structural analysis of ultrasharp silicon microneedles TU Ilmenau by liaoqinmei



                                                       Conor O’Mahony

                 Microsystems Centre, Tyndall National Institute, University College Cork, Cork, Ireland.

    Abstract — Emerging microneedle-based devices             silicon wafers. The wafer is then etched in a 29% w/v
have applications in transdermal therapeutic deliv-           aqueous KOH solution until mask undercut takes place,
ery, sensing and diagnostics. This paper assesses the         the eight resulting planes intersect and a needle shape is
failure mechanisms and reliability of wet-etched,             formed [3]. This needle is comprised of eight {263}
ultrasharp silicon microneedles. We believe that the          planes, a base of {212} planes and has a height:base
primary failure mode of these microneedles when               diameter aspect ratio of 3:2. Tip sharpness is a function
subjected to compressive forces is due to shearing            of mask and crystal axis alignment accuracy; closely
along the relatively weak {111} silicon plane. Howev-         aligned tip radii are generally of the order of 50-100 nm.
er, standard microneedle characterisation tests
neglect the interaction of needle tips with compliant
tissue, and we show that, under realistic usage
conditions, reliability of these microneedles is in fact
significantly higher than predicted using standard
mechanical analysis.

   Keywords : microneedle, KOH, reliability

   I - Introduction

    In recent years there has been considerable interest
in the development of microneedle technologies to
overcome the skin’s stratum corneum barrier in a
minimally invasive manner [1, 2]. To date, a number of
groups have investigated the applicability of our wet-
etched silicon microneedle technology to drug and                   Figure 1: Solid, 277 m tall silicon microneedle.
vaccine delivery, photodynamic therapy, DNA delivery,
electroporation, diagnostics and physiological signal            III - Experimental
monitoring. This paper assesses the structural robust-
ness and reliability of these silicon microneedles.           A. Structural analysis under compressive forces
    Two primary biomechanical parameters apply to the
safe and efficient design of microneedles; these are (a)          Microneedle failure under compressive forces was
the force required to insert the microneedle into skin,       assessed using an Instron 5565 computer-controlled
and (b) the force at which the structural integrity of the    force-displacement station. A steel rod was pressed
microneedle fails. In this paper, we measure skin inser-      against a single 300m tall microneedle at a rate of 1
tion forces, assess the structural properties of individual   m/s until the desired maximum load was reached.
silicon microneedles, and show that compressive failure
of this particular microneedle design occurs due to shear     B. In-vivo skin insertion force measurement
forces acting along the relatively weak {111} crystal
plane. By comparing these results with those obtained             Microneedle insertion tests were performed by using
from qualitative reliability tests in in-vivo human skin,     the force-displacement tool to press microneedle arrays
we demonstrate that such basic mechanical tests may           into the volar forearm of a 36-year-old male volunteer at
significantly underestimate safety factors for this type of   a rate of 1 mm/s until the desired load was reached. In
needle due to the brittleness of the material and the         order to assess penetration efficacy, approximately
unrepresentative nature of the interaction of a rigid         200μl of 2% w/v methylene blue dye was applied to the
surface with the needle tips. Finally, we show that the       site after removal of the array and left for 5 minutes
distribution of forces over the needle tip and conical        before excess dye was removed under running water.
structure during skin penetration leads to a very high        The site was examined using a stereozoom microscope.
degree of reliability.                                            Additionally, to mimic ‘real-world’ use, arrays of
                                                              microneedles were pressed into the volar forearm using
   II - Microneedle Fabrication                               moderate finger pressure and held in position using a
                                                              swaying motion for approximately ten seconds. Follow-
   Square oxide/nitride hardmasks were aligned to the         ing removal, 100 randomly selected microneedles were
<110> direction on 100 mm diameter <100> oriented             inspected for damage using an optical microscope.
   IV - Results                                                                                                                        Fn
A. Compressive force modelling & analysis                                                                                               Fs
   All microneedles were subsequently imaged using
scanning electron microscopy (SEM). Varying degrees
of damage are visible on every needle, Figure 2. The
damage has been characterised by measuring the size of
the broken particle ls, which we defined as the length
from the apex of the microneedle to a point defined as
the intersection of the longitudinal axis of the
microneedle and the broken plane. This parameter was
used in subsequent analytical analyses of tip failure.

                                                                    Figure 3: Resolution of forces acting on the weak {111}
                                                               silicon plane.
                                                                  The sheared area A is equal to .(a.b)/4. Using trig-
                                                               onometric relations, the angle of the needle sidewalls
                                                               (71.5o) and the angle of the sheared {111} plane (54.7o),

                                                                                                      A  1.03l s 
                                                                                                                              ;                                (1)

                    5m                                                                               l s  0.787        .                                     (2)
    Figure 2: Damage to needle after application of 0.5N to
the tip. The size of the broken particle ls is also defined.
                                                                   This dependence of shear particle size on compres-
    A closer inspection of the microneedles reveals that       sive force has been measured and is shown in Figure 4.
the damaged needle tips show a characteristic sloped
                                                                Sheared Particle Size, ls [microns]

shape, indicating that failure takes place along the                                                  60

relatively weak {111} plane, which has an angle of
54.7o to the horizontal, and which is a preferred cleav-
age plane in single-crystal silicon [4]. The lack of                                                  40
regular, sharp discontinuities in the force-displacement
curves also indicates that failure takes place through a                                              30
gradual ‘grinding’ mechanism of these planes, rather
than a sudden, catastrophic failure of the structure.                                                 20

    We therefore believe that these microneedles fail
due to the presence of a shearing component of the
compressive force along this weak {111} plane. This is                                                 0
shown schematically in Figure 3. The vertical compres-                                                       0         5          10        15       20   25
sive force, F, exerted on the microneedle may be re-                                                                       Applied Force [N]
solved into two vector components – one acting normal
to, and the other parallel to, the {111} plane. The
                                                                                                      Figure 4: Measured and analytically estimated shear par-
magnitude of the shearing component is given by Fs =                                                                      ticle sizes.
Fsin(54.7o), which will cause failure of the needle at a
point where the cross-sectional area of the fractured              The data has been fitted with Eqn. (1) using G =
{111} plane is A = Fs/G, where G is the shear strength         6.9GPa, a value that compares well with those reported
of the material.                                               elsewhere [5]. This simple model provides a useful
    In order to model the height of the sheared particle ls    estimate of the degree of damage likely to occur when a
as a function of applied compressive force F, we assume        compressive force is applied to the needle tip. It does,
that the shear area created by the intersection of the         however, become inaccurate at low forces, where the
{111} plane with the octagonal microneedle forms an            surface-to-volume ratio of the sheared particle becomes
ellipse of area A that is characterised by minor axis a        large and surface defects due to tip misalignment
and major axis b.                                              artifacts and the presence of rapidly converging plane
                                                               edges provide an increased probability of failure.
B. In-vivo insertion force analysis                                      inspected using an optical microscope at a magnifica-
                                                                         tion of 90X. No fracture or failure of any needle was
    Methylene blue dye is used as a skin staining agent                  observed.
and has previously been used to indicate the site of
microneedle penetration [6]. In this study, staining                        V - Discussion
clearly indicated the presence of conduits formed as a
result of microneedle insertion. For each test site, the                     These mechanical tests, commonly used in structural
number of conduits formed was recorded as a function                     analyses of microneedles, show that wet-etched silicon
of maximum application force per needle, and repre-                      needles subjected to compressive stress fail due to
sentative results are shown in Figs. 5-6 (averaged over                  continuous shearing along the relatively weak {111}
four test sites).                                                        plane, and that some damage occurs even at very low
                                                                         forces due to the ultrasharp nature of the needle tips.
                                                                         However, it is also clear that all of our in-vivo (human
                                                                         and animal) tests show that these conical needles are
                                                                         extremely reliable and that failure is very rare when
                                                                         applied to tissue.
                                                                             The discrepancy is due to the extremely high stress-
                                                                         es generated when nanometer-scale microneedle tips are
                                                                         pressed against a rigid surface, which easily exceed
                                                                         material fracture strengths. In reality, this type of simple
                                                                         rigid-body interaction does not occur when
                                                                         microneedles are used in practice. Instead, compliant
                                                                         skin tends to conform to the shape of the tip before,
                                                                         during and after insertion [7]. A simple model assumes
                                                                         that tissue conforms perfectly to the needle shape during
                                                                         the insertion process, that the needle-skin interfacial
    Figure 5: Methylene blue-stained in-vivo human skin after            area, AI, is equal to the octagonal cross-sectional area of
array application at 20 mN per needle. The spacing between
                                                                         the needle, and can be defined as a function of array
dye dots is 1 mm.
                                                                         displacement after initial skin contact s as

                          100                                                                              2
                                                                                              s       
                                                                             AI  2.828                ; s<h                   (3)
  Array penetration (%)

                                                                                        tan( 71.5 o ) 

                          70                                                 AI  Ab ; s ≥ h                                     (4)

                          60                                             where Ab is the area of the array baseplate associated
                                                                         with each needle, and is equal to 1 mm2 for this case.
                                                                         Note that for this particular shape, AI increases rapidly
                          40                                             as the needles are inserted into the skin.
                                0   10    20     30    40     50    60
                                    Applied force per needle (mN)

    Figure 6: Array penetration as a function of applied force
per needle for 81-needle arrays.
    Over 50% penetration has been achieved at forces as
low as 5 mN per needle, and over 90% penetration is
                                                                                     s    s
observed for forces of over 15 mN (1.2 N per array).
These figures are in close agreement with other work on
ultrasharp microneedles [7]. 100% penetration has never
been recorded; this is because at least one of the 81
needles on the array usually falls close to or on a hair                    Figure 7: Microneedle insertion model.
follicle and this prevents needle insertion at that point.
                                                                             To experimentally assess the practical effects of this
C. In-vivo reliability                                                   interaction, an array of 81 x 300m tall needles was
                                                                         pressed into human skin using the controlled force-
    Both before and after the procedure detailed in Sec-                 displacement setup described earlier. The resultant
tion III-B, 100 randomly selected microneedles were
stress acting on each needle was calculated as  =                                   instead our models show that the degree of failure is a
F(s)/AI, where F(s) is the experimentally measured                                   continuous function of applied force.
force-displacement relationship. This is graphically                                     Microneedles are commonly characterised by press-
illustrated in Figure 8, where the sharp drop in stress at s                         ing the structure against a hard surface such that a load
= 300m represents the contact of the base of the                                    is applied to the vertical axis of the needle. Although
microneedle array with the skin surface, i.e. s = h. This                            such compressive tests are reasonably accurate for
occurs even before penetration takes place, significantly                            needles with relatively flat or blunt tips, homogeneous
relieving stress on the needle itself at an early stage.                             materials and /or constant columnar cross-sectional
                                                                                     area, they are inappropriate for characterisation of
                            25                                 1.E+8                 needles of this type, where the very low interfacial area
 Force per needle, F [mN]

                                                                                     of the ultrasharp tips mean that extremely high pressures
                            20                                                       are exerted on the tips and the brittle nature of silicon
                                                               1.E+6                 means that corresponding fractures are easily attained at

                                                                       Stress [Pa]
                            15                                                       almost any force.
                                                                                         It has previously been shown that tissue tends to
                            10                                                       conform to the shape of the microneedle tip, and the
                                                                                     conical nature of these structures causes the
                            5                                                        microneedle-skin interfacial area to increase as skin
                                                                                     insertion takes place. Both conditions lead to a rapid
                            0                                  1.E+2                 decrease in compressive stress acting on the needle-skin
                                 0           2           4                           interfacial area as the microneedle penetrates into the
                                     Insertion Depth, s [mm]                         skin, and in addition this stress suddenly and signifi-
                                                                                     cantly drops further as the base of the array contacts the
    Figure 8: Measured stress as a function of needle dis-                           skin. Hence, the possibility of microneedle failure
placement.                                                                           actually decreases as skin insertion progresses.
                                                                                         This means that reliability of these microneedles,
    Even this rather basic model (which neglects com-                                when used in practical applications, is significantly
plex skin deformations and puncture mechanics [7]                                    higher than predicted using standard compression tests.
explains the high degree of reliability observed when                                In fact, repeated insertion into in-vivo human skin has
using these needles. Firstly, the maximum compressive                                failed to result in damage to arrays of microneedles, and
stress (<0.1GPa) is much less than that required to cause                            this agrees with our experience when using silicon
significant shear failure along the {111} plane. Second-                             microneedles for a wide range of in-vivo tests over a
ly, due to the conical shape of the needles, this stress                             significant timespan.
actually decreases during the initial stages of needle-                                  The results confirm the safety and reliability of sili-
skin engagement, reducing the likelihood of failure.                                 con microneedles for use in a wide range of biomedical
Once s ≥ h, the stress decreases sharply still further as                            applications.
the base of the needle comes into contact with the skin,
and is now several orders of magnitude below that                                    References
needed to cause failure. The final phase of the plot
shows an increase in stress as elastic deformation of the                            [1] R. Pettis, Therapeutic Delivery, 3(3), pp. 357-371,
skin occurs, although stress values are still well below                                 2012.
failure strengths.                                                                   [2] K. van der Maaden, W. Jiskoot and J. A.
    Insertion forces for these needles are comparable to                                 Bouwstra, J. Control. Release, 161(2), pp. 645-55,
the lowest reported in the field [7]; over 90% of the                                    2012.
needles on 81-needle arrays penetrate skin at forces of                              [3] N. Wilke, M. L. Reed and A. Morrissey, J.
20 mN per needle. These very low forces increase                                         Micromech. Microeng., 16, pp. 808-814, 2006.
reliability and also allow array designers to increase the                           [4] D. C. Miller, B. L. Boyce, M. T. Dugger, T. E.
number of needles on an array, thereby leading to                                        Buchheit, K. Gall, Sensor. Actuat. A, 138(1), pp.
increased skin permeability and device efficacy.                                         130–144, 2007.
                                                                                     [5] R. F. Cook, J. Mater. Sci, 41(3), pp. 841-872,
VI – Conclusion                                                                          2006.
                                                                                     [6] M. Haq, E. Smith, D. John, M. Kalavala, C.
    This paper has analysed the failure mechanisms and                                   Edwards, A. Anstey, A. Morrissey and J. Birchall,
reliability   of    wet-etched,     ultrasharp     silicon                               Biomed. Microdevices 11, pp.35-47, 2009.
microneedles. Using standard force-displacement                                      [7] N. Roxhed, T. C. Gasser, P. Griss, G. A. Holzapfel
methods, we show that the primary failure mode of                                        and G. Stemme, J. Microelectromech. Syst., 16,
these microneedles when subjected to compressive                                         pp. 1429–1440, 2007.
forces is by shearing along the relatively weak {111}
silicon plane. No single failure point is identified, and

To top