microsphere is literally

inflated, much like a balloon. This results in a decrease of the density from 1000 to 30 g/l.

After cooling down, the microspheres do not shrink because the thermoplastic shell hardens again.

Thus, the dramatic volume increase is irreversible.

The expandable microspheres are available with expansion temperatures in the range of 70 – 200

°C, whereby the volume increases with the temperature. If the temperature is increased above that at

which the highest expansion is obtained the microspheres will collapse.

The expandable microspheres can be used in contact with many chemicals, including solvents,

without negative effects on expansion. The available size range of unexpanded microspheres is from

5 – 90 µm.

Expancel beads are used in many different applications in the macro scale. For example, in

printing inks to form 3D patterns on paper, for the density decrease of plastics through a controlled

and predictable foaming process, and in reaction injection molding to counteract the shrinkage of

polyurethane.

6.2.3 Technology Demonstrator

There are two fundamentally different ways of measuring an amount of drug that is delivered, for

example, into dermal tissue:

(i) The first is to measure the flow rate of the drug (or parameters directly related to the flow

rate) during the delivery. Integrating the flow rate over time yields the total amount of

drug released into the tissue. This delivery scheme can be defined as being analog, i.e.

knowing rate and time it is theoretically possible to delivery any drug volume.

(ii) The second is to precisely define a number of unit drug volumes during fabrication (e.g.

via precisely fabricated liquid containers). A drug dose can then be achieved by adding up

a multitude of unit drug shots. In a shot, the whole liquid volume enclosed in the container

must be pressed out of the container, much like a microscaled syringe. This scheme is

somewhat digital and represented in Figure 6.6.


30

The analog method requires closed-loop sensing and processing of data to dose the liquid precisely

[67]. A digital system does not require data processing during delivery of a certain chemical dose.

This is an advantage when striving for low complexity of the device. This kind of actuation is

unidirectional and the actuator can be a one-shot device.

In this section the first tests of a novel actuator having the potential to be used in a digital dosing

and actuation unit are shown. This technology demonstrator does not address all of the above-

described objectives and functions but represents a first step to such a device.

The basic idea of the digital DAU is to replace the liquid stored in a container by solid matter as the

liquid is pressed out of the container. Expandable microspheres have the potential to fulfill this

function, as schematically shown in Figure 6.7. Bead expansion is triggered by a thermal signal.

Micromachined test devices were fabricated to demonstrate this concept [Paper 3]. In a one-shot

pump experiment, Expancel microspheres packed against a filter in a microchannel, immersed in

water were heated from 23 °C (i.e. the measured room temperature) to 70 °C under an applied counter

pressure of 100 kPa. After the beads were cooled down to room temperature again, the pressure was

varied between 20 kPa and 100 kPa whereby no volume variation of the microspheres was observed.

This shows that the microspheres expand regardless of a counter pressure of 100 kPa. It also means

that once the liquid to be pumped is displaced, it will not be possible for the liquid to flow back into

the location where it was stored before the actuation, regardless of 100 kPa of counter pressure and

cooling of the microspheres. The somewhat arbitrary value of 100 kPa is imposed by the performance

of the test set-up.

Liquid to bedelivered

Act

uato

r

Actuator

Storage

Delivery

Figure 6.6: Principle of a digital dosing system.

One-Shot Pump

Heat

pp

Liquid

Figure 6.7: Concept of a one-shot

pump based on expandable

microspheres


31

Figure 6.8: Expansion of expandable microspheres packed in a channel and immersed in water.

The expansion process is dependend on the geometry of the bead pack. Provided the depth of the

test channel is constant, the bead pack can be characterized by the ratio L/W, where L is the length of

the pack and W the width of the channel. It was observed that the relative volume expansion is

inversely proportional to the pack ratio L/W. Only the beads furthest away from the filter effectively

contribute to the volume expansion and the rest of the beads literally block the channel. Therefore, to

use the full potential of expandable microspheres, the L/W ratio of bead packs should be as small as

possible [Paper 3].

The DAU presented in this thesis differs from earlier shown discrete drug dosing systems as shown

earlier [68] where the drug is not pumped into the tissue but presented to the tissue via a normally

closed valve.

Other conventional micropumps based on, for example, thermo pneumatic, piezoelectric,

electrostatic, or electromagnetic actuators, require a relatively complex assembly which limits their

use in disposable devices [69, 70, 71]. Direct electrical actuation of liquids via electroosmotic flow

and dielectrophoresis is less suited for drug delivery applications because of the high voltages required

during actuation [72].

An innovative, low complexity, unidirectional liquid actuation concept was recently presented and

is based on controlled evaporation [73]. This principle has the great advantage of not requiring any

electrical actuation.

A dosing system based on the formation of gas bubbles via electrochemical processes during

electrolysis was presented earlier and was designed to dose nanoliter amounts of liquid [67]. The

application field of this device is the monitoring of clinically important substances via microdialysis.

There are two features of the DAU based on expandable microspheres that make it interesting for

transdermal drug delivery:

(i) high energy density, which allows the actuator to overcome high counterpressures

(ii) the fact that the delivered liquid cannot flow back into the DAU after actuation since the

liquid container is filled with expanded beads.

However, challenges remain, including the wafer level localization of the beads [74], the wafer

level integration of the DAU to the MTI, investigation of the thermal compatibility with drugs as well

as sterilizeability.

33

7 Micromachined Electrospray Emitter Tips

for Mass Spectrometry

Electrospray is a method by which ions, present in a solution, can be transferred to the gas phase.

In gas phase, the ions are then accessible to a mass spectrometer that detects their mass to charge ratio.

In this chapter the basic mechanisms of electrospray ionization are explored as well as a

biochemical application area, that is proteomics, where electrospray ionization mass spectrometry is of

great interest. Subsequently, we present an interface, which is designed to transfer biopolymers

present in a solution to gas phase for subsequent mass spectrometry detection.

7.1 Electrospray Ionization

Electrospray is an atmospheric pressure ionization method. The process is known since the

beginning of the century but in the 1960´s the first large molecules were transferred to gas phase [75].

The process involves an electric field concentrated at the apex of a hollow tip containing a solution of

the ions to be transferred to gas phase, as illustrated in Figure 7.1. The tip is often referred to as an

emitter. If the counter electrode generating the electric field has a negative potential, positive ions

accumulate at the emitter orifice. The high density of positive charges at the needle tip leads to the

formation of a Taylor cone [76]. Under a high electric field rE , the repulsive forces become stronger

than the surface tension at the tip of the cone and a liquid spray is generated, i.e. small droplets of

liquid are ejected from the cone.

There are three major steps that extract gas-phase ions from the droplets [77]:

(i) production of charged droplets at the emitter orifice

(ii) shrinkage of the charged droplets by solvent evaporation and repeated droplet

“explosion” due to charge concentration at the surface of the droplet

(iii) the actual mechanism by which the gas-phase ions are produced from the very small and

highly charged droplets.

The third step is not yet fully understood, but two mechanisms have been proposed in literature.

The charge-residue model states that the solvent evaporates and droplets break up until droplets with

only a single ion are formed [75]. The direct emission model suggests that droplets with a radius of

less than 10 nm can allow field desorption, i.e. they directly emit the gaseous ion of interest [78].


34

The opposing effects of the electrostatic force and surface tension of the liquid at the emitter orifice

were investigated [76, 79]. It was found that in the static case, i.e. before droplets are emitted, the

liquid surface takes the shape of a cone with a half angle of 49.3° at the apex. In the dynamic case, the

shape of the cone depends on the experimental conditions.

V+ -

++++

+++++

+

+++++

++++

+ ++++

+++ ++++++ ++ +++

++++++

++

+ +++

+

+ ++

++

+

+

+

+

+

++

+

++

+

+ +

+

+

++

+

+

+

+

+

+ +

+

+

+

--

--

-

-

-

--

-- -

--

---

--

-

-

--

-- --

---

++++

+

+

+

--

-

--

-

E

E

High Voltage

++++

++

-

--

--

rC

Emitter Tip

Solution

CathodeAnode

Cation

Anion

Figure 7.1: Schematic representation of the electrospray ionization process.

7.2 Proteomics

The term “proteome” refers to the total set of proteins expressed in a given cell at a given time; the

study of which is termed proteomics [80]. In the last few years mass spectrometry applied to

biological samples has seen tremendous activity and has become a wide spread technique within

proteomics [81]. Proteomics is complementary to genomics and now includes protein identification,

post-translational modifications, protein function, protein-protein interactions, etc. With the

accumulation of vast amounts of DNA sequences in databases, researchers are realizing that merely

having complete sequences of genomes is not sufficient to clarify biological function [81].

The promise and potential benefits of proteomics are such that corporate and academic researchers

are flocking the area. Apart from purely scientific interest, the disclosure of how protein cascades

change resulting from disorders or drug action has tremendous commercial potential for

pharmaceutical companies [80]. Proteomics directly contributes to drug development since almost all

drugs are directed against proteins.

The success of mass spectrometry (MS) within proteomics is a direct consequence of the

introduction of the soft ionization methods of electrospray ionization [82] and matrix-assisted laser

Micromachined Electrospray Emitter Tips for Mass Spectrometry

35

desorption/ionization (MALDI) [83]. Often, MS relies on the digestion of gel-separated proteins into

peptides by a sequence specific protease such as trypsin. The reason for analyzing peptides rather than

proteins is that the molecular weight of proteins is not usually sufficient for database identification

[81]. In contrast, peptides are easily eluted from gels and even small sets of peptides from a protein

provide sufficient information for identification.

In proteomics research there is a need for large scale, high throughput methods to support the

staggering number of protein analyses [84].

7.3 Micromachined Electrospray Emitter Tips

7.3.1 Objectives

The most obvious goal in investigating micromachined electrospray emitter tips is to create two-

dimensional arrays of emitters along with microfluidic functionality on the same chip. These two

features give the possibility to speed up protein/peptide detection and identification by automating

sample separation, preparation and ionization for mass spectrometry detection. In addition to speeding

up the detection process it also reduces sample volumes and reduces exposure of the analyte to the lab

environment.

The emitter tip itself should show very good spray performance, i.e. a high yield of conversion

from ions in their dissolved state to the gas phase. A high yield can be obtained by low sample flow

rates in the emitter [77]. Low flow rates create small droplets from which ions are extracted faster and

easier than from large droplets. Low flow rates are achieved by low potential differences between the

emitter and the counter electrode [79]. The required potential difference to generate a flow (for given

conditions such as the distance between emitter and counter electrode, physical liquid properties, etc.)

is dependent on how effective the electric field lines are concentrated at the emitter tip and thus

dependent on the base radius rC of the Taylor cone (Figure 7.1). Therefore, the lower rC , the lower the

potential difference required and the better the spray performance.

The differences in dimensions, e.g. orifice diameter or emitter tip length, should be minimized

when comparing tips of the same array and between different arrays, thereby avoiding variations of

spray conditions. Geometrical reproducibility is the key to automated, high throughput electrospray

mass spectrometry.

Table 7.1 is a summary of the objectives of the electrospray emitter work.

7.3.2 Technology Demonstrator

Process technology as presented earlier in chapter 3 was used to manufacture chip based

electrospray emitter tips. The out-of plane tips were arranged in a 2D array as illustrated in Figure 7.2.


36

Two types of emitters were fabricated and self-aligning principles were applied for the fabrication

of the emitter bore. The first type consists of silicon and is referred to as Si-tip; the second type

consists of silicon dioxide and is referred to as SiO2-tip, as depicted in Figure 7.3. The rim of the Si-

tip opening is extremely sharp, in the nanometer range. The circular rim of the SiO2-tip is defined by

the thickness of the grown oxide, which in this case is 1.5 µm in thickness. The rim of the tip opening

is less sharp compared to the edge of the silicon version. For characterization, the substrate wafer was

diced into 1×1 mm chips containing one tip and glued to the end of a fused silica capillary. With this

design the chip/capillary-assembly could be operated as a standard nanoelectrospray tip, as shown in

Figure 7.4.

TABLE 7.1:

LIST OF THE MAIN OBJECTIVES OF THE MICROMACHINED ELECTROSPRAY EMITTER TIPS FOR MASS SPECTROMETRY

IN DECREASING ORDER OF PRIORITY.

1. Enable two dimensional arrays of emitter tips for mass spectrometry

2. Enable integration of microfluidic functionality for each emitter tip without creating large

dead liquid volumes

3. Minimize Taylor cone base radius to the radius of the emitter orifice

4. High reproducibility of the emitter dimensions and geometry

Figure 7.2: Array of MS Tips


37

70 µ

m

70 µ

m

Figure 7.3: Zoom on both types of electrospray emitter tips. (left) Silicon tip. (right) Silicon dioxide tip.

Let us consider a sample solution and an arbitrary emitter tip as shown in Figure 7.5. The liquid

forms a certain contact angle θ C with the tip material depending on both the liquid and surface

properties. The emitter is characterized by the angle β, which defines how the channel geometry is

altered at the orifice. If β ≥ π/2-θC, the orifice at the apex of the emitter acts as a geometrical stop

valve [85]. In the hydrophilic case, θC < 90°, the liquid will fill the channel via capillary forces and

stop at the orifice. To wet the tapered shaft of the tip, the internal pressure ∆p of the liquid must be

increased so as the liquid meniscus whisks out to form θC between the shaft and the liquid. The larger

β, the larger the required pressure to obtain wetting of the shaft.

Let us now compare the tapered emitter presented in this thesis with flat nozzles, as illustrated in

Figure 7.6. In view of the above mentioned, the optimum size of the Taylor cone is achieved when its

base radius rC corresponds to the radius of the channel orifice in the emitter. It becomes clear now that

a sharp rim limits the Taylor cone base to the orifice whereas in the nozzle case the flat top of the

nozzle is wetted, provided that the flat nozzle top is hydrophilic. Thus the size of the cone is

increased, as schematically represented in Figure 7.6.

When visually examining the spray performance of the tapered tip emitters in an electric field it

was confirmed that the cone does not wet the shaft of the tip (Figure 7.7).


38

Sample container

Pressurized air

Metal connector

Metal bar

XYZ-stage

Mass Spectrometer

Inlet

High voltage

Figure 7.4: Operation of the micromachined nESI emitter demonstrator

Liquid

Tip structure

∆p

β

θ

∆p1

β

θ

β

θ

β

θ∆p2

Figure 7.5: Representation of the geometrical stop

valve and the pressure required to wick the fluid

meniscus, ∆p1 > ∆ p2. All surfaces are hydrophilic.

Liquid

Tip structure

Taylor cone

d d d

β=90°β≅160°β≅160°

Figure 7.6: comparison of the presented tapered

emitter tips and flat nozzle emitter. Because the angle

β is smaller for the nozzle case compared to the

presented sharp rim orifices, the flat nozzle top is

wetted, thus increasing the Taylor cone size. All

surfaces are hydrophilic.


39

Figure 7.7: Photograph of an electrospray of acetonitrile with 1% acetic acid from a silicon dioxide tip.

The presented micromachined electrospray emitter tips were extensively tested at Analytical

Chemistry, Royal Institute of Technology, Stockholm, Sweden and were successfully utilized for

protein and peptide analysis. A stable Taylor cone was observed during mass spectrometry and most

importantly, a higher repeatability compared to commercial pulled silica capillary tips resulted [86].

High repeatability resulted from the very reproducible emitter dimensions and is prerequisite for

automated, reproducible and robust operation in a high throughput analysis system. Commercial

players (Advion Inc., Ithaca, NY, USA) have recognized the need for such a system and started

commercializing micromachined emitter arrays [6].

Future work can include the integration of chemical functions on the chip containing the

micromachined emitter (array). For example, sample preparation, separation, or concentration.

41

8 Final Conclusions

Deep reactive ion etching is a powerful tool to sculpture truly three-dimensional structures; in

particular differently shaped tapered out-of-plane projections in the micro-scale (i.e. microneedles,

microspikes or microtips).

Medical and biochemical interfaces based on such microprojections show novel, beneficial

features.

The spiked EEG electrode, as an element of the Datex-Ohmeda anesthesia monitor, makes more

efficient surgical procedures plausible without compromising patients comfort and ease of use for

nurses.

The transdermal drug delivery system, and in particular the side-opened microneedle arrays,

considerably increase patient acceptance for transdermal drug delivery. Comfort for the patient is an

important condition when responding to the increased demand for home treatment and self-injection of

active compounds.

The micromachined electrospray ionization emitter tips combine spray efficiency of conventional

high-performance emitters with the possibility to batch fabricate two dimensional emitter arrays with

integrated microfluidics, for example in the separation of analytes.

Interestingly, the presented interfaces reduce the amount and influence of human interaction to

achieve a certain function when compared to their conventional counterparts. There is no need to

prepare the skin and apply gel prior to biopotential measurements. During the delivery of vaccines that

require shallow intradermal injections there is no risk for erroneous delivery too deep into the

subcutaneous tissue or even the muscle. The very reproducible dimensions of micromachined

electrospray emitter tips make that the automation of mass spectrometry analysis of biopolymers

conceivable.

Further research is required to evaluate the reliability and effectiveness of these concepts; an

important step towards applicable instruments. However, this step presents a worthy challenge with

great implications in accuracy, versatility, ease of use, comfort for the patient, as well as reduced

costs, should it prove successful.

43

9 Acknowledgements

With hesitation I consider this “book” as finished and look back on the time I spent in Stockholm.

It was full of adventure, instructive, exciting, sometimes dark and also very light… and cold.

The work was carried out at the Microsystem Technology Group, Department of Signals, Sensors

and Systems, Royal Institute of Technology, Stockholm, Sweden. Funding was gratefully received

from Datex-Ohmeda.

Without the professional or moral support of many persons, it would not have been possible to

come to this point.

In particular, I would like to express my sincere gratitude to:

My supervisor Göran Stemme for enthusiastic and generous guidance, for always being available,

and for enabling me to present this work in many places.

Pekka Meriläinen of Datex-Ohmeda for initiating our collaboration and for sharing his tremendous

experience and many ideas.

My colleagues and friends in the Microsystem Technology group. Wouter van der Wijngaart for

many stimulating discussions, for non-work related activities of the humpy-bumpy type and for being

“en riktig kompis”; Jessica Melin for forgiving me for always handing over manuscripts to be

proofread too close before the deadline (including this thesis…); Kjell Norén for his clever view on

packaging and supreme skills to produce mechanics and electronics; Sjoerd Haasl for optimizing the

world’s first clean-room walk, for his irrevocably positive thinking and for making work after

midnight less boring; Frank Niklaus for literally bleeding for my research and enduring a skin biopsy;

Hans Sohlström for his help in graphical and mathematical issues, especially the cover picture of this

book.

My ex-colleagues Peter Enoksson who got me started at the Microsystem Technology Group and

issued the most valuable ICP license and Helene Andersson for her different view on all sorts of

“small stuff”.


44

My colleagues at Datex-Ohmeda: Heli Tolvanen-Laakso and Magnus Kåll for working on the

spiked electrode project in Helsinki and for helpful discussions around the device.

Stig Ollmar of Medical Engineering, Karolinska Institute, for sharing his extensive knowledge of

skin impedance and for helping with the decisive first impedance measurements on the spiked

electrode.

Johan Sjödahl of Analytical Chemistry, Royal Institute of Technology, for characterizing the ESI-

tips and for explaining the “who’s who” of all kinds of “strange” molecules.

Peter Georén, Carina Lagergren, Anton Lundqvist and Anna-Karin Hjelm of the Department of

Chemical Engineering Technologies, Royal Institute of Technology, for instructions on how to do

impedance measurements on, hopefully still, healthy volunteers.

Marianne Ekman and Lennart Emtestam for their support in obtaining skin sections.

Martin and Emma for many superb hours together: in a kajak, on a mountain bike or on Norwegian

mountains (cramponer definitely rule!).

Roli and Oli back in Switzerland for being true friends.

My mother, father, sister and brother-in-law: I love you too!

45

1 0 Summary of the Appended Papers

Paper 1

The microfabrication of spike arrays is described. The shown packaging allowed for the preliminary

testing of the arrays as biopotential electrodes.

Electrode-skin-electrode impedance measurements confirmed that spiked electrodes do not need any

skin preparation or electrolytic gel to achieve comparable performance to standard wet electrodes.

Thus, spiked electrodes allow high quality recording of low level biopotentials such as signals

produced by the human brain (i.e. EEG recording). The application of the electrode on and removal

from the skin is fast and uncomplicated. Furthermore, the size of the spiked electrode can significantly

be reduced compared to standard electrodes.

Paper 2

Microfabrication methods for the machining of barbed spikes is described. The barbs are means of

attaching a chip to surfaces capable of being penetrated by silicon spikes. The attachment principle is

inspired by fishhooks and arrows whereby the barbs penetrate a material but offer resistance to

detaching forces.

The mechanical attachment of two types of barbed spike arrays was measured on non-biological

materials (Polyethylene foil, Aluminum foil and ParafilmTM) as well as on human skin. The maximum

adhesion force was 1,36 N achieved by a 2x2 mm2 chip containing 64 barbed spikes, 160 µm in

length, pressed into Polyethylene foil. Maximum adhesion to skin was 0,54 N.

Paper 3

Two novel concepts for controlled handling of liquids in microfluidic systems are presented: a one-

shot micropump and a normally open one-shot valve based on Expancel microspheres.

Expancel® microspheres are small spherical plastic particles that, when heated, increase their volume

considerably. It is shown that liquid volumes in the nanoliter range can be actuated against a counter

pressure of at least 100 kPa and that a channel can block fluid flow in a microchannel against

pressures of at least 100 kPa. The emphasis in relation to this thesis is on the one-shot pump since a

possible application of such a device is in transdermal drug delivery. The pumping principle is largely


46

independent on the liquid properties and prevents the delivered liquid from flowing back into the

pump.

A short review on different types of liquid handling (pumping and valving) is given in the introduction

of this paper.

Paper 4

The characterization of dry spiked biopotential electrodes is presented. It was found that the

electrode-skin-electrode impedance (ESEI) is strongly dependent on the electrode size (i.e. the

number of spikes) and the coating material of the spikes. Electrodes larger than 3x3 mm2 coated with

Ag/AgCl have sufficiently low ESEI to be well suited for electroencephalograph (EEG) recordings.

The maximum measured ESEI was 4.24 kΩ and 87 kΩ, at 1 kHz and 0.6 Hz, respectively. The

minimum ESEI was 0.65 kΩ and 16 kΩ, at the same frequencies. The ESEI of spiked electrodes is

stable over an extended period of time. The arithmetical mean of the generated DC offset voltage of

11.8 mV immediately after application on the skin and 9.8 mV after 20 to 30 min. A spectral study of

the generated potential difference was unstable at frequencies below approximately 0.8 Hz. Thus, the

signal does not interfere with a number of clinical applications using real time EEG.

Comparing raw EEG recordings of the spiked electrode to commercial high performance electrodes

(i.e. ZipprepTM) showed that both signals were similar. Due to the mechanical strength of the silicon

microneedles and the fact that neither skin preparation nor electrolytic gel is required, use of the

spiked electrode is convenient. The spiked electrode is very comfortable for the patient.

Paper 5

The first hollow out-of-wafer-plane silicon microneedles having openings in the shaft rather than

having an orifice at the tip are presented. These structures are well suited for transdermal microfluidic

applications, e.g. drug- or vaccine delivery. The developed deep reactive ion etching (DRIE) process

allows fabrication of two dimensional, mechanically highly resistant needle arrays offering low

resistance to liquid flows and a large liquid delivery area for the tissue. The presented process does

not require much wafer handling and only two photolithography steps are required. Using a 3x3 mm2

chip in a typical application, e.g. vaccine delivery, a 100 µl volume of aqueous fluid injected in 2 s

would cause a pressure drop of less than 2 kPa.

A short review on microprojections in the biomedical field is given.

Paper 6

Two novel types of micromachined nanoelectrospray emitter tips were designed, fabricated and tested.

The fabrication method of the hollow tips is based on a self-aligning deep reactive ion etch (DRIE)

process. The tips consist either of silicon dioxide or silicon and feature orifice diameters of 10 and

18 µm, respectively. The geometrical characteristics of both emitter types are favorable for the

Summary of the Appended

47

generation of stable electrospray ionization, i.e. wetting of the tip shaft is avoided and the base of the

Taylor cone is limited to the diameter of the orifice.

A silicon dioxide tip was operated in a bench top set-up to visually evaluate of the electrospray. Both

kinds of tips were also successfully used for the analysis of an insulin sample in an ion trap mass

spectrometer.

A short review on the use of microengineering in electrospray mass spectrometry is given in the

introduction of this paper.

49

1 1 Glossary

anatomy a branch of biology that deals with the study of the structure of organisms

anesthesia loss of sensation with or without loss of consciousness

anisotropic exhibiting properties (for example etch speed) with different values when

measured along different directions

aspect-ratio ratio of the length to the width of a structure (for example a hole or a

projection)

barb a sharp projection extending backward (as from the point of an arrow or

fishhook) and preventing easy extraction

biopolymer a polymeric substance (as a protein) formed in a biological system

biopotential potential created on the skin or within the body due to the excitation of active

cells such as nerves or muscles

cardiology the study of the heart and its action and diseases

collagen an insoluble fibrous protein that occurs in vertebrates as the chief constituent of

connective tissue fibrils and in bones. It is connective tissue of the dermis that

when properly prepared from animals by tanning, becomes the tough leather

that has many uses.

dendritic cells dendritic cells are located strategically at the interface to potential pathogen

entry sites and take up antigen, move into lymphoid tissues and trigger immune

response

electrocardiography detecting and recording the changes of electrical potential occurring during the

heartbeat. Used especially in diagnosing abnormalities of heart action

electroencephalography detecting and recording brain activity

electromyography detecting and recording activity associated with functioning skeletal muscle

follicle a) a small anatomical cavity or deep narrow-mouthed depression

b) a small lymph node

gauge the diameter of a slender object (as wire or a hypodermic needle)

gland cell, group of cells, or organ that selectively removes materials from the blood,

concentrates or alters them, and secretes them for further use in the body or for


50

elimination from the body

hypnosis any of various conditions that resemble sleep

hypodermic adapted for use in or administered by injection beneath the skin

intradermal situated, occurring, or done within or between the layers of the skin; also :

administered by entering the skin

invasive a medical procedure which requires puncture or incision of the skin or insertion

of an instrument or foreign material into the body.

isotropic exhibiting properties (for example etch speed) with the same values when

measured along all axes

keratin any of various sulfur-containing fibrous proteins that form the chemical basis of

horny epidermal tissues (as hair and nails)

neurology the scientific study of the nervous system especially in respect to its structure,

functions, and abnormalities

parenteral an administration route for drugs via injection. This may be intravenously (into

a vein), subcutaneously (under the skin) or intramuscularly (into a muscle)

pathogen a specific causative agent (as a bacterium or virus) of disease

physiology a branch of biology that deals with the functions and activities of life or of

living matter (as organs, tissues, or cells)

proteomics study of the total set of proteins in a given cell at a given time

subcutaneous being, living, used, or made under the skin

transdermal through the dermis

trauma tissue damage caused by external force

ventilator an adjustable positive pressure generator that acts as an artificial breathing

mechanism during anesthesia or respiratory depression

51

1 2 References

[1] G. T. A. Kovacs, Micromachined Transducers Sourcebook: WCB/McGraw-Hill, 1998.

[2] “Annual Report,” Instrumentarium Corporation, Helsinki 2000.

[3] http://www.datex-ohmeda.com/company/index.htm, .

[4] R. Eckert, J. M. Freyland, H. Gersen, H. Heinzelmann, G. Schürmann, W. Noell, U. Stauffer,

and N. F. d. Rooij, “Near-field fluorescence imaging with 32 nm resolution based on microfabricated

cantilever probes,” Applied Physics Letters, vol. 77, pp. 3695-3697, 2000.

[5] G. Schürmann, W. Noell, U. Stauffer, and N. F. deRooij, “Microfabrication of a combined

AFM-SNOM sensor,” Ultramicroscopy, vol. 82, pp. 33-38, 2000.

[6] G. A. Schultz, T. N. Corso, S. J. Prosser, and S. Zhang, “A Fully Integrated Monolithic

Microchip Electrospray Device for Mass Spectrometry,” Analytical Chemistry, vol. 72, pp. 4058-

4063, 2000.

[7] R. Ohigashi, K. Tsuchiya, Y. Mita, and H. Fujita, “Micro capillaries array head for direct

drawing of fine patterns,” presented at IEEE MEMS 2001, Interlaken, Switzerland, 2001.

[8] K. Okano, K. Hoshina, M. Iida, S. Koizumi, and T. Inuzuka, “Fabrication of a diamond field

emitter array,” Applied Physics letters, vol. 64, pp. 2742-2744, 1994.

[9] K. Uemura, S. Kanemaru, and J. Itoh, “Fabrication of a vacuum-sealed magnetic sensor with a

Si field emitter tip,” Journal of Micromechanics and Microengineering, vol. 11, pp. 82-83, 2001.

[10] P. Rangsten, C. Ribbing, C. Strandman, B. Hok, and L. Smith, “Field-emitting structures

intended for a miniature X-ray source,” Sensors and Actuators: A Physical, vol. 82, pp. 24-29, 2000.

[11] L. Licklider, X.-Q. Wang, A. Desai, Y.-C. Tai, and T. D. Lee, “A Micromachined Chip-Based

Electrospray Source for Mass Spectrometry,” Analalytical Chemistry, vol. 72, 2000, 367-375.

[12] A. Desai, Y.-C. Tai, M. T. Davis, and T. D. Lee, “A MEMS Electrospray Nozzle for Mass

Spectrometry,” presented at International Conference on Solid-State Sensors andActuators,

Transducers ´97, Chicaco, June, 1997.

[13] J. Thaysen, A. Boisen, O. Hansen, and S. Bouwstra, “Atomic force microscopy probe with

piezoresistive read-out and a highly symmetrical Wheatstone bridge arrangement,” Sensors and

Actuators A: Physical, vol. 83, pp. 47-53, 2000.

[14] G. Binning, H. Rohrer, C. Gerber, and E. Weibel, “Surface Studies by Scanning Tunneling

Microscopy,” Physical Review Letters, vol. 49, pp. 57-61, 1982.


52

[15] C. Beuret, T. Akiyama, U. Stauffer, N. F. deRooij, P. Niedermann, and W. Hänni, “Conical

diamond tips realized by a double-molding process for hig-resolution profilometry and atomic force

microscopy applications,” Applied Physisc Letters, vol. 76, pp. 1621-1623, 2000.

[16] P.-F. Indermühle, G. Schürman, G.-A. Racine, and N. F. deRooij, “Self-sharpening tip

integrated on micro cantilevers with self-exciting piezoelectric sensor for parallel atomic force

microscoy,” Applied Physics letters, vol. 70, pp. 2318-2320, 1997.

[17] J. G. E. Gardeniers, J. W. Berenschot, M. J. d. Boer, Y. Yeshurun, M. Hefetz, R. v. t. Oever,

and A. v. d. Berg, “Silicon Micromachined Hollow Microneedles for Transdermal Liquid Transfer,”

presented at The 15th International Confererence on Micro Electro Mechanical Systems, Las Vegas,

USA, 2002.

[18] D. V. McAllister, F. Cros, S. P. Davis, L. M. Matta, M. R. Prausnitz, and M. G. Allen, “Three-

Dimensional Hollow Microneedle and Microtube arrays,” presented at The 10th International

Conference on Solid-State Sensors and Actuators, Sendaj, Japan,, 1999.

[19] B. Stoeber and D. Liepman, “Fluid Injection Through Out-of-Plane Microneedles,” presented at

1st Annual International IEEE-EMBS Special Topic Conference on Microtechnologies in Medicine &

Biology, Lyon, France,, 2000.

[20] W. Trimmer, P. Ling, C. Chee-Kok, P. Orton, R. Gaugler, S. Hashmi, G. Hashmi, B. Brunett,

and M. Reed, “Injection of DNA into plant and animal tissues with micromechanical piercing

structures,” presented at IEEE Micro Electro Mechanical Systems 1995, 1995.

[21] K. Chun, H. G., H. Toshiyoshi, H. Fujita, Y. Kikuchi, J. Ishikawa, Y. Murakami, and E.

Tamiya, “An Array of Hollow Microcapillaries for the Contolled Injection of Genetic Materials into

Animal/Plant Cells,” presented at MEMS 99, Orlando, FL, 1999.

[22] J. Chen, K. D. Wise, J. F. Hetke, and J. S. C. Bledsoe, “A Multichannel Neural Probe for

Selective Chemical Delivery at the Cellular Level,” IEEE Transactions on Biomedical Engineerin,

vol. 44, 1997.

[23] K. S. Lebouitz and A. P. Pisano, “Microneedles and Micro-lancets Fabricated Using SOI wafers

and Isotropic Etching,” presented at Meeting of the Electrochemical Society, Boston, USA, 1998.

[24] N. H. Talbot and A. P. Pisano, “Polymolding: Two Wafer Polysilicon Micromolding of Closed-

Flow Passages for Microneedles and Microfluidic Devices,” presented at Soli-State Sensor and

Actuator Workshop, Hilton Head Island, South Carolina, USA,, 1998.

[25] L. Lin and A. P. Pisano, “Silicon-Processed Microneedles,” IEEE Journal of

Microelectromechanical Systems, vol. 8, 1999.

[26] J. D. Brazzle, I. Papautsky, and A. B. Frazier, “Micromachined Needle Array for Drug Delivery

or Fluid Extraction,” IEEE Engineering in Medicine and Biology, vol. Nov/Dec, 1999.

[27] P. A. Stupar and A. P. Pisano, “Silicon, Parylene, and Silicon/Parylene Micro-Needles for

Strength and Toughness,” presented at The 11th International Conference on Solid-State Sensors and

Actuators, Munich, Germany, 2001.

References

53

[28] J. D. Zahn, D. Trebotich, and D. Liepmann, “Microfabricated Microdialysis Microneedles for

Continous Medical Monitoring,” presented at 1st Annual Interantional IEEE-EMBS Special Topic

Conference on Microtechnologies in Medicine&Biology, Lyon, France, 2000.

[29] P. K. Campbel, K. E. Jones, R. J. Huber, K. W. Horch, and R. A. Normann, “A silicon-based,

three-dimensional neural interface: manufacturing processes for an intracortical electrode array,” IEEE

Transactions on Biomedical Engineering, vol. 38 , pp. 758-, 1991.

[30] Q. Bai, K. D. Wise, and D. J. Anderson, “A High-Yield Microassembly Structure for Three-

Diomensional Microelectrode Arrays,” IEEE Transactions on Biomedical Engineering, vol. 47, pp.

281-289, 2000.

[31] J. Chen and K. D. Wise, “A Silicon Probe with Integrated Microheaters for Thermal Marking

and Monitoring of Neural Tissue,” IEEE Transactions on Biomedical Engineering, vol. 44, pp. 770-

774, 1997.

[32] T. E. Bell, K. D. Wise, and D. J. andersson, “A flexible micromachined electrode array for

cochlear prosthesis,” Sensors and Actuators A: Physical, vol. 66, pp. 63-69, 1998.

[33] P. Vettiger, G. Cross, M. Despont, U. Drechsler, U. Dürig, W. Häberle, W. P. King, M. I.

Lutwyche, H. Rothuizen, R. Stutz, R. Widmer, and G. K. Binnig, “The "Millipede" - More than one

Thousand Tips for Parallel and Dense AFM Data Storage,” presented at Solid-State Circuits

Conference, 2000.

[34] M. Madou, Fundamentals of Microfabrication: CRC Press, 1997.

[35] H. Elderstig and O. Larsson, “Polymeric MST - high precision at low cost,” Journal of

Micromechanics and Microengineering, vol. 7, pp. 89-92, 1997.

[36] Y. Xia and G. M. Whitesides, “Soft Lithography,” Annual Reviews in Materials Science, vol.

28, pp. 153-184, 1998.

[37] B. H. Weigl, R. Bardell, T. Schulte, F. Battrell, and J. Hayenga, “Design and Rapid Prototyping

of Thin-Film Laminate-Based Microfluidic Devices,” Biomedical Microdevices, vol. 3, pp. 267-274,

2001.

[38] H. Lorenz, M. Despont, N. Farni, N. LaBianca, P. Renaud, and P. Vettiger, “SU-8: a low-cost

negative resist for MEMS,” Journal of Micromechanics and Microengineering, vol. 7, pp. 121-124,

1997.

[39] F. Lärmer and P. Schilp, “Method of anisotropically etching silicon,” in German Patent No.

4,241,045. Germany, 1994.

[40] E. H. P. Klaassen, K.; Noworolski, J.M.; Logan, J.; Maluf, N.I.; Brown, J.; Storment, C.;

McCullcy, W.; Kovacs, G.T.A., “Silicon Fusion Bonding and Deep Reactive Ion Etching a New

Technology for Microstructures,” presented at Solid-State Sensors and Actuators, 1995 and

Eurosensors IX.. Transducers '95. The 8th International Conference on, Stockholm, Sweden, 1995.

[41] H. Jansen, M. d. Boer, B. Otter, and M. Elwenspoek, “The Black Silicon Method IV: The

Fabrication of Three Dimensional structures in Silicon with high Aspect Ratios for Scanning Probe


54

Microscopy and Other Applications,” presented at IEEE Micro Electro Mechanical Systems,

Amsterdam, the Nederlands, 1995.

[42] A. A. Ayón, C. C. Lin, R. A. Braff, and M. A. Schmidt, “Etching Characteristics and Profile

Control in a Time Multiplexed Inductively Coupled Plasma Etcher,” presented at Solid State Sensor

and Actuator Workshop, Hilton Head Islan, South Carolina, 1998.

[43] G. L. Wilkes, I. A. Brown, and R. H. Wildnauer, “The Biomechanical Properties of Skin,” CRC

Critical Reviews in Bioengineering, vol. August, pp. 453-495, 1973.

[44] J. Banchereau and R. M. Steinman, “Dendritic cells and the control of immunity,” Nature, vol.

392, pp. 245-252, 1998.

[45] W. M. Sams, “Structure and Function of the Skin,” in Principles and Practice of Dermatology,

W. M. Sams and P. J. Lynch, Eds.: Churchill Livingstone, 1996.

[46] J. G. Webster, Medical Instrumentation, Application and Design. New York: John Wiley &

Sons, 1998.

[47] B. H. Brown, R. H. Smallwood, D. C. Barber, P. V. Lawford, and D. R. Hose, Medical Physics

and Biomedical Engineering. London: Institute of Physics Publishing, 1999.

[48] H. Viertiö-Oja, P. Meriläinen, M. Särkelä, P. Talja, H. Tolvanen-Laakso, and A. Yli-Hankala,

“Spectral Entropy, Approximate Entropy, Complexity, Fractal Spectrum, and Bispectrum of EEG

During Anesthesia,” presented at International Conference on Memory, Awarness and Consciousness,

New York, 2001.

[49] R. F. Schmidt, Grundriss der Neurophysiologie: Springer Verlag Berlin-Heidelberg-New York.

[50] R. S. C. Cobbold, “Transducers for Biomedical Measurements: Principles and Applications,” in

Biomedical Engineering and Health Systems: Wiley-Interscience Series.

[51] C. Gondran, E. Siebert, P. Fabry, E. Novakov, and P. Y. Gumery, “Non-polarisable dry

electrode based on NASICON ceramic,” Medical & Biological Engineering & Computing, vol. 33, pp.

452-457, 1995.

[52] A. C. Metting van Rijn, A. P. Kuiper, T. E. Dankers, and C. A. Grimpergen, “Low-cost Active

Electrode Improves the Resolution in Biopotential Recordings,” presented at 18th Int. Conf. of the

IEEE Engineering in Medicine and Biology Society, Amsterdam, 1996.

[53] B. Alizadeh-Taheri, R. L. Smith, and R. Knight, “An Active, Microfabricated, Scalp Electrode-

Array for EEG Recording,” presented at The 8th International Conference on Solid-State Sensors and

Acutators, and Eurosensors IX, 1995.

[54] C. Fendrock, “Disposable, pre-gelled, self-prepping electrode,” . USA, 1994.

[55] A. D. Legatt, “Impairement of Common Mode Rejection by Mismatched Electrode Impedances:

Quantitative Analysis,” American Journal of EEG Technology, vol. 35, pp. 296-302, 1995.

[56] L. A. Geddes, L. E. Baker, and A. G. Moore, “Optimum Electrolytic Chloriding of Silver

Electrodes,” Med & biol Engng, vol. 7, pp. 49-56, 1968.

[57] J. R. Macdonald, “Impedance Spectroscopy,” Annals of Biomedical Engineering, vol. 20, pp.

References

55

289-305, 1992.

[58] W. M. Saltzman, Drug Delivery: Engineering Principles for Drug Therapy: Oxford University

Press, 2001.

[59] M. R. Prausnitz, “Reversible Skin Permeabilization for Transdermal Delivery of

Macromolecules,” Critical Reviews in Therapeutic Drug Carrier Systems, vol. 14, pp. 455-483, 1997.

[60] T. Maeder, “Down with the Bad, Up with the Good,” Scientific American, pp. 22-23, 2002.

[61] F. O. Nestle, J. Banchereau, and D. Hart, “Dendritic Cells: On the Move from Bench to

Bedside,” Nature Medicine, vol. 7, pp. 761-765, 2001.

[62] B. W. Barry, “Novel mechanisms and devices ro enable successful transdermal drug delivery,”

European Journal of Pharmaceutical Sciences, vol. 14, pp. 100-114, 2001.

[63] P. Helgesson, “Nåldyna ger smärtfria injektioner,” in Dabens Nyheter. Stockholm, 2002.

[64] “Nyhetsmorgon”, Swedish Television, TV4, 2002.

[65] “Direkt”, Swedish Television, TV3, 2002.

[66] S. Babiuk, M. Baca-estrada, L. A. Babiuk, C. Ewen, and M. Foldvari, “Cutaneous vaccintation:

the skin as an immunologically active tissue and the challenge on antigen delivery,” Journal of

Controlled Release, vol. 66, pp. 199-214, 2000.

[67] S. Böhm, W. Olthuis, and P. Bergveld, “An integrated micromachined electrochemical pump

and dosing system,” Journal of Biomedical Microdevices, vol. 1, pp. 121-130, 1999.

[68] J. T. Santini, A. C. Richards, R. scheidt, M. J. Cima, and R. Langer, “Microchips as Controlled

Drug-Delivery Devices,” Angewandte Chemie International Edition, vol. 39, pp. 2396-2407, 2000.

[69] A. v. d. Berg and T. S. J. Lammerink, “Micro Total Analysis Systems: microfluidic aspects,

Integration Concept and Applications,” Topics in Current Chemistry, vol. 194, pp. 21-48, 1998.

[70] S. Shoji and M. Esashi, “Microflow Devices and Systems,” Journal of Micromechanics and

Microengineering, vol. 4, 1994.

[71] H. Andersson, W. v. d. Wijngaard, P. Nilsson, P. Enoksson, and G. Stemme, “A Valve-Less

Diffuser Micropump for Microfluidic Analytical Systems,” Sensors and Actuators: B Chemical, vol.

72, pp. 259-265, 2001.

[72] W. E. Morf, O. T. Guenat, and N. F. deRooij, “Partial electroosmotic pumping in complex

capillary sytems Part 1: Principles and general theoretical approach,” Sensors and Actuators B:

Chemical, vol. 72, pp. 266-272, 2001.

[73] C. S. Effenhauser, H. Harttig, and P. Krämer, “A Disposable Micropump Concept for Small and

Constant Flow Rates,” presented at Micro Total Analysis Systems, 2001.

[74] H. Andersson, P. Griss, and G. Stemme, “Expandable microspheres - surface immobilization

techniques,” Sensors and Actuators B: Chemical, vol. Accepted for publication, 2002.

[75] M. Dole, L. L. Mach, R. L. Hines, R. C. Mobley, L. D. Ferguson, and M. B. Alice, “Molecular

Beams of Macroions,” Journal of Chemical Physics, vol. 49, pp. 2240-2247, 1968.

[76] G. I. Taylor, “Disintegration of Water Drops in an Electric Field,” Proceedings of The Royal


56

Society London, vol. A280, pp. 383-397, 1964.

[77] R. B. Cole, Electrospray ionization mass spectrometry. New York: Wiley-Interscience, 1997.

[78] J. V. Iribarne and B. A. Thomson, “On the evaporation of small ions from charged droplets,”

Journal of Chemical Physics, vol. 64, pp. 2287-2294, 1976.

[79] M. S. Wilm and M. Mann, “Electrospray and Taylor-Cone theory, Dole's beam of

macromolecules at last?,” International journal of Mass Spectrometry and Ion Processes, vol. 136, pp.

167--180, 1994.

[80] A. Dove, “Proteomics: translating genomics into products?,” Nature Biotechnology, vol. 17, pp.

233-236, 1999.

[81] A. Pandey and M. Mann, “Proteomics to study genes and genomes,” Nature, vol. 405, pp. 837-

846, 2000.

[82] M. Yamashita and J. B. Fenn, “Electrospray Ion Source. Another Variation of the Free-Jet

Theme,” J. Phys. Chem.,, vol. 88, pp. 4451-4459, 1984.

[83] M. Karas, D. Bachmann, U. Bahr, and F. Hillenkamp, “Matrix-assisted ultraviolet laser

desorpton of non-volatile compounds,” Int. J. Mass Spectrom. Ion Processes, vol. 78, pp. 53-68, 1987.

[84] R. D. Oleschuk and D. J. Harrison, “Analytical microdevices for mass spectrometry,” Trends in

analytical chemistry, vol. 19, pp. 379-388, 2000.

[85] P. F. Man, C. H. Mastrangelo, M. A. Burns, and D. T. Burke, “Microfabricated Capillarity-

Driven Stop valve and Sample Injector,” presented at MEMS98, Heidelber, 1998.

[86] J. Sjödahl, P. Griss, J. Melin, Å. Emmer, J. Roeraade, and G. Stemme, “Characterization of

micromachined Hollow Tips for 2D Nanoelectrospray Massspectrometry,” presented at the 15th

International Symposium on Microscale Separations and Analysis, HPCE 2002, Stockholm, Sweden.

Micromachined Interfaces for Medical and Biochemical ...9144/FULLTEXT01.pdf · Micromachined...

Documents

Transcript of Micromachined Interfaces for Medical and Biochemical ...9144/FULLTEXT01.pdf · Micromachined...