INTRODUCTION

INTRODUCTION
PHC5 28/11/2005 6:15 PM Page 80
5
Laser Sources for Confocal Microscopy
Enrico Gratton and Martin J. vandeVen
INTRODUCTION
Laser assisted confocal microscopy has made a lot of progress over
the past few years. Laser systems have become more modular and
compact. There is an ever-increasing number of available laser
excitation lines as well as an improvement in user friendliness
and ease of use. At the same time, expansion of web resources
has provided easy access to a wealth of information. Our goal is
both to aid the experienced and novice microscopist in quickly
locating and sorting through the relevant laser information and to
provide a means of avoiding common problems and pitfalls in the
use of laser excitation in the various fluorescence techniques
such as fluorescence correlation spectroscopy (FCS), fluorescence
lifetime imaging microscopy (FLIM), fluorescence loss in photobleaching (FLIP), fluorescence recovery after photobleaching
(FRAP), optical coherence tomography (OCT), second harmonic
generation (SHG), single molecule detection (SMD), and single
particle tracking (SPT). In this chapter we describe the characteristic properties of a number of lasers commonly used in fluorescence microscopy. We concentrate on the characteristics of lasers
in relation to their use as an illumination source for microscopy.
Compared to other sources emitting electro-magnetic radiation,
such as hot filaments, arc lamps, and light-emitting diodes (LEDs),
lasers have a number of unique properties, which make them an
almost ideal light source for use in confocal microscopy. These
properties are:
• high degree of monochromaticity
• small divergence angle
• high brightness
• high degree of spatial and temporal coherence
• plane polarized emission (for many types)
• a Gaussian beam profile (in some cases this requires special
optics).
In the 40 years since the realization of the first experimental
laser, a wide and still rapidly expanding variety of lasers has been
developed. Currently very rapid development of miniaturized,
easy-to-use, tunable “pocket” lasers is taking place. These small
convenient lasers are in the process of replacing many of the large
laser systems still in use.
Available laser systems cover an extremely wide range, differing from each other in physical size, principle of operation, and
optical, temporal, and mechanical properties, such as beam diver-
gence, output power, polarization properties, duty cycle, stability
of the beam, and vibration sensitivity. These characteristics are
related to the mechanical design, emission wavelengths and tunability, ease of operation, maintenance costs, reliability, and safety
aspects. This chapter introduces the microscopist to the operation
of the laser, the most important laser parameters, their influence
on the quality of the confocal image, and methods to create wavelength-tunable light sources. In addition, laser systems for second
harmonic generation and optical tweezers are described.
LASER POWER REQUIREMENTS
First, we need an order of magnitude estimate of the emission
intensity that can be obtained in fluorescence microscopy using
1 mW of input light. The amount of laser power needed depends
crucially on the quantum efficiency of the contrast medium being
studied. The most common contrast factors are sample fluorescence and backscatter.
It is convenient to express the quantities in terms of photons/
(s * pixel * mW) of incident light at a given wavelength because
the intrinsic dark noise of modern detectors is often specified in
similar units. Also, expressing the flux per pixel provides a quantity that is independent of the illuminated area. The following are
useful relationships:
-19
• Energy of one photon: hn = hc/l = 4 ¥ 10 J at l = 500 nm
• 1 mW of light intensity at 500 nm represents 2 ¥ 10
+15
photons/s
On a widefield image of 1000 ¥ 1000 pixels, 1 mW of incident
light, uniformly distributed, is equivalent to
• flux per pixel = 2 ¥ 10
9
photons/(s * pixel * mW) at 500 nm
There are two considerations. First, how many photons will
be emitted per pixel? Secondly, how many photons can be tolerated per pixel before saturation of the fluorescent molecules
occurs?
Let us analyze the first question. Using fluorescein, one of the
most common probes, the molar extinction coefficient is about
100,000/cm of optical path. Assuming an effective optical path of
about 1 mm (the depth of field), the molar extinction is about 10.
The local concentration of fluorescein can vary, depending on the
spatial location and the degree of labeling. Assuming that a
Enrico Gratton • University of Illinois at Urbana-Champaign, Department of Physics, Laboratory for Fluorescence Dynamics, Urbana, Illinois
Martin J. vandeVen • Department of Cell Physiology, Microfluorimetry Section, Biomedical Institute, Hasselt University and Trans national University
Limburg, and Institute for Materials Research IMO/IMOMEC, Diepenbeek, Belgium
80
Handbook of Biological Confocal Microscopy, third edition, edited by James B. Pawley, SpringerScience+Business Media, New York, 2006.
PHC5 28/11/2005 6:15 PM Page 81
Laser Sources for Confocal Microscopy • Chapter 5
concentration of 10-5M is reasonable, the optical density (OD) of
a 1 mm path length is ª 10-4. The number of photons absorbed is
then
photons absorbed = ( flux per pixel) ¥ (OD) =
2 ¥ 105 ( s * pixel * mW ) at 500 nm.
Assuming a quantum yield of 0.8 and a collection efficiency
of 10%, the detector receives
photons at the detector = 1.6 ¥ 10 4
(s * pixel * mW of incident light).
Given the quantum efficiency for a good detector (10% at
500 nm), the final detected photon flux should be about
flux detected = 1600 photons (s * pixel * mW of light).
This flux should be compared with the dark noise of most
detectors, which can vary between 10 and 100 equivalent
photons/(s * pixel). In our estimation, the only quantities that can
vary over a wide range are the power of the laser and the effective
concentration of the probe. Lasers can have up to watts of power
and the concentration of the probe can be higher than we have
assumed. The efficiency of detection is usually smaller than we
estimate and the noise can be larger. The purpose of our calculation is to give a rough idea of the kind of power that a laser must
furnish to be usable for fluorescence detection in confocal laser
scanning microscopy (CLSM). Tsien and Waggoner (Chapter 16,
this volume) find an optimal power with the best signal-to-noise
ratio (S/N) with respect to autofluorescence and elastic and inelastic scattering of 76 mW at 488 nm and 590 mW, as long as triplet
formation is neglected. Therefore, a laser power of 1 to 2 mW
spread over 106 pixels at the specimen position should be more
than sufficient for most applications. Effectively, 10 to 100 mW is
common in confocal. Assuming a 10% optical path efficiency a
laserhead output power of >~1 mW suffices.
There are two different types of saturation effects. One is
related to the number of molecules that can absorb light in a given
area for a certain incident flux. In a given pixel, assuming a volume
of 1 mm3, the volume is 10-15 L. At a molar concentration of 10-5,
we should have approximately 6000 molecules/pixel. Since the
number of photons absorbed per milliwatt of incident light is about
2.5 ¥ 10+5/s on a single pixel in widefield, each molecule is excited
about 40 times per second. From the photophysical point of view,
the decay of fluorescein (and in general any singlet single state
decay) is very fast (4 ¥ 10-9 s), so that the ground state should be
repopulated very rapidly. However, in the confical microscope for
a pixel dwell time of about 1 ms, the 40 ¥ 4 ns = 160 ns dead time
represents 16% of the pixel period.
There are many possible photochemical processes that are
either irreversible or have a cycle time of several milliseconds to
seconds. In this latter case, even if the quantum yield for these
effects is very low (below 0.001), and the exposure time is on the
order of seconds, molecules lost to the long-lived state will
severely limit the overall peak excitation intensity that can be used
before the output looses its linear relationship with the input.
For quantitative microscopy this is the most important limitation.
Hess and Webb (2002) found that their FCS data implied a nonGaussian three-dimensional (3D) volume and distortion of the
calibration of the excitation volume at a power level of 10 to 100
mW at 488 nm, for one photon Rhodamine Green excitation and
5 to 10 mW at 980 nm, for the two-photon case (Rhodamine Green
or Alexa 488, Molecular Probes).
Having discussed the power requirements, we continue with
a concise description of the basic elements of a laser, its principle
81
of operation, and other important practical aspects, such as heat
removal and mechanical and optical stability.
In general, confocal microscopes work best at <1 mW of continuous wave (CW) beam power at the specimen and useful images
of living cells have been made with <100 nW (see Chapter 19, this
volume). When pulsed lasers are used for two-photon excitation
it is important for the pulse frequency to be high enough (80–
100 MHz) so that many pulses occur during the 1- to 4-ms pixel
dwell time. The intensity instability of this light when it reaches
the specimen must be small compared to the statistical uncertainty
that will occur when the signal photons are detected; if the signal
level is only 100 photons/pixel, then Poisson statistics will limit
accuracy to 10%, while 104 detected photons will yield 1%
accuracy.
THE BASIC LASER
The acronym laser stands for light amplification by stimulated
emission of radiation. Laser action, that is, the emission of coherent radiation, has been observed for a large number of different
media, but all lasers have several features in common (see
Fig. 5.1).
• Active laser medium: Atoms, molecules, and atomic ions in
•
•
pure or mixed gases, vapors, liquids, or solids, confined to the
gain medium volume where they can be excited by an external source of energy.
Excitation source: An external source of energy used to pump
the laser medium. This can be another laser, arc lamp or flash
lamp, electron beam, proton beam, electrical current, electrical discharge or radio frequency (RF) excitation, etc. The
choice of pump source is determined by the optical, thermal,
and mechanical properties of the active medium and the wavelength region of interest.
Optical resonator: The laser medium is enclosed between two
parallel mirrors, which form a Fabry–Perot interferometer. The
mirrors are placed at each end of the laser medium. One mirror,
the high reflector (HR), totally reflects (R = 99.9%); the other
mirror, the output coupler (OC), only partially reflects. To a
first approximation, the ratio of these two reflectivities is the
gain of the cavity that they form. For instance, if the reflectivity of the OC is 95%, on average a given photon will pass
through it 1 time in 20 and, in the absence of other loss mechanisms, the cavity gain will be 20. In small semiconductor
lasers the physical polished sides of the devices may act like
HR
LASER MEDIUM
+ PUMP SOURCE
OC
LASER BEAM
OPTICAL RESONATOR CAVITY
FIGURE 5.1. An optical resonator is formed between a highly reflective
mirror (high reflector, R = 99.99%), HR, and a mirror with a reflectivity of, for
example, 80%, the output coupler (OC). Within this resonator we find the active
lasing medium (a crystal, semiconductor, liquid or gas) and its energy supply:
the pump source, an arc lamp or flash lamp, another laser.
PHC5 28/11/2005 6:15 PM Page 82
82
Chapter 5 • E. Gratton and M.J. vandeVen
effective mirrors. Laser emission generated by electronic transitions that are relatively rare require cavities with higher gain.
FAST
PUMPING
Principle of Operation
A particle of the active laser gain medium becomes excited when
it absorbs pump energy and goes to an excited level [Fig. 5.2(A)].
It then returns to the ground state via non-radiative relaxation
processes and also by emission of radiation [Fig. 5.2(B)]. Under
normal conditions, a Boltzmann equilibrium describes the population of the various energy levels: the higher the energy level the
lower the population of that level. When an excited, metastable
level with a long lifetime exists in the laser medium, energy will
accumulate in this level. If the excitation is intense enough, the
Boltzmann distribution normally present will “invert” for the population of this metastable state (i.e., there will be many more electons in the excited state than Boltmann would predict). In the laser
cavity, photons emitted from this energy level will strongly interact with the population of the metastable level [Fig. 5.2(C)],
forcing it to release energy and return back to the lower level [Fig.
5.2(D)]. This process is called stimulated emission of radiation,
that is, the interaction of the light with the excited particle
increases the likelihood that the particle will return to the ground
state. The stimulated emitted light has a high degree of monochromaticity, because emission occurs from a well-defined transition. In addition, the photon that results from this stimulated
de-excitation process is in phase with the electromagnetic wave
traveling in the laser medium. As a result, the emitted radiation has
excellent spatial and temporal coherence and is highly directional.
A convenient way to let the electromagnetic radiation interact
with the laser medium is a resonant cavity (Fig. 5.1). At optical
wavelengths, this is achieved by a Fabry–Perot type interferometer. Two plane-parallel mirrors, one highly reflective, the other
semi-transparent, are separated by a distance equal to an integral
multiple of half the lasing wavelength. Because the electromagnetic radiation interacts repeatedly with the laser medium, this
resonant cavity increases the probability of stimulating deexcitation. It also provides the necessary feedback to make the
emission grow coherently. Because the emission of the first photon
going in precisely the correct direction to reflect back and forth in
the cavity is a low probability event, getting the laser to fire is a
chaotic phenomenon that exhibits a threshold effect; lasers won’t
start or work stably below a certain output power level.
FAST
FAST
FAST
PUMPING
SLOW
SLOW
LASER
EMISSION
LASER
EMISSION
FAST
A
3-STATE LASER
B
4-STATE LASER
FIGURE 5.3. Fast non-radiative relaxation processes couple an excited state
with a lower metastable energy level. Laser emission occurs in the gain medium
from the metastable state. Dependent on the gain medium properties this laser
process is described by a three-level or four-level lasing scheme.
So far we have only discussed a three-level laser, that is,
ground-state, upper excited-state, and lower excited-state [Fig.
5.3(A)]. In a four-level laser [Fig 5.3(B)], as in the helium-neon
(He-Ne) laser, the population inversion can be obtained more
easily (Arecchi and Schultz-Dubois, 1972). The titanium–sapphire
(Ti : Sa) vibronic laser is effectively also a four-level laser due to
its broad energy bands (see later section). Other improvements
relate to the replacement of the Fabry–Perot mirrors by corner cube
reflectors or crossed-roof prisms to increase mechanical and
thermal stability. For further information, see Arecchi and SchultzDubois, 1972; Stitch, 1979; Bertolotti, 1983; Bass and Stitch,
1985; Kudryashov and Weber, 1999; Webb and Jones, 2004;
Hodgson and Weber, 2005.
An interesting approach is the increasing use of optical fibers
that act as both lasing medium and cavity at the same time; a very
compact design is described later in this chapter.
Pumping Power Requirements
In order to sustain laser action, the gain of the optical resonator
needs to be larger than the losses due to resonator walls and other
optical elements. The minimum necessary pumping power, P, is
proportional to n3. This means that, as one shifts from the infrared
(IR) through the visible (VIS) towards the ultra violet (UV), an
ever increasing amount of energy is needed to obtain laser action.
This limits the possible pumping mechanisms.
Laser Modes: Longitudinal (Axial) and Transverse
• Axial or longitudinal modes: Separated by a distance, L, the
GROUND
STATE
EXCITED
STATE
SIMULATED
EMISSION
LIGHT
AMPLIFICATION
ABSORPTION
SPONTANEOUS
EMISSION
METASTABLE
STATE
STIMULATED
COHERENT
EMISSION
A
B
C
D
FIGURE 5.2. Optical (de)excitation processes. (A) Transition from ground
state to excitated state upon absorption of a photon, an electron moves to a
more outward shell, timescale < femtoseconds. (B) Relaxation to a lower level,
for example, ground state under spontaneous emission of a photon, electron
returns to a lower level, timescale nanoseconds. (C) Light driven interaction
inside laser gain medium of a photon with electrons in excited metastable state
with a long lifetime (D) Stimulated coherent emission and light amplification.
two plane-parallel mirrors of the Fabry–Perot interferometer
cavity form an optical resonator. This separation distance can
be long (meters, as in big frame ion lasers) or very small
(micrometers, as in very compact diode lasers). A number of
standing wave patterns each consisting of an integer multiple,
m, of half wavelengths, l/2, exists in an optical cavity of length
L: m = L/(l/2). The frequency, n, for the mth vibration along
the long axis of the laser is, therefore, n = mc/2 L, where c is
the speed of light in the laser cavity. The frequency spacing
between adjacent longitudinal modes is c/2 L, that is, the
inverse of the laser cavity round trip time. A very large number
of longitudinal modes can exist in the laser cavity unless
bandwidth-limiting devices such as Fabry–Perot etalons are
installed in the cavity.
PHC5 28/11/2005 6:15 PM Page 83
Laser Sources for Confocal Microscopy • Chapter 5
• Transverse modes: These modes vibrate perpendicular to the
long axis of the laser and are known as transverse electromagnetic modes, TEMmn, where m and n are integers describing the nodal points in the two directions perpendicular to the
laser axis. For each transverse mode, many longitudinal modes
can exist. (For an in-depth derivation, see laser handbooks by
Arecchi and Schultz-Dubois, 1972; Stitch, 1979; Demtröder;
1996; Bass and Stitch; 1985; Hecht and Zajac, 2003; Silfvast,
2004). An adjustable diaphragm or pinhole inside the laser
cavity is sometimes used to select the TEM00 lasing mode and
control its intensity. No transverse modes exist in single (longitudinal) mode fiber lasers.
The TEM00 mode is desired for most light microscopy experiments. It has a Gaussian beam profile with no phase shifts across
the beam. The maximum intensity is in the center of the beam, it
has complete spatial coherence, and the smallest possible beam
divergence. Only this mode can be focused to a diffraction-limited
spot. Donut-shaped transverse modes TEM*01 (TEM10 in overlap
with TEM01), such as produced by some helium–cadmium
(He-Cd) lasers operating at 325 nm, are not desirable because they
possess no intensity in the center of the beam. In this case, spatial
filtering is necessary (see later section).
Polarization
The output of many lasers is linearly polarized with the polarization vector vertical.
• Lasers with Brewster surfaces: A convenient and inexpen-
sive way to minimize reflection losses and to generate linearly
polarized light is the installation of a Brewster surface (i.e.,
one that is tilted so that the normal of the plane and the incoming beam form a specific angle, qBrewster) at the end of the resonator. Horizontally polarized light incident on such a
plane-parallel plate will be completely reflected. Vertically
polarized light will be completely transmitted without any
reflection losses (Fig. 5.4). In laser resonator cavities, this plate
is usually a quartz or a fused-silica plate. In solid-state lasers,
the semiconductor rod itself is sometimes cut at the Brewster
angle to minimize reflection losses. In gas lasers, the exit
B
HR
B
GAIN MEDIUM
ION LASER
BREWSTER
WINDOW
OC
LASER
BEAM
BREWSTER
WINDOW
FIGURE 5.4. Emission of linearly (vertically) polarized laser light from an
ion laser resonator cavity equipped with Brewster angle plasma-tube windows.
Only vertically polarized light is amplified. It experiences no reflection losses
at the Brewster angle: BBrewster = tan-1n, with n the refractive index of the pure
crystalline quartz window. Horizontally polarized light suffers reflection losses
and stays below the lasing threshold. Windows normal to the optical axis of
the laser would introduce reflection losses and could prevent the laser from
operating at all or introduce a set of reduced length resonator cavities. Tungsten disks or BeO tubes with a central bore with or without focusing magnetic
fields keep the lasing plasma volume centered in the cavity. Gas return paths
are also provided in the disks. HR, highly refelecting mirror; OC, output coupling mirror.
•
83
windows are usually oriented at the Brewster angle to obtain
vertically polarized light. Dust often collects on Brewster
windows that point upward, damaging the coating. To operate
properly, they must be kept absolutely clean using a dust cover
and a slightly positive dry air or nitrogen pressure in the laser
head. In the jet-stream dye laser, the jet is placed at Brewster
angle in order to minimize losses due to reflection of the
pumping beam, thereby maximizing the pumping efficiency.
For the same reason, the tuning element of the dye laser is also
placed at Brewster angle. Brewster surfaces can also be the
origin of very dangerous reflections. Eye protection should
always be worn when protective covers are removed from a
laser.
Lasers built without Brewster windows will still show some
polarized output due to birefringence within several optical
components. However, the plane of polarization may change
with time and, even though the total intensity stays constant,
should this light pass through a polarizer, large output intensity fluctuations may be observed.
Randomly polarized beams contain two orthogonal components but the phase and ratio of the two components vary rapidly
in time, creating polarization noise. Dichroic mirrors, and, in fact,
any mirror or lens surface, not at right angles with the incoming
radiation will reflect vertically and horizontally polarized light differently (directions are taken with respect to the plane of incidence
formed by the reflected and transmitted beams).
A convenient way to depolarize laser emission is to install a
polarizer or polarizing beam-splitter and a l/4-wave plate. When
placed at an angle of 45° with respect to the incoming linear polarization, this 1/4-wave plate converts linearly polarized light into
circularly polarized light. These waveplates are usually designed
for a specific wavelength. Achromatic retarders, such as Fresnel
rhombs, are sometimes preferred, but are quite expensive (Driscoll
and Vaughan, 1977; Hecht and Zajac, 2003). Another advantage
of this arrangement in optical microscopy is that it prevents
reflected or backscattered light from reaching the detector. The
reflected light passes through the 1/4-wave plate in such a way that,
on return, it is blocked by the input polarizer because its plane
polarization is now orthogonal to the main transmission direction
of the input polarizer. Phase randomizing will briefly be discussed
in a later section. For an in depth treatment see Chapter 31 by
Harris and Delaney. In fiber lasers, optical isolators are inserted
for the same reason, that is, to prevent backreflections. The
polarization-independent isolator consists of a small cylindrical
package containing a fiber pigtail, non-spherical collimating
lenslet, a slightly wedged birefringent platelet, a magneto-optic
Faraday rotator at 45°, exit birefringent platelet also at 45°, and
refocusing lenslet followed by another fiber pigtail.
Coherent Properties of Laser Light
Laser beams illuminate objects coherently. The process of
stimulated emission imposes coherence effects on the emitted laser
light waves. All parts of the electromagnetic wave reach a crest
that is, are in phase, at a given point in space at the same time.
Spatial and temporal coherence is present. This coherence
stretches for a certain distance and time and depends on the spectral width or pulse duration of the laser light. Beyond a certain distance or time interval synchronization differences arise. When a
choice exists, a short coherence length laser should be used for
CLSM or scrambling devices should be inserted into the optical
path.
PHC5 28/11/2005 6:15 PM Page 84
84
Chapter 5 • E. Gratton and M.J. vandeVen
• Temporal coherence: The coherence time is the time interval
•
•
•
•
during which light traveling over a certain distance maintains
a phase difference of less than p (or 180°), tcoh. = 1/Dn.
Coherence length: The path traveled by the wave during the
coherence time is called the coherence length, Lc ~ c/Dw, where
Dw is the spectral width of the laser. Longitudinal coherence
length Lcoh. = c/Dn. Lateral coherence length, lcoh. = l/f, with f
the beam divergence. For a typical gas laser, for example, a
He-Ne laser with a bandwidth of 1.5 GHz, this leads to Lcoh of
~20 cm. In single-mode operation with Dw ~ 1 MHz, Lcoh is
~50 m. A diode laser with a spectral width of 0.1 nm at 780 nm
possesses a coherence length of 1 mm. Dye lasers equipped
with tuning elements, diode lasers, and other lasers such as the
Ti : Sa systems, usually have bandwidths on the order of tens
of gigahertz and an Lcoh of a few millimeters or less. Depending on the type of line narrowing element installed, Lcoh can
be increased by a factor 103 to 106. For practical confocal
microscopy, a shorter coherence length is preferred to eliminate the influence from out-of-focus defects. For example, dust
on semi-transparent surfaces, lenses and mirrors creates interference fringes when a laser with long coherence length is used
(Hell et al., 1991). (See Tables 6.1 and 6.2 which contain
columns with spectral width and pulse length, from which the
coherence length follows.)
Spatial coherence: Spatial coherence occurs when a constant,
time-independent, phase difference exists for the optical field
amplitude over a given distance at two different points in
space.
Coherence surface: The coherence surface is the region of
space for which the absolute phase difference of the optical
field is less than p. The well-known “speckle” pattern consisting of a pattern of darker and brighter light spots is visible
when a matte wall surface is illuminated by a distant He-Ne
laser. The interference effects caused by irregularities in the
surface will be noticeable.
Coherence volume: The coherence volume is the product of
coherence length and the coherence surface. Interference
between superimposed coherent waves will only be visible
within this volume.
Phase Randomization: Scrambling
the Coherence Properties of Laser Light
A long coherence length of lasers will cause laser speckles (i.e.,
interference effects) and scatter from out-of-focus particles and
these will interfere with the image. Normally in a fluorescence
CLSM, this scattered light is removed by the dichroic mirror as
the emitted fluorescence has a different wavelength than the illuminating light source. However, a polarizing beam-splitter and l/4wave retardation plate can be equally effective for the removal of
light backscattered by dust or optical surfaces. The angular position of this 1/4-wave plate and the polarizer is important
(Szarowski et al., 1992). For generating circularly polarized light,
it is only necessary that the incoming plane of polarization and the
principal axis of the 1/4-wave plate be at a 45° angle. Light
reflected or scattered from an illuminated dielectric surface has its
direction of rotation reversed. On its return passage through the
plate, the light becomes linearly polarized again, but its plane of
polarization is now rotated by 90° and it is removed from the
optical path by the polarizing beam-splitter.
Another method to reduce speckle comes from color confocal
microscopy where one uses several lasers. As the coherence prop-
erties of the various laser light sources are not the same, averaging over the three wavelengths and over several planes will reduce
the effect of the laser speckle (Cogwell et al., 1992b).
Measures to Reduce the Coherence Length of
Laser Light
Many applications work better and image quality improves if the
coherence of the laser light is reduced. One way to achieve this is
to place a rotating diffuser wedge in front of the beam expander
(Hard et al., 1977). Rotation or vibration of this wedge is essential to average out local diffuser properties. Alternatively, focusing
the laser beam into a multi-mode optical fiber bounces the light
around inside the fiber and mixes the various propagation modes
to scramble the Gaussian intensity pattern (TEM00) to a multimode pattern with a homogeneous spot intensity. This eliminates
or reduces the coherence surface. A short focal length lens or
graded index lens (GRIN) recollimates the light at the fiber
exit.
To eliminate the speckle pattern due to temporal coherence
properties, a piezo-electric driver can be used to induce vibrations
in a section of the fiber (Ellis, 1979). In the context of medical
imaging, Connor, Davenport and Gmitro (1992) used a rotating
diffuser to eliminate the speckle pattern by coupling the laser beam
into a bundle of small, multi-mode fibers. The latest method in this
area is the introduction of ultra-thin single-mode image fibers. A
fiber-optic bundle with a total diameter of 200 to 500 mm consisting of 2000 to 5000 micro image fibers, each with a core size of
1.0 to 1.4 mm (Kiat et al., 1992), can be used as a phase randomizer in conjunction with a rotating diffuser. Although these ultrathin fiber bundles propagate shorter wavelengths much more
strongly than longer, leaky ones, this is not a problem as long as
monochromatic laser light is used.
As light is scattered in all directions by most diffusers, they
reduce the intensity of the laser light. However, holographic diffusers have been designed to carry strong forward intensity lobes.
Among others, Lasiris, Inc. and Physical Optics Corp. (see Table
5.3) produce holographic laser beam homogenizers, having a
conversion efficiency of 80% to 90% and low backscatter (Bains,
1993).
Laser coherence effects can also be eliminated by spot scanning the field of view with a tiny, focused, single-mode laser beam.
Alternatively, the condenser plane can be scanned with the same
tiny, focused, single-mode laser beam while illuminating the whole
field of view. Coherence scrambling occurs because the laser spot
rapidly scans the back-focal plane and the field is continuously
illuminated with a mix of beam angles (Ellis, 1988).
Heat Removal
Most of the laser excitation energy is converted into heat. This
must be removed to prevent thermal destruction of the active laser
medium. Small laser systems can use convective air cooling, but
larger laser systems need forced-air or water cooling. Especially
for the largest systems, fans and turbulent water-flows may introduce vibrations in the system (microphonics). These unwanted
mechanical vibrations are inevitably coupled to the mirror mounts
of the resonator cavity, causing increased noise in the optical
output. Pumps that recycle liquid laser-media are another source
for vibration. In this case, the vibration is transferred via hoses to
the active medium in the resonator. To minimize these effects, the
hoses should be clamped or fastened to the support table as near
as possible to the laser head.
PHC5 28/11/2005 6:15 PM Page 85
Laser Sources for Confocal Microscopy • Chapter 5
The heat generated will also put thermal stress on the mechanical parts of the resonator. Poorly designed laser systems quickly
loose their proper alignment or may need an unacceptably long
warm-up time. The installation of a laser system in a room with
large daily or annual temperature fluctuations or a ceiling fan continuously blowing cold, dusty, air directly onto a laser will also
hamper operation and cause performance to deteriorate. The use
of thermo-electrically (TEC) cooled diode lasers that are in
thermal contact with their enclosures may create hot surfaces
that should be labeled as such. Diode arrays may require water
cooling. Because of their monolithic design, they are less susceptible to mechanical instabilities but thermal stress has to be avoided
at all cost. The same is true for fiber lasers, which, due to the fiber
length, are sensitive to temperature fluctuations unless counter
measures such as negeative temperature coefficient materials are
used.
Other Installation Requirements
Manufacturers usually describe the electrical power requirements
and flow rates for cooling water (Gibson, 1988, 1989; Rapp, 1988).
Before installation of a laser system, adequate control of room temperature and air conditioning should be available. Walls should be
painted to eliminate concrete dust from settling on delicate optical
surfaces. For emergency situations, a master switch for the laser
power should be installed with easy access at eye height and
labeled as such. Heat from large fan-cooled laser power supplies
can sometimes be plumbed directly into the intake of the air conditioning system, with backflow protection, if this is allowed. Similarly, acoustic noise can be very tiring and noisy power supplies
should be placed outside the experiment room.
An exhaust for noxious fumes such as ozone or dye vapor
should exist and backup systems should be established to prevent
interruptions of power or coolant flow. Mechanical vibrations due
to nearby traffic can be eliminated by installing the system on commercially available vibration-free, air-cushioned laser tables or on
tables isolated using sand-filled containers, tennis balls, or motorcycle inner tubes. When equipment used for radio-frequency
crystal-growth, nuclear magnetic resonance (NMR) equipment,
building air-conditioning machines, large elevators, or other large
laser systems are present, it is often wise to equip each laser with
its own stabilized power line to prevent large voltage spikes from
reaching laser power supplies. Hidden cables drawing a large
current (for elevators, for example) in ducts near your facility may
ruin a carefully planned sensitive system by disruptive induction
currents. Stray magnetic fields may affect the flow of ions in gas
lasers.
Movement of fiber-optic components changes the beam propagation properties of the fiber (bundle) and should be prevented
for best stability (unless the device is used intentionally to
scramble the laser light polarization and reduce the coherence
length).
Attenuation of Laser Beams
Laser damage levels for materials are usually given in mW/cm2
while laser output is given as total average power (mW) for a
small, for example, 2-mm diameter beam. Although intensities in
the milliwatt range can be attenuated with neutral density filters,
high-power lasers will easily destroy this type of filter. The
absorbed light may even bleach or heat them so much that they
fracture or explode. Polaroid material can be used only for low
light intensities and but this material usually has a peak transmis-
85
sion of only 30% to 50%. For intensities above about 10 mW, better
attenuators are Glan–Thompson polarizers. When a laser emits
only polarized light, one rotatable polarizer, set at a proper angle
will suffice. For randomly polarized lasers a set of two polarizers
can be used. The first one passes 50% of the total intensity. The
second polarizing analyzer transmits a continuously adjustable
amount of light depending on its orientation angle. In all these
cases one must be careful to ensure that light reflected from the
various crystal surfaces is also absorbed by beam stops. Initial
alignment should be done with only the lowest laser power available. If a polarizer is used to attenuate an initially randomlypolarized laser beam, fluctuations in the intensity may be observed.
These are due the varying nature of the interaction between
the cavity modes and therefore the polarization state inside the
laser.
Glan–Thompson polarizers contain Canada balsam to hold
them together. This material may absorb slightly, specially in the
UV. Power levels above several watts (power = energy/time = area
of pulse) will destroy this type of polarizer. For the highest
expected power levels, air-spaced Glan–Taylor polarizers are
strongly recommended. A new type of polarizing element is the
Microwire™ polarizer commercialized by Moxtek Inc. (see
Table 5.3). It fits in very tight spaces but is somewhat more
fragile.
STABILIZATION OF INTENSITY, WAVELENGTH,
AND BEAM POSITION IN LASERS
Sources of Noise in Lasers
Stability, both temporal and positional (beam pointing), is a very
important aspect of any CLSM laser light source. As beam pointing affects the amount of light coupled into the delivery fiber,
pointing instability can cause intensity instability. Other intensity
instabilities are related to backreflections from the fiber that can
be avoided (Cogwell et al., 1992a) by employing a wedged fiber
input (e.g., Point Source, see Table 5.3). Figure 5.5 gives examples
of noise in the output of several laser systems. The most stable are
the diode laser and the forced-air cooled (fan) small-frame
argon-ion laser. The argon-ion laser-pumped titanium : sapphire
laser output was coupled into a fiber-optic attached to the detector. The average relative intensity fluctuations are defined as
N-1 S i(|Ii - ·IÒ| /·IÒ), where N is the number of points. For the diode
laser this quantity is about 0.1% (Franceschini et al., 1994), and
for the Ti : Sapphire laser at the fiber output, about 2%. Both would
be acceptable for microcopical applications in which less than
50 photons were detected per pixel.
It is imperative to keep lasers well-aligned with very clean
windows. Misaligned mirrors, a half-illuminated reference monitor
diode, or a dust speck may create a very unstable output. As misalignment can cause unusual modes to appear, these laser lines will
fluctuate independently, and the total output becomes unstable. For
instance, the two UV lines of an old water-cooled argon-ion laser
were observed to alternate between bright and dim with a period
of a few seconds to a minute. Realignment helped but the tube had
to be replaced. Transverse mode swapping is seen in He-Cd tubes
that need replacement.
Noise in the optical output of a laser can be created by a variety
of phenomena. Several factors produce high-frequency (noise) or
slow variations (drift) in beam power. These include power-supply
variations, thermal drift of the cavity, and mode competition.
Examples include:
PHC5 28/11/2005 6:15 PM Page 86
86
Chapter 5 • E. Gratton and M.J. vandeVen
• Solid-state lasers: The important noise sources are microphonic noise and 1/f noise related to thermal fluctuations in the
lasing rod. Furthermore the Ti : Sa rod temperature bath should
be set at the correct temperature, for example, 18°C. Failing
to do so creates condensation droplets on the pump
surface. Fiber lasers should be kept in a stable temperature
environment.
A
B
C
0
100
200
300
TIME (SEC.)
FIGURE 5.5. Measured laser emission stability (arbitrary units). Normalized
and shifted intensities for (A) a diode laser (Sony SLD104AV, l = 780 nm; top
curve); (B) a forced-air cooled small frame argon-ion laser, Omnichrome model
532, running at 488 nm and 7 mW power (center curve); (C) Mira™ (Coherent) Ti:Sa laser running at 780 nm and 1.2 W (bottom curve). The normalized
noise for the diode laser is 0.1% and for the Ti:Sa laser about 2%. The attenuated Ti:Sa laser was focused into a fiber-optic.
All lasers suffer from noise introduced by their powersupplies.
Switching-mode powersupplies introduce switching ripple, typically at tens of kilohertz. This type of power supply was developed to take advantage of better transformer efficiencies and
smaller physical sizes (Forrester, 1994a,b).
Planar optical elements and filters should be inserted at a small
angle with respect to the optical axis to prevent beam reflections
from returning to the laser cavity where they might increase the
noise level of the laser (Cogswell et al., 1992a).
Images created on commercial CLSM instruments may show
a fine pattern of irregular vertical lines. Possible sources include
resonances in the scanning mirror systems and high frequency
ripple in the switching power supplies. However, sometimes the
problem is not the light source but the detector system (French and
Gratton, personal communication). Last but not least, the main
laboratory power supply lines may be unstable.
Other sources of noise are caused by external influences: traffic
vibrations, etc. A simple means to reduce its effect is a welldamped, stable support platform (see above).
Laser Beam Intensity Stabilization in
Current- or Power-Control Mode
• Dye lasers: Noise and drift in dye lasers can be caused by air
•
•
•
bubbles and inhomogeneity in the dye, the dye pump, and by
the laser pump source.
Gas lasers: A major source of noise comes from plasma oscillations and from microphonics generated by cooling water
turbulence. Forced-air cooling for smaller lasers can avoid
most microphonics, but the fan may introduce some new noise.
Commercial confocals often mount the lasers on a heavy stone
slab (Zeiss) or place them in a separate, shielded drawer
(Leica). Also any optical pump source can introduce noise.
He-Cd lasers: Strong plasma oscillation between 225 and 400
kHz may create an output variation of up to 12%.
Semiconductor or diode lasers: Semiconductor laser diodes
require a current source with the highest stability and the
lowest electronic noise (Malzahn, 2004). Furthermore, temperature stability is imperative for adequate intensity and
wavelength stability (Hodgson, 1994). Forward-biased semiconductor junctions are inherently thermally unstable; an
increase in current increases the temperature, which, in turn,
increases the current even more. Therefore, current supply
design is very important and must be suited to the specific
laser. Internal damage in laser diode junctions due to heat
buildup ages the device, resulting in a reduced output and slope
efficiency. 1/f noise is due to trapping of carriers in the device
(Mooradian, 1993). Drive-current fluctuations also degrade the
performance and will modulate the beam intensity. Mistreated
fiber-optics can also increase the amount of noise. On the other
hand, an imposed high frequency (MHz) modulation allows
the laser to be used for time-resolved and frequency-domain
imaging applications.
Continuous wave gas lasers can easily be intensity stabilized using
either tube current stabilization or external modulation of the light
intensity (Miller and Hoyt, 1986).
• Constant-current mode: An electronic feedback loop directly
•
controls the tube current and minimizes current drift. As this
system takes no account of how temperature may affect cavity
gain, best results are obtained after the laser temperature stabilizes.
Constant output power mode: The drive current is controlled
via a signal obtained from a built-in monitor diode. The
stabilization circuit typically consists of a beam-splitter–
photodiode assembly mounted behind the output coupler. The
monitor diode picks off a small amount of laser intensity.
Changes in intensity are compensated for by opposing changes
in the drive current. When not properly aligned, large intensity
fluctuations may result. Multi-line lasing depends on an interplay of different gain and mode patterns specially when the
bore holes in the internal BeO laser disks become larger
with time (Fig. 5.4). A constant output power supply in combination with an aging plasma tube will increase in tube
voltage and current supplied, damaging the tube further and
even more rapidly. Both effects are even stronger in mixed gas
lasers.
Even if the laser beam is stable, external noise can be introduced to the system by room dust, traffic vibrations, or movement
of fiber-optic components and may still adversely affect intensity
stability. The long-term stability of the largest argon-ion models is
better than 1% in this control mode. But this light control mode
does not work so well when multiple emission lines are monitored
simultaneously (Brelje et al., 1993).
PHC5 28/11/2005 6:15 PM Page 87
Laser Sources for Confocal Microscopy • Chapter 5
87
Laser Beam Intensity Stabilization with External
Pockels Cell Modulator
Diode Laser Intensity and Wavelength
Stabilization
Some manufacturers provide special accessories to improve the
intensity stability of CW lasers. These devices are external modulators incorporating a fast feedback system (Miller and Hoyt, 1986;
Miller, 1991; see Fig. 5.6), and, of course, they reduce the total
output power. Intensity fluctuations of up to 50% of the maximum
power can be corrected in this way, but only with a 50% reduction
in the total available power. Laser intensity stabilizers can regulate the output of CW and mode-locked lasers to within 0.025%,
with a noise attenuation of 400 : 1 and a bandwidth from direct
current (DC) to several megahertz in some systems.
Depending on the polarization state of the laser light, up to
80% transmission is obtained. In this system (Fig. 5.6), a photodiode is illuminated with a fraction of the laser light deflected by
a beam-splitter. A servo-control unit compares the detector signal
with a user-selected set point. The difference signal is amplified
and drives an electro-optic (Pockels cell) modulator. The amplified electrical signal causes a rotation of the plane of polarization.
Depending on the voltage applied, more or less light can pass
through the modulator creating a variable beam attenuator. Ideally
the laser beam should be stabilized as close as possible to the
sample position rather than at the exit of the laser head. Otherwise,
vibrations or dust may alter the stability of the beam after it leaves
the source.
Stabilization units should be able to operate at a variety of
wavelengths with UV capability when necessary. They should be
protected against driver signals beyond a certain maximum, so that
a runaway situation does not occur, for example, if the beam is
blocked or temporarily interrupted. When the stabilizer is driven
by an external control signal, complex intensity sequences may
illuminate the sample. Commercial laser power controller (LPC)
stabilizers can be obtained from BEOC and several other manufacturers (see Table 5.3).
When randomly-polarized laser light has to be stabilized, the
external Pockels cell modulator must be placed between two
crossed polarizers. Unfortunately, such a modulator can not work
with low-repetition-rate pulsed lasers. Another simple way to
remove laser noise is similar to that typically used in steady-state
fluorescence spectroscopy. The effect of source fluctuations is cancelled by dividing the detected signal from the sample by a signal
derived from the laser sampling monitor detector.
Incorporating a Fiber Bragg Reflector (FBR) or Grating (FBG)
in a (pump) diode laser stabilizes wavelength, improves coupling
efficiency, reduces unwanted reflections and the effects of
temperature and drive current fluctuations, as well as simplifying
manufacturing (Guy and Painchaud, 2004). The FBR consists of a
stack of thin coating layers forming a bandpass reflectivity filter at
either the high reflector or output coupler end. Even simpler is lens
coupling the diode-laser output into a section of FBG fiber pigtail.
A periodic or non-periodic (chirped) refractive index variation of
the Bragg grating can be present in its core, creating a narrow band
low-reflectivity filter that reflects a small fraction of the light from
the diode laser back into it. However, due to interference effects,
only a certain wavelength — the designed lasing wavelength — is
actually fed back into the laser diode. All other wavelengths are out
of phase and die off. In this way, the attachment of the proper FBG
fiber pigtail creates a diode laser that is stabilized in wavelength
and intensity. Naturally, if the FBG temperature or internal strain
changes, the lasing wavelength will vary slightly. A typical temperature coefficient is 0.012 nm/°C. A temperature increase from
18°C to 40°C will cause the lasing wavelength to shift 0.26 nm,
but in practice it may well be 5-fold larger. With temperature compensating packaging and a strain-inducing material with a negative
temperature coefficient, this effect can be reduced to 0.001 nm/°C.
NOISE REDUCTION AND
INTENSITY STABILIZATION
CIRCUIT
LASER HEAD
FEEDBACK LOOP
ELECTROPOLARIZER
OPTIC
MODULATOR
PHOTODIODE
BEAM
SPLITTER
FIGURE 5.6. Example of a laser beam intensity stabilizer and laser noise suppressor. A beam-splitter delivers a fraction of the laser output beam to a fast
photodiode detector. Electronic feedback loops for low- and high-frequency
signals control the electro-optic modulator (EOM) intensity regulator. The
rotated plane of polarization exiting from the EOM passes through an angleadjustable Glan-Taylor polarizer. It defines a single plane of linear polarization. A detected intensity fluctuation alters the drive voltage to the crystal.
Beam Pointing Improvement Via Active Laser
Cavity Stabilization
The mirror orientation changes when the laser cavity warms up.
This causes the laser beam to wander and the beam intensity to
vary. Active cavity stabilization corrects for misalignment of the
resonator via a feedback mechanism (Peuse, 1988). The advantage
is that the resonator structure becomes independent of changes in
the environment, for example, temperature, and as a result provides an extremely short warm-up time (seconds), enabling handsoff operation. In addition, there is less chance that setting the
power supply for a constant optical output will provide damaging
power levels to a tube that has low gain because it is misaligned.
Large frame ion lasers currently come standard with active resonator stabilization systems such as PowerTrack™, ModeTrack™
and ModeTune™ (Coherent) or BeamLok™ Z-Lok™ (for modehop-free single-frequency operation) and J-Lok™ (for jitter reduction; Spectra Physics). These systems result in 1-min warm-up
times, hands-off operation, and long-term intensity stability of
better than 1%. Beam-pointing stability is obtained using a quadrant photodiode detector.
The horizontal and vertical positions of the high reflecting
mirror are continuously tuned for optimum laser power with independent (magnetic or piezo) actuators. A small oscillation (dither
signal) is superimposed on both actuators. A sensor photodiode
signal reflects the power level of the beam. The filtered dither
signal (lock-in detection) creates vertical and horizontal error
signals. This oscillation is compared with a microprocessor-stored
reference. The horizontal and vertical difference signals drive the
mirror positions independently to obtain maximum power. The
minimum detected oscillation (dither) signal corresponds with
optimum alignment (Miller, 1991).
Beam Delivery and Positioning
Fiber-optic Coupling
Continous wave laser beams can be delivered to the scan head or
microscope frame via single-mode, polarization-preserving, dual-
PHC5 28/11/2005 6:15 PM Page 88
88
Chapter 5 • E. Gratton and M.J. vandeVen
circular stress-rod design, fiber-optic (the so-called “Panda-eye
pattern type,” see Chapter 26, this volume). One should use fiberoptics whenever laser light has to be transported, eye-safe, over
large distances, especially when the laser is located in a dusty environment (i.e., not a clean room). A single-mode fiber core typically
has a diameter of 3 to 6 mm for VIS wavelengths. This creates a
point-like excitation spot for confocal microscopy. Multi-mode
fibers with core diameters ranging from 0.1 to 1 mm may also be
used at times but beam characteristics such as beam width and
polarization state are altered. However, fibers have been found less
suitable for carrying UV light or for preserving the pulse width of
femtosecond pulsed lasers. For femtosecond pulse lasers, the beam
should be enclosed in a dust-free tube and reflected from dispersion controlling mirrors (e.g., from Newport Corp). Picosecond
systems are less affected (see Chapter 26, this volume). Handy
beam couplers are available from many suppliers (OZ Optics Ltd.,
see Table 5.3). The Cell-viZio from Mauna Kea Technologies (see
Table 5.3) uses a bundle of 30,000 fibers of 2 mm each.
Ultrafast pulses can currently be delivered by fiber-optic
without dispersion using two methods: by pre-compensation or by
photonic-bandgap fibers. Ultrafast pulses create non-linear effects
in glass fibers and the fibers suffer from dispersion effects. Even
at low pulse energies, a non-linear Brillouin and Raman background is generated. Pulse trains can only be transported without
detrimental effects from chromatic dispersion under special conditions. By precompensating for the expected fiber pulse chirp with
a dual prism or grating configuration (Fork et al., 1984; Zeek
et al., 1999), the initial shape of the pulse can be preserved after
it exits the fiber, but this adds to the complexity of the system and
creates power loss. This precompensation uses the fact that a short
pulse consists of many frequencies described by a Fourier sum,
with proper amplitudes and phases.
In holey fiber coupling, single-mode, hollow-core (= air-filled)
photonic-bandgap or crystal fibers (PCF), non-linear effects occur
at a much higher threshold than in glass. As long as the zerodispersion region of ~812 nm for the waveguide matches the laser
wavelength, a perfect femtosecond pulse is transmitted without
distortion (Tai et al., 2004). Tuning outside this range gives pulse
broadening again (Hitz, 2004c). The spectral width of the 0 to
350 mW Ti : Sa laser stays unchanged at 812 nm after passing
through a 1.5-m length of HC-800-01 bandgap fiber (Crystal Fibre
A/S, see Table 5.3). Dunsby and colleagues (2004) describe an
electronically tunable white light continuum, 435 to 1150 nm,
generated by injecting 80 MHz 120 femtosecond light into a
micro-structured fiber.
Ultrafast Lasers Assuming the proper coatings, direct mirror
coupling into the scan head can maintain the femtosecond pulse
width and multi-photon intensity for ultrafast lasers such as Ti : Sa
systems (Fermann et al., 2002). On the other hand the dispersion
material used in acousto-optic modulator (AOM) or acousto-optic
deflector (AOD) devices or the glass in thick objectives may cause
inappropriate pulse broadening due to group velocity dispersion
(GVD). A GVD compensating prism or grating pair or a
Gires–Tournois interferometer (GTI) can control the GVD and
compensate the widening to some extent. Instead of prism and
grating pairs, chirped mirrors could be used. These limit tunability but improve stability against environmental effects such as
temperature changes and reduce high-frequency noise. In a chirped
mirror, the Bragg wavelength increases with increasing penetration depth. Double-chirped mirrors that produce an additional
chirp in the coupling of the incoming and reflected wave reach
almost perfect dispersion compensation.
Multi-photon ultrafast (100 fs) lasers possess a phase that
depends on the spectral region. To match the slightly different twophoton absorption properties of fluorophores in different environments one could tune the laser but this changes the beam properties
such as intensity, phase distortions, etc. A better approach is to use
a charge-coupled device (CCD) computer-assisted pulse shaper
either to create a transform-limited flat phase across the spectrum
or to modify the phases within the pulse via multi-photon intrapulse interference (MII), making selective excitation possible
(Dantus, 2003; Dela Cruz, 2004). MII can also limit the amount
of three-photon processes, thereby limiting deep UV damage to
living cells.
Direct Mirror Coupling and Pulse Width and
Pulse Shape Control
Intensity Control
Combinations of CW Lasers All commercial confocal systems
these days offer facilities to combine the output of several lasers
using dichroic mirrors and intensity-balancing neutral-density
filters. A device for selecting wavelength and intensity, such as an
acousto-optic tuning filter (AOTF) or acousto-optic beam splitter
(AOBS), is then added. One of the advantages of having many
separate lasers is that the intensity of each line is servo-controlled
independently. In the competing Kr/Ar ion laser, it is only the total
optical output that is stabilized. The portion of this total associated
with each individual line can vary widely with time. For example,
Cogswell and colleagues (1992a,b) combined a 632.8 nm He-Ne
with a 532 nm frequency-doubled Nd-YAG laser and the 442-nm
line from a He-Cd laser for true-color confocal reflection
microscopy. Issues that may arise are spectral purity and polarization of the excitation.
Wavelength Selection and Intensity Control
Wavelength Selection
In commercial confocal systems, several laser beams are often
combined via dichroic mirrors onto the same optical path leading
towards an AOTF or AOBS (Chapter 3, this volume). Both allow
very fast wavelength and intensity control on a microsecond
timescale. We found, however, that the spectral purity of the
AOTF–fiber combination is sometimes insufficient. When viewing
mirror-like solid surfaces covered with labeled biosensor molecules excited at 488 nm, a little bit of 514-nm laser light was
observed passing through the fiber-optic and masquerading as fluoroscein isothiocyanate (FITC) 509 nm emission. A thin 488 nm
interference filter placed in front of the fiber-optic entrance in the
AOTF module remedies this problem.
Continuous wave laser beams can be deflected with acousto-optic
devices/deflectors/modulators (AOD, AOM) for intensity and
position control. The functioning and practical use of these devices
has been discussed in great detail in Art and Goodman (1993).
Draaijer and Houpt (1988) used these devices for video-rate
scanning in confocal microscopy and their system was commercially available through Noran and still supported by Visitech Intl.
Ltd. and by Prairie Technologies, Inc. (see Table 5.3). Intensity
control via an electro-optic modulator is described in Chapter 3.
For use of polarizers to control laser intensity, see above under
Attenuation of Laser Beams.
Polarization of the Laser Light
We measured a degree of polarization of the light entering the
Zeiss AOFT at better than 1000 : 1. Using a compact home-built
polarization sensitive power meter at the sample position the laser
PHC5 28/11/2005 6:15 PM Page 89
Laser Sources for Confocal Microscopy • Chapter 5
light was still very well polarized, better than 200 : 1 with the principal axis 5° counterclockwise from the positive x-axis of the stage
(i.e., left to right direction).
89
LASER REQUIREMENTS FOR BIOLOGICAL
CLSM-RELATED TECHNIQUES
Optical Tweezers
SPATIAL BEAM CHARACTERISTICS
Although lasers provide very small beam divergence, the uniformity of the intensity across the beam may be relatively poor.
Fortunately, this is generally not a serious problem because the uniformity of illumination can be easily improved by spatial filtering.
Filtering is very efficient when there is good spatial coherence of
the beam.
Several devices are available for spatial filtering; the most
common is the Gaussian spatial filter (Melles Griot, see Table 5.3),
which consists of a focusing lens that converges the beam toward
a very small pinhole. After this pinhole, a second lens, one focal
length away, is used to regenerate a parallel beam. By choosing
the focal lengths of the two lenses carefully, this configuration can
also be used to expand or decrease the beam diameter but a pinhole
is always essential for spatial filtering. Spatial filtering takes place
because diffraction effects at the pinhole produce an out-going
wavefront, which is the Fourier transform of the pinhole and is not
affected by the spatial properties of the light impinging on the
pinhole (Fig. 5.7). It should be noted that dust particles moving in
the vicinity of the pinhole can modulate the transmitted intensity,
so great care must be taken to ensure that this does not occur. For
most practical purposes, a single-mode optical fiber functions as a
spatial filter.
For large size beams from, for example, excimer lasers, an
external beam homogenizer consisting of a prism/lens combination can be used. The optics recombine sections of the beam to
produce a homogeneous rectangular shape (Austin et al., 1989).
Edge-emitting diode lasers typically require extensive astigmatic and anamorphic corrective optics to obtain a circular, parallel beam. Commercial packages are available to implement this
(Ingeneric GmbH, see Table 5.3). However, the use of a spatial
filter, a single-mode fiber-optic, or an overfilled back aperture
plane can also produce a symmetric beam.
Single-cell manipulation is feasible by trapping the cell in an
optical box formed by one or more focused laser beams. Each focal
spot has a diameter of about 1 mm. Taken to the extreme, optical
trapping can take place in a completely enclosed environment.
Ashkin and colleagues (1987) reported the optical trapping of
particles and viruses although the technique was also used decades
earlier for levitation of particles in light-scattering experiments. A
simple design is given, for example, in Smith and colleagues
(1999). A beam power of 20 to 100 mW TEM00 at the specimen is
sufficient. For biological specimens infrared light should be used.
A careful choice of the trapping wavelength is essential to reduce
optical damage to biological samples as much as possible. Shorter
wavelengths are absorbed by the specimen, longer wavelengths by
water. Therefore, traps commonly use a Nd-YAG (Hoffmann et al.,
2000; Reichle et al., 2001) or a Nd-YLF 100 mW laser operating
at 1064 nm. Other useful lasers may be a 632.8 nm, 25 mW He-Ne
laser, or a diode laser beam. An improved design is given by
Bechhoefer and Wilson (2002). Fiber lasers with their diffractionlimited output and excellent beam quality are also very suitable for
laser tweezers applications (Woods, 2003).
Optical trapping of particles larger than the illumination wavelength can be explained with geometric optics. When a transparent spherical particle, having a refractive index larger than the
surrounding medium, refracts a ray of light, a change in momentum of the refracted photon occurs. Due to the principle of conservation of momentum, the particle receives an equal but opposite
change in momentum. This force pushes the particle away from
the direction of the refracted photon.
When a particle, somewhat smaller than the focal spot of the
laser beam, is suspended in the intensity gradient of a Gaussian
beam profile, then the intensity of the light on one side of the particle is larger than on its opposite side. This makes the particle
move to the side with the higher intensity (i.e., the center of the
laser beam) and it seems as if the laser beam attracts the particle.
Heating and Damage Effects in
the Cellular Environment
INCOMING LENS
BEAM
SPATIAL
FILTER
LENS
CLEANED-UP
LASER BEAM
FIGURE 5.7. Diagram of a spatial filter for by a pair of lenses and a small
pinhole. High frequency noise is removed form the laser beam. The spatial
intensity is cleaned up.
Cells attached to biosensor surfaces may suffer additional heating
effects because of one-sided contact with a buffer solution and
close contact with coated or absorbing semiconductor surfaces.
Objects caught in an optical trap are also subjected to more severe
heat stress. For instance, a 100 mW Nd-YVO4 laser focused into
watery media heats it by 0.8 K/100 mW, but, add a polystyrene
bead in glycerol, and it jumps to 4 K/100 mW. Even when this
amount of heat is not damaging to the cell, it influences the calibration of the trap, that is, the force–displacement relationship
(Boas, 2003; Peterman, 2003). Several studies report cell damage
conditions (König et al., 1996) of NIR multi-mode optical traps.
Neumann (1999) shows that for the spectral region from 790 to
1064 nm, an oxygen-mediated, one-photon process is causing the
damage. Wavelength and power level effects monitored over time
by Leitz and colleagues (2002), indicate that the 700 to 760 nm
spectral region is unsuitable due to induced photochemical effects,
but that 810 nm is better because only photothermal effects occur.
These can be controlled by reducing the trap energy to 36 mW. Cellular response to NIR light with varying pulse length is reported
by König and colleagues (1999 and Chapter 38, this volume).
PHC5 28/11/2005 6:15 PM Page 90
90
Chapter 5 • E. Gratton and M.J. vandeVen
Zhang and colleagues (1999) look at cell viability in CW Al : GaAs
diode laser traps.
Total Internal Reflection Microscopy
For imaging cell adhesion and other biological processes such
as vesicle transport and observing thin self-assembled layers
for biosensor applications, total internal reflection provides a
background-free fluorescent imaging technique. Penetration of the
evanescent wave beyond the boundary plane is limited to roughly
100 nm. Objects farther from the interface are not excited and stay
dark. Two methods exist to create total internation reflection fluorescence (TIRF) conditions. One is to inject the light through a
very high, ≥1.45, numerical aperature (NA) objective lens and the
other utilizes an external prism coupling to the slide on the far side
of the specimen. The through-the-objective lens method requires
only 40 mW CW at 488 nm for green fluorescent protein (GFP)
imaging with reduced bleaching effects. For the prism coupling
technique, 20 mW CW was barely sufficient and would have
benefitted from higher power. Other suitable lasers include 50 mW
green 532 nm; 70 mW He-Ne laser, or 200 mW 488 water-cooled
argon-ion lasers. It must be noted that, at these power levels, some
published configurations of the external-beam prism method can
be very hazardous to the eye.
tion rate lasers with 1 to 10 mJ pulse energies (Arrigoni, 2004a;
also see Chapter 38, this volume).
TYPES OF LASERS
The earliest lasers were solid-state lasers using ruby as the active
laser medium (Bertolotti, 1983). Subsequently, a wide variety of
lasers were developed (Weast and Tuve, 1971; Arecchi, 1972;
Brown, 1981; Bass and Stitch, 1985; Eden, 1988; Fermann et al.,
2002; Silfvast, 2004; Webb and Jones, 2004). Essentially, all CW
gas lasers (Bloom, 1968) and some solid-state lasers with emission in the visible part of the electromagnetic spectrum, meet the
minimum intensity requirements estimated above for fluorescence
microscopy. The list of available wavelengths continues to expand
(Weber, 1999). For trapping and label-free SHG microscopy, IR
wavelengths are the most suitable. Some other important parameters are the output power at each wavelength, efficiency, and stability. Table 5.1 lists the major types of CW lasers. Table 5.2 lists
the major options for pulsed-laser systems.
CONTINUOUS WAVE (CW) LASERS
Continuous wave (CW) lasers can be divided into several classes:
Confocal Raman CLSM for Chemical Imaging
Raman spectroscopy and imaging is used in research on immobilized molecules and biosensor interfaces. It requires a modehop-free tuning range of several gigahertz, a narrow line width of
1 MHz, and output power of several hundred milliwatts. Cellular
autofluorescence can present a large background. In order to
reduce its influence, red excitation from a diode laser is a good
choice. Stry and colleagues (2004) utilized a Tiger ECDL laser
system (see Table 5.2), with grating tuning from 775 to 785 nm,
and also from 730 to 1085 nm with 1 W output power.
Coherent anti-stokes Raman scattering (CARS), complementary to CLSM, maps via vibrational contrast the intracellular water
distribution. It uses the beat frequency of two lasers, preferably
operating in the IR to avoid damage to cells as much as possible.
Multi-photon lasers emitting picosecond pulses with widths of 2
to 5 ps are better for CARS because their spectral linewidth
matches the typical Raman bandwidth of ~20 cm-1 much better
than femtosecond pulses (Nan et al., 2004). They were able to
visualize the CH2 stretch vibrations from lipid tails (see also
Chapter 33, this volume).
Non-Linear Confocal Microscopy
Second harmonic imaging is a very good method to probe external membranes of living cells. Laser system considerations and
characteristics are given below under Wavelength Expansion Techniques. The SHG images show little speckle background and can
be created with about the same laser intensity as used for fluorescence imaging. One can track microtubules with SHG and intracellular inhomogeneities with THG (Hogan, 2004). Yang and
Mertz (2002) used, for example, a mode-locked 860 nm Ti : Sa
laser (see below) with a ~100 fs., 82 MHz optical pulse train that
delivered 10 mW at the sample (see also Chapter 40, this volume).
• gas lasers
• dye lasers
• solid-state lasers.
Gas Lasers
Three major types of CW gas lasers are available (Bloom, 1968):
• argon-ion, krypton-ion, and a mixture of argon and krypton
(mixed gas)
• helium–neon
• helium–cadmium
• alkali metallic vapor lasers.
The CO and CO2 lasers that emit around 5 mm and at 10.6 mm
are not discussed here. Commercially available, but not widely
used and therefore also omitted from further discussion, are the
neon laser that can emit 1 W of multi-line UV (CW) at 339.2
(0.3 W), 337.8 (0.2 W), and 332.4 nm (0.5 W) and the xenon laser
emitting between 488 and 540 nm but usually only operated in
pulsed mode. In general, krypton and helium–neon lasers are used
only if red excitation is necessary, but the latter also now provide
lines in yellow, green (GreNe), and orange (594 nm). With the
increasing availability of red-absorbing dyes, their importance is
growing. Several leading confocal microscope manufacturers still
equip their systems with red and green helium–neon lasers emitting a few milliwatts and air-cooled argon-ion lasers. Argon–
krypton mixed-gas ion lasers often form a cheaper alternative to
the purchase of separate argon and krypton systems plus the additional requirements for power and beam-steering optics. It is worth
realizing however, that the lifetime of a specific red laser line in a
mixed gas laser may be much less than the stated lifetime of the
strongest laser line (green, 514 nm).
Argon-ion
Nanosurgery and Microdissection
Multi-photon imaging systems can be used for multi-photon ablation of cellular structures and tissue. It requires high (kHz) repeti-
Because its emission wavelength matches the absorption peak of
fluorescein and other popular dyes, the argon-ion laser is still by
far the most common laser in microscopy. Lately, however, highly
efficient 488 nm small-footprint, solid-state lasers have appeared.
40–200 GHz
1/20 MHz
2 GHz
370–1100
400–850
700–950
20–40 l
60–100
500 GHz
mW
Spectral
Width
10–2000
200–1200
CW Av.
Power
2.5
2/h
1/3-day
rms, %
Noise
0.1
<0.005
<0.03
nm
Wavel.
Reset.
0.1
nm
Abs.
Acc.
<0.5
%
ASE
12–32
<25
%
Conv.
Eff.
100 Mhz/h
MHz/b
Freq.
Stab.
<1.3
0.6
mm
Diam.
<0.5
<1.5
1.5
mrad
TEM00
TEM00
Mode
Beam Parameters
Div.
V, >98
V
%
Pol.
(Continued)
Legends: Wavel(ength); CW: Continuous Wave; Av(erage), Power; Abs(olute) Acc(uracy); Wavel(ength) Reset(tability); ASE: Amplified Spontaneous Emission; Conv(ersion) Eff(iciency); Freq(uency) Stab(ility); Reset(ability);
Beam Parameters Diam(eter) (e-2 = 13.5% intensity level), Div(ergence) (full angle), Mode, Pol(arization).
1
Standing wave 3 mirror cavity dye laser, Invar bar temperature stabilized, Argon-ion and Verdi DPSS pump power 5 W UV or R, B, G, pump wavelength and output power depend on dye. 20–40, 60–100 and 500 GHz 1, 2 and
3 plate birefringent tuning, noise rated between 10 Hz and 100 kHz.
2
Combined dye + TI : Sa resonator design, single frequency ring dye laser, accommodates both low (<8 W) and high power (>10 W) pump lasers, LiO2 and KTP frequency doubling extends range down to 270 nm, Autoscan II
complete automation. Example of costs for a second hand 899 system, without Ti : Sa: ~ 12,000 USD.
3
Ti : Sa/Dye ring laser: Pumped with 308 nm excimer or 532 nm Nd:YAG; invar bar resonator, Linewidth for active/passive single frequency operation 1/20 MHz, broadband 3 plate birefringent filter 2 GHz, noise between 10 Hz
and 1 MHz, Labview and C++ software control.
5–15 W
560–650
Radiant Dyes
& CW Dye L.
Coherent 8992
Radiant Dyes3
& Dye
Ti:Sa
nm
mJ
390–930
Wavel.
Range
Pump
Power
Coherent 5991
Manufactory
& Model
Tunable CW Dye lasers
TABLE 5.1. Continuous Wave Lasers
PHC5 28/11/2005 6:15 PM Page 91
Laser Sources for Confocal Microscopy • Chapter 5
91
(nm)
(h) & (W)
(mW)
CW
Av. Pwr
(mW)
CW
Av. Pwr
>5,000 h
<31 A
1 phase
Coherent &
Enterprise II
–6104
Water-cooled
LASOS Laser
& Argon-ion5
SpectraPhysics
& Advantage6
163Caircooled
<20 A
<70 A/ph
3 phase
B
Large frame
Water-cooled
Sabre DBW253
Dual Brewster
Window (DBW)
or Tunable
Sealed Mirrors
(TBM)
457.9
488
514.5
ML VIS
454.5–
514.5
458
488
514
ML Red
ML white
647.1
676.4
752.5
ML B.G
MM B.G
457.9
465.8
472.2
476.5
488.0
496.5
501.7
514.5
520.8
528.7
530.9
568.2
ML-VIS
<40 A/ph.
3 phase
<55 A/ph
3 phase
ML UV
>5,000 &
300C-308
Small frame
Water-cooled
Coherent
Ion lasers1
Small frame
Water-cooled
70C Innova2
Series
40 mW
all lines
80
200]
350
1,000
40
300
5,000
6,000
300
70
100
600
1,500
600
350
2,000
70C-5
Argon
50
500
120
750
70C-K
Krypton
Gas lasers: Argon- and Krypton-Ion Pure Gas and Mixed Laser Systems
Wave-length
Life &
Power Cons.
30
2,500
250
130
150
250
130
100
250
30
70CSpectral
50
(mW)
CW
Av. Pwr
ML-VIS
351.1
363.8
MLUV
454.5
457.9
465.8
472.2
476.5
488.0
496.5
501.7
514.5
520.8
528.7
(nm)
Wave-length
8,000
8,000
550
300C308
250
250
750
140
560
180
240
950
2,400
950
480
3,200
(mW)
CW
Av. Pwr
454.5
457.9
465.8
472.2
476.5
488.0
496.5
501.7
514.5
528.7
ML
VIS
275.4
302.4
334.5
275.4–
305.5
300.3–
335.8
351.1
363.8
333.6–363.8
(nm)
Wave-length
25,000
800
1,500
800
1,300
3,000
8,000
3,000
1,800
10,000
1,800
7,000
3,000
1,800
1,700
1,600
180
380
500
Sabre
(mW)
CW
Av. Pwr
0.69
0.66
<1
<0.1
<1 pp
<1.3
<2.6
1.8 @*
514.5
1.5
mm
Diam.
<0.75
6 pp
0.2
(3 long
term)
current
reg.
0.5 light
reg.
(rms, %)
Noise &
Stabil.
<0.95
<1.05
<0.8
<0.8
0.4 @
514.5
0.5 (Ar)
0.8 (Kr)
mrad
Div.
TEM00
TEM00
TEM0
TEM0
TEM0
Mode
Pol.
V
500 : 1
V
100 : 1
V
100 : 1
Beam Parameters
<1.2
<1.3
<1.2
M2
Qual.
92
Manufactory
& Model
TABLE 5.1. Continuous Wave Lasers (Continued)
PHC5 28/11/2005 6:15 PM Page 92
Chapter 5 • E. Gratton and M.J. vandeVen
Wave-length
(nm)
Life &
Power Cons.
(h) & (W)
(mW)
CW
Av. Pwr
(mW)
CW
Av. Pwr
(mW)
CW
Av. Pwr
7
(nm)
Wave-length
>50,000
Melles Griot&
Green HeNe
Yellow HeNe
Orange HeNe
Lasos Lasert &
Red HeNe
632.8
543.5
591.1
611.9
RGB
325, 441.6
RG, UV
>10
1
1
1
20–50
20–50
30
(mW)
CW
Av. Pwr
2
(nm)
Wave-length
(mW)
CW
Av. Pwr
0.5%
<0.5
<0.5
(rms, %)
Noise &
Stabil.
0.7
0.88
0.75
0.88
1–2
mm
Diam.
<1.4
0.81
0.92
2.2
0.5–1
mrad
Div.
TEM00
Single/
Multi
Mode
Pol.
500 : 1
V
500 : 1
random
H
100 : 1
Beam Parameters
<1.2
M2
Qual.
(Continued)
Legends: Lftm. (Lifetime); Pwr = Power; Cons(umption); FWHM (Full Width Half Maximum); Beam Parameters Diam(eter) (e- = 13.5% intensity level), Div(ergence) (full angle), Mode, Pol(arization); Qual(ity); ph(ase); ML
Multiline; MM = Multi Mode; pp: peak-to-peak; reg(ulation), RGB: Red Green Blue, TBM: Tunable Bragg Mirror.
1
Spectra-Physics produces similar water-cooled on lasers both in pure and mixed Argon/Krypton with Z-Lok automatic single frequency stabilization and J-Lok jitter reduction. Examples are the BeamLok 2060 and 2080 pure
gas models and the Stabilite models 2017 and 2018-RM, as well as BeamLok 2060 mixed gas lasers.
2
Innova Series V plasma tube, SuperInvar resonator, Model 70C: 3 phase with <40 A/phase, 8.5 l cooling water flow rate/min., Model 300C: 3 phase with <55 A/phase, 9.6 l/min. cooling water, sealed mirrors, with PowerTrack,
ModeTrack, ModeTune, modehop-free operation with etalon.
3
3 phase with <70 A/phase, SuperInvar stable resonator, with Sentry system management complete automation for wavelength selection, mode-control, search and tune, 5 min warm-up, NuTrack active cavity length stabilization
for single frequency operation, Sabre Purelight Low Divergence beam diameter and beam divergence, noise level between 10 Hz and 2 MHz.
4
Single phase and <31 A, SuperInvar resonator, 8 l/min. cooling water, Series V plasma tube, Single line and UV emitting systems also, I nnova: Series V plasma tube, SuperInvar resonator, Model 70C: 3 phase with <40 A/phase,
8.5 l cooling water flow rate/min., Model 300C: 3 phase with <55 A/phase, 9.6 l/min. cooling water, sealed mirrors, noise level with PowerTrack, ModeTrack, ModeTune; mode-hop-free with etalon.
5
Warm-up time 15 min., stability <1% over 2 h., noise between 20 Hz and 2 MHz, beam pointing stability <30 mrad/3K.
6
Advantage Model 163C–A 02, Noise level from 10 Hz–3 kHz, 0.3% from 10 Hz to 2 MHz.
7
Many He-Ne suppliers exist. He-Cd are made by The Cooke Corporation, Melles Griot and Kimmon Electric, Vapor lasers. HeNe: Green 543 nm; Orange 594 nm; Red 633 nm Model LGK-7654-8 20 min warm-up time, noise
between 30 Hz and 10 MHz, 5% power stability over 8 h.
8
Black + White Knight RGB -450M, RGB 441.5/537.8/635.5 nm, other non-RGB version as indicated. Warm-up time 20 min., noise between 10 Hz and 10 MHz, stability 3%/2 h.
>20,000
n.a. &
<600–700
Cooke Corp. &
He–Cd, White+
Black Knight8
Gas lasers: Helium–Cadmium (He-CD), Green/Yellow/Oramge/Red Helium–Neon (He–Ne)
Manufactory
& Model
TABLE 5.1.2. Continuous Wave Lasers
PHC5 28/11/2005 6:15 PM Page 93
Laser Sources for Confocal Microscopy • Chapter 5
93
Guided Color6
& Fiber laser
Lumics GmbH7
National Laser
& LasNOVA8
Unique-m.o.d.e
AG9 &
UMFL491
Fiber Lasers
Novalux Inc.4
& Protera
xxx–x
Power
Technology
Inc.5 &
IQ series
Toptica &
PVLS 500
PVLS 3000
Blue Sky Res
& ChromaLase 4881
Coherent2 &
Radius 375-8
Radius 405-50
Sapphire 460
488-2003
Radius 440-16
Radius 635-25
Microlaser &
Lepton series
L4 xxx–xx–TE/
Manufactory
& Model
>10,000
>5000
>5000
Twice that
of argonion
>20 k
>10 k
5
<15
<20
<20
<20
<15
5–10
>5
10
490 ± 3
490
491.5 &
<1
10
3.5
50
V1
<0.3
<0.5
<0.5
<1
<0.2
<0.2
0.2
<0.5
<0.5
<0.5
<0.5
<30
<25
<30
<30
<30
<30
<30
<0.6
<0.6
<0.6
<0.6
<1
8
50
25
200
200
16
25
6
36
12
3
5
5–20
5
8
50
16
4
<2
<2
<2
<2
<2
<2
<2
<5
<2
<5
<5
<2
%
mrad/°C
<4
Power
Stab.
Pointing
Stab
<1%
rms, %
Noise
25
mW
CW
Av. Power
491
488 ±
0.5 &
0.1
375 ± 5
405 ± 5
408
460 ± 2
488 ± 2
440 ± 5
635 ± 7
375 ± 5
408 ± 10
440 ± 10
473 ± 10
460 ± 2
488 ± 2
532 ± 2
375 ± 5
405 ± 10
440 ± 10
473 ± 5
405 ± 10
<12
<15
nm
W
h
Wavel. &
Spectr W.
Power
Cons.
Lifetime
Diode lasers: Continuous Wave
20–100
<5 Hz
<100 Hz
<5 Hz
<5 Hz
MHz
Mod.
Freq.
100 MHz
dig. TTL
MHz
Pulse
Mod
<200
<5
<200
<200
msec
Rise
time
0.75
0.7
2.1–4.3
0.7
0.7
0.7
2.5 ¥ 4
1.3
2.5 ¥ 4
2.5 ¥ 4
0.7
0.7
1¥3
1
6 stand.
or
1/2/4/
1¥3
4.7 ¥ 1.6
0.7
mm
Diam.
<1.2
<1.2
<0.2
<0.2
<1.2
<1.2
0.7 ¥ 0.3
0.7 ¥ 0.3
<0.7
<0.7
<0.7
<0.7
<1
<1.2
<0.8
<0.5
<0.5
<0.5
<0.5
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
Mode
Beam Parameters
0.6 ¥ 1
0.7 ¥ 0.3
<1
mrad
Div.
random
random
50 : 1 or
random
100 : 1
V100 : 1
V100 : 1
100.1
100 : 1
>100 : 1
>100 : 1
>100 : 1
>100 : 1
V100 : 1
V100 : 1
V100 : 1
100 : 1
100 : 1
H
100 : 1
Pol.
<1.1
<1.1
<1.4
<1.3
<1.2
<1.2
<1.2
<1.1
<1.1
M2
Qual.
94
TABLE 5.1. (Continued)
PHC5 28/11/2005 6:15 PM Page 94
Chapter 5 • E. Gratton and M.J. vandeVen
>10,000
<40
<60
<15
<60
<100
<60
<20
<20
50–300
5/8/10
45
5/10/15
>10
442 ± 3
442
473 ± 5
561
1064
532
350
15
10
5–40
5
50
500
2000
500
200
947
1064
1319
1444 &
1 kHz
532
430 ± 1
473
488 ± 1
605
evaluation
488 ±
0.3 & 0.1
50
100
473
532
<0.75
<0.5
<1
<0.5
<2 pp
<0.5
<3 pp
<2 pp
<40
<30
<20
<30
<10 /h
<2
<2
<2
~2.5
~2.5
<1
Q-Sw
IR + 532
0.7
0.4 ¥ 0.3
0.25
0.4 ¥ 0.3
0.67
0.67
0.65
2.5
<0.8
3
<0.8
<1.2
<1.1
0.5
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
<5
<6
<6
<1
0.1
0.1
TEM00
TEM00
TEM00
TEM00
3 (3 h)
3 (3 h)
<0.1
<0.1
<0.1
<0.1
<0.5
<0.5
V
V100 : 1
V100 : 1
V100 : 1
V 100 : 1
100 : 1
<1.1
<1.2
<1.1
<1.2
<1.2
<1.2
<1.1
<1.1
<1.1
<1.1
<1.1
<1.2
Legends: Pwr = power; Cons(umption); Wavel(ength); Spectr(al) W(idth) given as FWHM (Full Width Half Maximum); CW: Continuous Wave; Av(erage) Power; Pointing & Power Stab(ility); Mod(ulation) Freq(uency); Beam
Parameters Diam(eter) (e-2 = 13.5% intensity level), Div(ergence)(full angle), Mode, Pol(arization), Beam Qual(ity). DPSS: Diode Pumped Solid State lasers.
1
Noise between 20 Hz and 20 MHz, power stability over 8 h.
2
Several models with varying output available, Model 375-8 CDRH; Model 405, noise between 210 Hz and 10 MHz, modulation may save tube life.
3
Noise between 20 Hz and 2 MHz, power stability after 2 hours and ±3°C.
4
Model 460-5; Model 488-15: Warm-up time <5 min, Spectral width <<0.01 nm, Noise between 20 Hz and 2 MHz typically only 0.03%, IR emission <0.1 mW, warm-up time 2 min.
5
IQ series: Instrument Quality, beam circulization available as well as analog and digital modulation.
6
Fiber Laser (Guided Color Technologies GmbH): noise level when stabilized otherwise >10%.
7
Lumics LU0490F005.
8
Noise level between 30 Hz and 100 MHz.
9
Warm-up time <10 min, noise between 10 Hz and 10 MHz, optional single-mode fiber coupling.
10
Noise level between 10 Hz–100 kHz pp, <3% between 100 kHz–5 MHz, <0.5% between 5–100 MHz, 3% long term (3 h) stability.
11
Ring geometry DPSS, single frequency (1 kHz) with PZT over 20 GHz tuning range, noise eater option.
12
Warm-up time 3 min, to beam-pointing stability 10–15 min, noise between 20 Hz and 10 MHz, TEC cooled.
13
Warm-up time <3 min, Noise between 20 Hz and 10 MHz, pointing stability for ±2°C.
14
Noise between 10 Hz and 10 MHz, power and pointing stability after 2 h, 473 nm laser data from CrystaLaser Inc., warm-up time 1 min.
15
Evaluation kit for GaAsSb 5 W 808 nm pumped DPSS laser.
16
Warm-up time <1 min, noise level between 20 Hz and 20 kHz.
17
Kit comes complete with Nd:YVO4 or Nd:YAG crystal, pump diode, mirror sets, KTP doubling crystal, 670 nm alignment laser diode, IR — VIS converter.
ALPHALAS
DPSS–17
DPSS Kits
Rainbow Ph.14 &
BluePoint430
473 laser
BluePoint488
Toptica15 &
Eksel 110
TorsanaLas.16 &
Starbright 488
LaserQuantum
& Torus 532
Melles Griot12 &
85 BTA
85 BTC
85 BCA13
85 YCA
Cobolt AB &
Blues10
Samba
INNOLIGHT11
& Mephisto QTL
Mephisto Mephisto
YLF Mephisto E
Continuous Wave DPSS Lasers
PHC5 28/11/2005 6:15 PM Page 95
Laser Sources for Confocal Microscopy • Chapter 5
95
4
Hamamatsu7 &
PLP-10
Becker-Hickl6
& BDH series
Pulsed diode lasers
LambdaPhysik &
LPXPro
TuiLaser AG
& Excistar5
Pulsed excimer lasers
SpectraPhysics
& VSL337
& Model
Manufactory
h
Lifetime
90
0.8, 2.4
0.9, 0.5,
10
70, 50,
20, 20,
30, 30,
30, 30,
50 pk
power
3.5 k
3.5 k
308
351
mW
Av. Power
CW
1.5
CW
Av. Power
38 k
28 k
375, 405,
475, 635,
680, 780
375, 405,
440, 470,
635, 650,
670, 780,
850 All
±10
0.1
nm & Mhz
Spectr W
Wavel. &
4 / 0.04
and
0.02 cm-1
0.03, 0.07
-0.15 cm-1
mW
Spectral
Width
308
351
337.1
W
Cons.
Power
330–780
360–950
198–320
850
nm
mJ
<65@355
<150@53
2 @ 40 Hz
@1–30 Hz
Wavel.
Range
Pump
Power
5–35 (gr)
5–80 (pr)
8–150 (gr)
8–230 (pr)
50–650 (gr)
@<100 Hz
Pulsed nitrogen lasers
PrecisionScan
LambdaPhysik
&
ScanMatePro1
SpectraPhys.2
& Duo
SpectraPhys.3
& Cobra
Cobra Stretch
Pulsed dye lasers
Manufactory
& Model
<0.03
<0.5
rms, %
Noise
<0.03
<0.5
<2
<2
<2
<0.0005
<0.0005
20
50
80
2 Hz–100
10,000
10,000
<4 pp
<0.0001
<0.0001
1–20 Hz
%
Rate
MHz
Stab.
Rep.
<0.5
<0.5
<0.5
<0.5
%
ASE
Stab.
Pk-Pk
<0.005
<0.005
<0.005
nm
Wavel.
Reset.
mrad/C
Pointing
<0.03
nm
<0.5
6 pp
(rms)
(%)
Noise
Abs.
Accur.
100
60, 60,
50, 50,
n.a & 8
8
n.a. & 400
320
4k
ps
Width
Pulse
<28
<28
<20
28
%
Conv.
Eff.
100–500
3¥6
3¥6
30 kW
& >120
mW & mJ
& Pk E.
Pk Pwr
<0.01
<0.05
<0.01
<0.05
<0.01
<0.05
mm
cm-1/°C
<1 ¥ 2
<1 ¥ 2
3¥7
mm
Diam.
2¥3
Diam.
Freq.
Stab.
Div.
15–34,
13–22,
10–35,
10–30,
12–32
3¥8
mrad
Div.
Mode
Mode
Beam Parameters
<0.5
<1
<1.5
4
0.5
mrad
Beam Parameters
%
Pol.
>98
>98
>98
Pol.
M2
Qual.
<1.1
M2
Qual.
96
TABLE 5.2. Pulsed Lasers
PHC5 28/11/2005 6:15 PM Page 96
Chapter 5 • E. Gratton and M.J. vandeVen
<3 k
<3 k
<3 k
<3 k
510.6
578.2
530 (SHG)
780–790,
1060.1065
400, 635
670, 805
375, 405,
440, 470
& ±3
635,
637, 655,
660, 665,
670, 690,
735, 757,
785, 806,
830, & ±7
10 k
20 k
20 k
35 k
1.0, 1.0,
1.0, 1.0
(wide)
3.5,
8.0, 6.0,
8.0, 4.0
2.0, 8.0,
3.5, 6.0,
8.0, 10,
4.0
(wide)
600
160
150
<1
3 pp
<5
<5
<2
<3
<1%
3% pp
0.01
0.01
0.01
0.01
40–80
0.1–1500
2.5 k to 3 3
<70, 70,
90, 90
narrow
<90,
100, 70,
90, 90,
70, 70,
130, 90,
70, 120,
100
narrow
<50
<90
<80
40
80
80
140
600
7000
6000
4
4
4
4
>90
>90
>90
>90
>90
>90
>90
>90
>90
(Continued)
Legends: Power Cons(umption); Wavel(ength); Spectr(al) W(idth); CW: Continuous Wave; Av(erage) Power; Abs(olute) Accur(acy);Wavel(ength) Reset(tability); ASE: Asynchronous Spontaneous Emission; Conv(ersion)
Eff(iciency); Freq(uency), Pointing, Power and Peak-to-Peak (pk-pk & pp) Stab(ility); Rep(itition) Rate; Peak Power (Pk Pwr) & Pulsek E(nergy); Beam Parameters Diam(eter) (e-2 = 13.5% intensity level), Div(ergence) (full
angle), Mode, Pol(arization), (Beam) Qual(ality).
1
Model ScanMatePro 2 pumped by both <40 Hz 308 nm Excimer and <50 Hz 355 as well as 532 nm Nd-Yag, 28% pump efficiency at 560 nm with Rhodamine 6G, frequency extension to 198 nm.
2
Nitrogen laser pumped, pulse width >4 ns.
3
Cobra series and PrecisionScan pumped by 355 nm and/or 532 nm at 10–100 Hz with Nd-Yag laser, LambdaLok (LLOck) frequency stability <0.01 cm-1 in Table LLock for PrecisionScan conversion efficiency at 570 nm, Labview
and Sirah control software, SHG and THG packages available.
4
Air-cooled VSL-337, int. and ext. trigger capabilities.
5
Excistar S-500 suitable as pump source for cuvette type dye laser.
6
Warm-up time 3 min., average power at 80 MHz pulse repetition rate, pulse width at 50 MHz @ 1 mW power, more red wavelengths available.
7
Output peak power after fiber pigtail, wide pulse (listed) and narrow pulse operation.
8
MDL: Modulated diode lasers, modulation frequency range 100 kHz–1.5 GHz.
9
Repetition rate 80 MHz but other frequencies possible, pulse width with narrow pulse listed; with high power adjust pulse width is ~ <300 ps for wavelengths below 470 nm; above 470 nm pulse width in high power adjust mode
<400 ps.(wide pulse listed), selected 50 ps devices , narrow distributed feedback bandwidth available, 440 nm also available with 3 mW average output.
10
80, 40, 20, 10 and 5 MHz pulse repetition rates and adjustable rates, ASE <50 mW @ 780 nm, <60 mW @ 1060 nm, fiber coupling passes <5% IR in multimode and <0.1% for single-mode fiber-optic, spectral linewidth 4 nm
or 0.1 nm (narrow).
11
Water consumption 2 l/min, 80% of plane-plane cavity output within 0.6 mrad, Timing jitter (ps): LS10-10/20-10: 5 ps, LS 20-50: 2 ps, LS 35: 3 ps.
Oxford
Lasers11 &
LS 10-10
LS 20-10
LS 20-50
LS 35
Pulsed vapor lasers
PicoQuant10 &
PicoTA
PicoQuant9 &
PDL LDH xxx
PicoQuant8 &
MDL 300
PicoQuant &
PDL LDH
Pulsed diode laser
PHC5 28/11/2005 6:15 PM Page 97
Laser Sources for Confocal Microscopy • Chapter 5
97
1,064/532
Optronics8
<1.3 k
1,064/532
>10 k
pump
diode
>10 k
<700
<1.1 k
<1 k
W
h
>10 k
(pump
diode)
Power
Cons.
Lifetime
ALPHALAS &
Yag or YVO4
Kits
Coherent &
Verdi
Elektronik
Laser System3
& VersaDisk
GWU
Lasertechnik4
& simoLAS
Lumera Laser
& UPL-20
SpectraPhysics5
& Empower
Spectra Physics6
& Millennia-Pro
Spectra Physics7
& Vanguard
Becker & Hickl1
& BDL-405
Coherent2 &
Evolution with
Nd:YLF crystal
Pulsed DPSS lasers
Manufactory
& Model
500
500
532 &
<5 MHz
532
532
6k
16 k
10 k
1,064 ± 40
1,064
532
527
2/5/6/8/
10/18 k
30 k
2/5/6/8/
10 k
2000
5–100 k
2–15 k
15 k
30 k
75 k
90 k
0.3–1.3
mJ
Peak
Energy
1,030
515 / <5
527
527
527
527
405
nm/Mhz
Wavel. &
Spect W
3–20
12
20
15
18
0.04%
<10 ps
ns / %
Jitter/
Noise
n.a. & <0.03
n.a. & <1
n.a. & 0.04
<2
25
<2
1
Q-sw.
5–15
50–100 kHz
76 or 80
1–10 kHz
n.a. & <1
10 kHz
Q-sw
20–8
ps
Pulse
Width
160 ± 1
<1
MHz
Mode
lock
n.a. & <1
<2
20–50
%
mrad/C
20–50
Power
Stab.
Pointing
Stab.
Q-sw.
10–30 k,
50–100 k
<12
<10
200 ns
200 ns
200 ns
200 ns
>60
mW & mJ
Pulse
Width
2.25
1.4
2.3
3
5
5
7
7
0.5
Mm
Diam.
TEM00
<0.5
<1
<0.5
<2
<0.5
<0.7
TEM00
TEM00
Multi-
TEM0
TEM00
TEM00
special
version
Mode
Pol.
V/99
V/99
V/99
H
99
V/99
H
H
H
H
H
Dir/%
Beam Parameters
mrad
Div.
<1.1
<1.3
<1.1
<1.2
<1.1
M2
Qual.
98
TABLE 5.2. Pulsed Lasers (Continued)
PHC5 28/11/2005 6:15 PM Page 98
Chapter 5 • E. Gratton and M.J. vandeVen
Amplitude Sys
& t-pulse
Coherent9 &
ChameleonXR
Coherent
Mira 900 fs10, V
or Innova 310
or Sabre 14 W
Mira 900 ps, V
or Innova 310
or Sabre 14 W
Doubling
Coherent
Vitesse 800-211
Vitesse 800-5
DelMarVent&
SciCRF-65P
Femtolasers
Productions
GmbH & FS
Scientific XL12
High Q Laser
& FemtoTrain13
PicoTrain DPSS
Quantronix &
Integra-i14
SpectraPhysics15
& Mai-Tai
SpectraPhysics
Tsunami fs16
w. MillenniaV
w. MillenniaX
Tsunami ps
w. MillenniaV
w. MillenniaX
XL optics set
Doubled fs sel
w. MillenniaV
w. MillenniaX
Doubled ps sel
w. MillenniaV
w. MillenniaX
R&D Ultrafast
Lasers Ltd &
FemtoRose 10017
FemtoRose 10
Time-Bandw&
Tiger fs
Pallas ps
Pulsed lasers, ultrafast
<500
<500
3 phase
50 A/ph
<2300
1.3 k
800
1,400
1.3 k
800
1.4 k
2–15%
700–980
>1.0
>1.8
345–540
345–540
780–860
710–980
700
1,500
710–980
700–1,000
970–1,080
500
500
>3.2
>6
700
1,400
710–980
700–1,000
n.a &
0.12 nJ
0.22 nJ
n.a. &
0.4 nJ
0.75 nJ
85
170
150
1.5 k
800–900
1,064
1,600 &
>80 nJ
710–990
>400
>100–10 k
800 & >50
n.a & 1
880
1,230–1,270
n.a. &
20 nJ
800 center
>200
>650
180–250
800 ± 1
240–320
350–500
>1 k
1k
720–980
1,030
n.a &
<0.1
n.a &
<0.2
n.a.
<2
n.a &
<0.1
n.a. &
<0.15
n.a. &
<0.1
<2
<2
<0.1
<2
<2
1%/°C
1%/°C
50–200
75–200
70
–100 ps
80–150
<10
<2 k
<130
<1
<2
80
100
<1
80
<2
<1
2
<1.2
<1.1
<2
1.7
30 mrad/
100 nm
<100
7.5 k
<2
1.25
1.45
0.8
1.2
10
75
60–1 k
<20
<65
<0.1
<200
<3 k
<140
<200
<1.5%
<1
<1
11
80
76
<3
<5
<5
<3
<5
<5
1
90
2
50
TEM00
TEM00
TEM00
near
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
TEM00
H
99.8
500:1
V
H
H
H
H/99
H
H/99.8
<1.1
<1.1
(Continued)
<1.8
<1.2
<1.2
<1.3
<1.2
<1.1
PHC5 28/11/2005 6:15 PM Page 99
Laser Sources for Confocal Microscopy • Chapter 5
99
Power
Cons.
W
Lifetime
h
740–840
720–980
1,240–1,270
780–840,
740–950,
710–950
780–810
750–840
& >60 nm
nm/Mhz
Wavel. &
Spect W
400–450
<250
<1 k
<250
250–1 k
mJ
Peak
Energy
n.a &
4.4–6.1
ns / %
Jitter/
Noise, rms
%
mrad/C
n.a. & <2
Power
Stab.
Pointing
Stab.
<5
MHz
Mode
lock Freq.
80–100
70–140
ps
Pulse
Width
<15 fs
<25 fs
20–200
<20
<100
<65 s
mW & mJ
Pulse
Width
~1
2–3
Mm
Diam.
Div.
~1
<2
<2
mrad
~
TEM00
TEM00
TEM00
Mode
Pol.
H
H
H
Dir/%
Beam Parameters
M2
Qual.
Legends: Power Cons(umption); Wavel(ength); Spectr(al) W(idth) at Full Width at Half maximum (FWHM); Av(erage) Power; T(iming) Jitter; Pointing and Energy Stab(ility); Rep(itition) Rate; Peak Power (Pk Pwr) & Pulsek
E(nergy); Beam Parameters Diam(eter) (e-2 = 13.5% intensity level), Div(ergence)(full angle), Mode, Pol(arization), (Beam) Qual(ality).; fs: femtosecond, ps: picosecond, stabil(ity),V(erdi).
1
Warm-up time 3 min., average power at 80 MHz pulse repetition rate, pulse width at 50 MHz @ 0.5 mW power, more red wavelengths available.
2
Ti:Sa pump laser, pulse repetition rate 1–10 kHz,. Peak energy at 1 kHz pulse repetition rate, noise over 8 hours < 1% (rms).
3
Noise level defined between 10 Hz to 1 GHz, consumes 600 W, linewidth without etalon < 1 GHz,, birefringent tuner for 1000–1060 nm region.
4
Single-mode Nd:YVO4 ring laser.
5
Nd:YLF pump laser for ultrafast amplifiers.
6
Ti:Sa pump laser, noise between 10 Hz and 100 MHz, Quiet MultiAxial Mode doubling (QMAD), laserhead closed chiller cooled for 5, 6, 8 an 10 W versions.
7
Model Vanguard 2000-HM-32, Nd:YVO4 lasing crystal, warm-up time 30 min, to full specifications 1 h., cooling air flow 300 cfm.
8
Nd:YVO4 crystal, non-linear crystal, set of mirrors, Cr:YAG passive Q-switch, Glan-Taylor prism.
9
Pumped by Verdi lasers, with Power Pulse automatic pulse width optimization, and Power Track active alignment, with MRU miniature air re-circulator unit for easy access to red wavelengths.
10
V(erdi) model V8, Innova Argon-ion model 310-8 or V10 pump or Sabre ion laser 14 W pump laser, Optima version provides complete automated optimization, version S(imple) has manual starter, external Mira model 9300 harmonic generator with LBO crystal doubler with BBO frequency mixing crystal.
11
Ti:Sa oscillator with built-in Verdi pump laser, noise between 10 Hz and 10 Mhz.
12
(FemtoSource) Scientific SL.
13
Ti:Sa model IC-xxx-400 many options w.r.t. crystals, pulse width, repetition rate, and PicoTrain, Nd:YVO4 crystal, output stability over 24 h.
14
Pump source for OPA, includes Nd:YLF pump laser, stretcher, seed oscillator with multi-pass regenerative amplifier, compressor.
15
Peak power at 800 nm, closed loop chiller, StabiLok beam pointing stability.
16
Millennia V and X pump lasers, broadband optics, extra mirror set for wavelength range 970–1080 nm, average power at 970 nm, noise between 10 Hz and 2 MHz.
17
Birefringent tuner.
18
5 W DPSS pump recommended, KLM mode-locking.
19
3–10 W DPSS or ion laser pump power.
20
Pump power 6–10 W Ytterbium fiber pump, options available w.r.t. emission center, bandwidth and average power, FemtoRainbow 100 OPO KTP based delivers ~ 100 fs pulses.
CrF-kit-65P
DelMarVentures
FemtoStart18
TISSA-20-10019
KapteynMurnane Labs.
& TS20
Avesta Project &
TIF-kit-xx
Pulse lasers: Ultrafast kits
Manufactory
& Model
100
TABLE 5.2. Pulsed Lasers (Continued)
PHC5 28/11/2005 6:15 PM Page 100
Chapter 5 • E. Gratton and M.J. vandeVen
780
800
1,150–1,400
IMRA &
FemtoLiteA25
FemtoL.FX-10
Toptica1 &
FFS-F-C Sys
1100–1250
1350–1600
525–665 KTP-ring-SHG
1050–1330 KTP linear/
ring-IR 1350–1600
CTA linear
222.5–450
445–1750 & <0.075
870–1550 (s)
1570–5600 (i)
1050–1510(s)
1590–3100(i)
417–2400
245–1700
1048
524
1150–1600 (s)
1600–2630 (i)
400–1150 (SHG)
300–400 (FHG)
1.5 k
4k
250
200
250(s)
200(s)
>200–240
high power
version
25
10
100
CW Av. &
Pk Power
mW & kW
<2 (8 h)
<5 stability
<0.2
n.a. & ±10
rates
n.a. & <2
Time-jitter
& Noise
rms, ns/ %
<7
80 MHz
10 Hz
76 MHz
or other rep.
50
50
Rep Ratel
ML
MHz
n.a.
<500 ps
n.a. & 5–50 ps
80–150 ps
>1 @ 800 nm
1 kHz
n.a. &
<130 ps
<130 ps
350 mj @
355 nm
<1.2 ps
<200
180 fs
120 fs
100 ± 10 ps
ML
mJ & ps
10–50
10–50
10–50
10–50
60 mJ @
1.3 mm &
<130 f
n.a. & <2 k
n.a. & <2 k
120 mJ @
1.3 mm & <130
50 mJ &
3–6 ns
n.a. & <250
ML & Q-sw
mJ & nsec
50 mJ &
3–6 ns
n.a. & <250
Q-sw
mJ & nsec
Pulse E & Width (FWHM)
<2.0
~4–6
<1.25
Diam.
mm
<1.0
<1.5
<1.30
Div.
mrad
TEM00
TEM00
TEM00
Mode
Beam Parameters
99.9
H(s)
V (i)
H >99
V
99%
H >99
H
Pol.
Dir/%
Qual.
M2
Legends: Av(erage) & Peak (Pk) Power; Rep(itition) rate; ML: Mode-Locked; Q-switched); Pulse E(nergy); Beam Parameters Diam(eter) (e-2 = 13.5% intensity level), Div(ergence) (full angle), Mode, Pol(arization), (Beam)
Qual(ality);. i: idler; n.a.: not applicable; s: signal.
1
FFS-F-C laser system: FemtoSecond Scientific Er doped fiber laser.
2
500 mW MIRA 900 pumping threshold, ring cavity version gives more output power, active length stabilization, sealed and purgeable, intracavity doubling for VIS range, NCPM crystal.
3
Precision II models 355 nm pump laser, temperature stabilized, narrow bandwidth related replacement for a dye laser.
4
720–800 nm 1.2 W 100 fs pump at 1200 nm.
5
Millennia X pump, LBO crystal, active stabilization, frequency doubler for visible, automated.
6
OPA 9400/9450 and 9800/9850: RegA wavelength extension for Vis (blue) – NIR; RegA wavelength extension for NIR — MIR. RegA 9000/9500: High repetoition rate, < 260 kHz, ultrafast amplifiers, high spatial beam quality,
one RegA can pump two OPAs. Other systems are: Hidra: custom high energy; Legend: Flexible Ti:Sa regenerative amplifier; Libra: One box, ultrafast Ti:Sa amplifier.
7
gap free tuning from UV < 300 nm — MIR for wavelength extension of Legend and Libra, fs, lowest optical noise, automated, white light initiated OPA.
8
Traveling Wave Optical Parametric Amplifier: femtesecond (listed) and picosecond pumped at e.g. 800 and 355 nm as well as 400 nm.
Quantronix8 &
TopasPlus Hep
Topas White
SpectraPhysics &
Eclipse
Coherent6 &
OPA 9800/9850
Coherent7 &
OperA
Optical parametric amplifiers (OPA)
Continuum &
Sunlite EX3
GWU-Lasertechnik
& Synchro fs4
GWU-Lasertechnik
& Synchro ps
Spectra-Physics5 &
OPAL 1.3
OPAL 1.5
Coherent
Mira-OPO2
Ps and fs
Optical parametric oscillators (OPO)
Wavelength
nm
Manufactory
& Model
Ultrafast fiber lasers
PHC5 28/11/2005 6:15 PM Page 101
Laser Sources for Confocal Microscopy • Chapter 5
101
PHC5 28/11/2005 6:15 PM Page 102
102
Chapter 5 • E. Gratton and M.J. vandeVen
Their contrasting characteristics are described near the end of this
section. As a pump source for Ti : Sa lasers, the argon-ion laser has
been completely replaced by solid-state lasers.
In general, argon-ion laser stability is better than that of a
krypton laser, with less gas slushing. The output power of krypton
is at best about 30% that of argon under identical lasing conditions. Commercial systems provide a large variety of emission
wavelengths and output powers (Fig. 5.8; Ginouves, 2002).
Fragile quartz plasma tubes have been replaced by versions
with rugged metal–ceramic (beryllium oxide, BeO) envelopes for
the small- and medium-frame lasers, which usually do not use a
magnet to confine the plasma to the central core of the tube. External copper disks remove the heat.
Large systems with high output power use alumina (Al2O3)
ceramic tubes with brazed tungsten/copper disks inside, crystalline
quartz–coated windows, and quartz–metal hard seals (Fig. 5.4).
This significantly increases their reliability. Other systems use
rugged metal/ceramic BeO tubes. The central bore defines the laser
beam. The plasma discharge pumps ions towards the cathode end
of the tube, creating a pressure gradient between the anode and
cathode. Return holes around the outer edge of the disks form an
internal gas return path to maintain a uniform gas pressure along
the length of the plasma tube. Typical gas pressure inside a gas
laser is about 1 Torr.
The formation of color centers in the Brewster windows has
been reduced by the use of better quality window materials,
increasing the lifespan of this type of laser to 3000 to 10,000 h.
However, tube lifetime may only reflect the life of the strongest
laser line and not that of the weaker lines needed on a daily basis.
The stability of these lasers, especially the small- and mediumsized models, is good because they can be cooled by convection
(passive cooling) or forced air and their electric power requirements are modest. Some models are portable and easy to use. The
larger frame lasers provide active resonator stabilization.
The wall-plug efficiency of large argon-ion models is very low
(0.1% typical) and the heat generated must be removed with a large
amount of cooling water. The turbulent flow causes some vibration in these systems while insufficient water flow may cause the
cooling water to boil and lead to the destruction of the plasma tube.
The large amount of heat generated also puts more strain on the
2
4p P
1/2
2
4p D
2
4p P
3/2
457.9 nm
454.5
465.8
3/2
496.5
2
4p D
5/2
476.5
472.7
4
4p D
5/2
2
4p S
1/2
Argon-ion Laser Lines in the UV
Commercially available argon lasers have emission wavelengths
extending from 275 to 528 nm (Hecht, 1992). Higher discharge
currents are necessary to populate the doubly-ionized argon levels
sufficiently for deep UV emission. Wavelength selection is
obtained by installation of an angle-tuned quartz prism or other
appropriate optical elements. The wavelength range is extended
down to 229 nm by inserting a frequency-doubling crystal in the
laser cavity and thereby producing a CW intra-cavity frequencydoubled argon-ion laser system (FreD™, Coherent, see Tables 5.1,
5.3). Reported power levels range from 40 mW at 228.9 nm to 1 W
at 257 nm for the model Innova Sabre MotoFreD™.
Argon-Ion Laser Lines in the Visible Region of
the Spectrum
The most common spectral region for lasers in microscopy is
between 488 and 514 nm, although argon-ion laser emission can
be obtained at 454.5, 457.9, 465.8, 472.7, 476.5, 488.0, 496.5,
503.7, 514.5, and 528.7 nm.
Krypton-ion
When strong red emission at 647.1 nm is needed to expand the
spectral coverage of an argon-ion laser, a krypton laser is the
system of choice (Brelje et al., 1993). Its stability is slightly less
than that of a comparably sized argon-ion system and the gas retention in the graphite disks is slightly larger. However, with active
stabilization now available and with ceramic tube technology
firmly in place, both drawbacks have now been largely overcome.
Based on the FreD™ argon-ion laser, see above, we may see the
introduction of an intracavity frequency-doubled krypton-ion laser
as a source for UV light at 376, 338, 323, 284, and 266 nm.
Mixed Gas Argon–Krypton
488.9
488.0
2
4s P
1/2
514.5
resonator cavity. Longer warm-up times are necessary when
no active stabilization is incorporated in the design. The operational lifetime of the higher power laser tubes is now well above
2000 h.
Using proper interference filters (inserted with their shiny side
towards the laser head to prevent damage) or emission line selection by means of an AOTF or AOBS, single, strong laser lines can
be selected. Dual line emitting argon-ion lasers simultaneously
emitting at 454 to 457 nm and 514 nm are also offered (Table 5.1).
When the experiment requires a reliable, stable source of deep
UV radiation at power levels of tens to hundreds of milliwatts,
these ion laser systems provide a better alternative than any current
solid-state UV laser system, as these require more steps to produce
UV light. The area where the argon-ion laser was very valuable,
as a 5 to 10 W pump source for solid-state Ti : Sa lasers, has been
completely overtaken by solid-state lasers.
2
4s P
3/2
Combining the best properties of the argon and krypton ion lasers
creates a reliable laser with stable output and the broad spectral
coverage needed for multicolor CLSMs. Brelje and colleagues
(1993) used an air-cooled 15 mW Kr-Ar ion laser for multicolor
immunofluorescence microscopy and described a range of applications. Several companies produce systems delivering several
tens of milliwatts, for example, Melles Griot (Table 5.3).
Helium–Neon
Ar
+
FIGURE 5.8. Schematic energy diagram for an argon-ion laser.
The use of these stable, inexpensive lasers in microscopy has been
somewhat limited by their relatively low intensity and predominantly 632.8 nm red emission. Several manufacturers have recently
PHC5 28/11/2005 6:15 PM Page 103
Laser Sources for Confocal Microscopy • Chapter 5
introduced He-Ne lasers with emission at 534, 594, 612, and
632.8 nm and lines in the infrared, for example, 1152 nm (Fig. 5.9).
To better excite Texas Red, Alexa 594, and several other dyes, the
red He-Ne at 632.8 nm and the green He-Ne (GreNe) at 543 nm
have been joined by the orange He-Ne at 594 nm. Typical powers
for the 632.8 nm line range from 0.5 to 10 mW with a maximum of
75 mW. Although increased competition is expected from solidstate, semiconductor, and diode lasers, the He-Ne laser still competes well thanks to its good beam quality, shorter wavelengths,
and long lifetime. Invar-stabilized He-Ne frames are now available
from Research Electro-Optics Inc. (Table 5.3). The 632.8 nm can
be used to create an autofocus feedback circuitry that automatically
adjusts for specimen vertical movement (ASI, Zeiss, Table 5.3).
Helium–Cadmium
Helium–cadmium (He-Cd) lasers have found several applications
in microscopy. Usually they operate at three emission wavelengths: 325, 354, and 442 nm (Fig. 5.10), with power levels up to
several hundred milliwatts (Hecht, 1993c). The shortest wavelength requires special optics and is very rarely used in fluorescence microscopy. Membrane probes such as Indo-1 and Fura-2
can be conveniently excited with the 354 nm line. The emission at
442 nm is ideal for excitation of cyan fluorescent protein, flavins,
and other fluorescent molecules. The power available at 325 and
354 nm is generally on the order of a few to tens of milliwatts,
while at 442 nm up to 150 mW is now available.
The stability of these lasers is substantially less than that of the
argon-ion lasers. In the 325 nm region, intensity fluctuations of 10%
to 20% are not uncommon but the emission tends to be more stable
at 442 nm. Although most He-Cd systems had an operational life
of a 1 to 1.5 years (1000 to 2000 h), a new mirror technology has
increased the lifetime of the coatings to > 5000 h for 325 nm
mirrors. The less powerful, smaller frame lasers, operating at
325 nm, may still have a limited life span. One manufacturer now
supplies a sealed, multiple-wavelength version with up to 10,000 h
lifetime (see Table 5.2).
The He-Cd laser experiences heavy competition from the
numerous blue diode lasers coming to market in the wavelength
region spanning roughly 400 to 450 nm.
Alkali Vapor Lasers
Diode–pumped, alkali lasers (DPAL) can generate blue laser
emission by a two step up-conversion scheme (Beach et al., 2004;
Krupke, 2004). Cesium and rubidium vapor CW lasers pumped by
diode lasers either at 852 or 859 or with 876 or 921 nm (Cs) or at
780 or 795 or with 761 or 776 nm (Rb), emit at 455 or 459 (Cs)
and 420 or 422 nm (Rb). With their low cost and compact design,
potential 50% efficiency, high average power (>100 mW), easy
scalability, and good beam quality, they will be competition for
thin disk and fiber lasers and show a great potential.
broad wavelength band. This band can be extended by changing
dyes. Tuning of the laser is accomplished by a prism, a diffraction
grating, and a stack of birefringent plates inserted at the Brewster
angle or by etalons (Demtröder, 1996; Mollenauer and White,
1987). The laser medium is easy to handle, inexpensive, and consists of a fluorescent dye dissolved in a solvent. Because of socalled “forbidden transitions,” the molecules tend to pile up in the
first excited triplet state and thereby become unavailable for lasing.
Rapid pumping or the use of a dye-jet stream reduces the population of this triplet state. In order to keep the quantum yield high,
the dye is usually cooled to minimize competing processes, such
as collisional and vibrational de-excitation. Intrinsic optical efficiencies as high as 20% or 30% can be obtained (not counting the
efficiency of the pump laser) reaching an output power of several
hundred milliwatts.
Operating a dye laser is sometimes rather cumbersome, involving bottles of spent laser dye and waste materials, dye spillage, and
often large quantities of dye solvent that must be stored in a safe
storage area. Cuvettes holding laser dye have to be cleaned regularly because the pumping beam bakes dye molecules onto the
optical surfaces. Although solid-state lasers have largely replaced
the dye laser, they still can fill several spectral gaps between
390 nm and the IR.
The intensity stability of dye lasers is lower than that of its
lamp or laser pump source. Two percent to 5% intensity stability
is considered good. Beam pointing stability is heavily influenced
by the regular tuning done on these systems. Distances should be
kept as short as possible or fiber-optic beam transport should be
considered.
The spectral regions covered, tuning curves, and the output
power offered can be found in the manufacturer’s documentation.
Solid-State Lasers
An ever-growing number of power-efficient solid-state lasers is
being offered for use in confocal microscopy. They are particularly
successful in replacing inherently inefficient gas lasers. Their versatility and ease of operation either for use as CW or pulsed direct
light sources or for use as pump sources for other lasers guarantees continued rapid development. There is a trend towards smaller
more efficient all-in-one-box designs and a drive to increase total
average power by creating laser arrays.
• Semiconductor lasers: Most are based on gallium mixtures,
•
Dye Lasers
As all dye lasers for confocal microscopy are optically pumped by
other powerful lasers, they are not so much lasers as wavelength
shifters. Suitable pump sources are excimer, ion, neodynium–
yttrium aluminum garnet (Nd-YAG), neudynium–yttrium lithium
fluoride (Nd-YLF), and metal vapor lasers. For CW operation, the
dye is circulated to prevent heating and bleaching, to reduce the
competing triplet-state population, and to remove dye molecule
aggregates. Compared with gas lasers, the emission spectra are
quite broad and this permits them to be tuned to lase over a fairly
103
•
emit a single line in the red or NIR and can be pumped electrically with very high efficiency. SHG generation in an appropriate crystal creates UV, blue, and green emission. The
number and versatility of these lasers increases daily.
Non-tunable solid-state laser: Another class of laser is
formed by the Nd-YAG, Nd-YLF, and neodymium–yttrium
ortho vanadate (Nd-YVO4) systems operating at 1064, 1047,
and 1064 nm, respectively. These can be used as a stand-alone
light source or serve as a pump for dye-lasers and titanium :
sapphire amplifier systems.
Tunable solid-state ultrafast lasers: These are epitomized by
the titanium : sapphire laser systems generating very short
pulses down to a few to tens of femtosecond. They provide a
very broad tuning range, comparable with that of dye lasers,
but they are much easier to operate, generate a comparable or
higher output, provide significantly better stability, and can be
built in a range of configurations. They can also generate
stable, high-power picosecond and femtosecond pulses as we
will discuss later.
PHC5 28/11/2005 6:15 PM Page 104
104
Chapter 5 • E. Gratton and M.J. vandeVen
TABLE 5.3. Sample Listing of Manufacturers
ALPHALAS GmbH, Göttingne, Germany, +49-(0)551-770-6146, http://www.alphalas.com/
Amplitude Systemes, Talenc, France, +33-(0)5-4000-3447, http://www.amplitude-systemes.com
AVESTA PROJECTS Ltd., Troitsk/Moscow region, Russia, +7-(0)95-334-0078, http://www.avesta.ru/
ASI, Applied Scientific Instrum., Eugene, OR, (800)-706-2284, http://www.ASIimaging.com
BATOP GmbH, Weimar, Germany, +49-(0)172-3520029, http://www.batop.com/index.html
BEOC, Brocton Electro-Optics Corp., Brockton, MA, (508)-559-9958, http://www.brocktoneo.com/services.htm
Blue Sky Research, Miltipas, CA, (408) 941-6068, http://www.blueskyresearch.com/index.htm
Bromba GmbH, Münich, Germany, +49-(0)89-3266-8930, http://www.bromba.com/indicard.htm
Cambridge Laser Laboratories, Fremont, CA, (510)-651-0110, http://www.cambridgelasers.com/
CDP Inc. or Systems, represented by Edinburgh Instrum. Ltd. Inc, http://www.edinst.com/agencylaser.htm and PicoQuant GmbH.
Clark-MXR, Inc., Dexter, MI, (734)-426-2803, http://www.cmxr.com/gate.htm
Cobolt AB, Stickholm, Sweden, +46-(0)8-545-91-230, http://www.cobolt.se/
Coherent Inc., Laser Group, Santa Clara, CA, (800)-527-3786, http://www.cohr.com/ and http://search.coherentinc.com/
cgi-bin/MsmFind.exe?RESMASK=MssRes.msk&CFGNAME=MssFind.cfg&QUERY=brochures
Continuum, Santa Clara, CA, A division of Excel Technology Comp., (888)-532-1064 http://www.continuumlasers.com/mainswf.html
The Cooke Corp., Auburn Hills, MI, (248)-276-8820, http://www.cookecorp.com/
CSK Optronics, Torrance, CA, (310)-574-8181, csk@usa.com
Crystal Fibre, A/S, Birkerød, Denmark, +45-(0)4348-2820, http://www.crystal-fibre.com/
CrystaLaser Inc., Reno, NV, (775)-348-4820, http://www.crystalaser.com/homenew.html
Del Mar Ventures, San Diego, CA, (858)-481-9523, http://www.femtosecondsystems.com/index.php
Edmund Optics Inc., Barrington, NJ, (800)-363-1992, http://www.edmundoptics.com/
Edmund Scientific, Tonawanda, NY, (800)-728-6999, http://scientificsonline.com/Default.asp?bhcd2=1099033491
Electronik Laser Systems GmbH, http://www.els.de/Products/vdisk.html
Electroi Optical Components, Inc., Santa Rosa, CA, (707)-568-1642, http://www.eoc-inc.com/ir_sensor_cards.htm
Evergreen Laser Corp., Durham, CT, (860)-349-1797, http://www.evergreenlaser.com/Default.htm
Excel/Quantronix Corp., East Setauket, NY, (631)-784-6100, http://www.quantron.com/
Femtolasers Productions GmbH, Vienna, Austria, +43-(0)1-503-70020, and Femtolasers Inc., Harvard, MA, (978)-456-9920, http://www.femtolasers.com/
Fianium Ltd., Southampton, UK, +44-(0)2380-458776, http://www.fianium.com
Guided Color Technologies GmbH, Jena, Germany, +49-(0)3641-675350, http://www.gc-tec.com/
GWU-Lasertechnik mbH, Erftstadt, Germany, +49-(0)2235-955220, http://www.gwu-group.de/laser/index.html
High Q Laser, Hohenems, Austria, +43 (0)5576-43040, and High Q Laser (US) Inc., Watertown, MA, (617)-924-1441, http://www.highqlaser.at/ and
http://www.HIGHQ-US.com/
Holo-Spectra Inc., Van Nuys, CA, (818)-994-9577, http://www.lasershs.com/index.html
IMRA America, Inc. (IMRA: Institut Minoru de Recherche Avancee, now IMRA Europe S.A.), Ann Arbor, MI, (734)-930-2590, http://www.imra.com/
Ingeneric Gmbh, Aachen, Germany, +49-(0)241-963-1343, http://www.ingeneric.com/
INNOLIGHT GmbH, Hannover, Germany, +49-(0)511-760-7270, http://www.innolight.de/products/start.htm
Kapteyn-Murnane Laboratories, Inc., Boulder, CO, (303)-544-9068, http://www.kmlabs.com/
Kimmon Electric USA., Englewood, CO, (303)-754-0401, Edinburgh Instrum. Ltd. rep http://www.edinst.com/agencylaser.htm
Lambda Physik AG (sub. of Coherent Inc.), Göttingen, Germany, +49-(0)551-69380, http://www.lambdaphysik.com/
Laser Diode, Inc. (part of Tyco Electronics Co.), Edison, NJ, (732)-549-9001, http://www.laserdiode.com/
Laser Innovations, Santa Paula, CA, (805)-933-0015, http://www.laserinnovations.com/coherent_lasers.htm
Laser Quantum Ltd, Stockport, Cheshire, UK, +44-(0)161-975-5300, http://www.laserquantum.com/
Laser Resale Inc., Sudbury, MA, (978)-443-8484, http://www.laserresale.com/
LaserMax Inc., Rochester, NY, (585)-272-5420, http://www.lasermax-inc.com/
Lasermet Ltd., Laser Safety Solutions, Bournemouth, UK, +44-(0)1202-770740, http://www.lasermet.com/
Lasiris Inc. (Div. of StockerYale), StockerYale Canada, Montreal, Quebec, Canada, (800)-814-9552
National Laser Corp., Salt Lake City, UT, (801)-467-3391, http://www.nationallaser.com
Lasos LaserTechnik GmbH, Jena, Germany, +49-(0)3641-29440, http://www.lasos.com/index.htm
Lexel Laser Inc. (div. of Cambridge Laser Laboratories), Fremont, CA, (510)-651-0110, http://www.lexellaser.com/
LFW, Laser Focus World Buyers Guide, PennWell Corp., 2004, Nashua, NH, (603)-819-0123, http://lfw.pennnet.com/home.cfm
LG Laser Technologies GmbH, Kleinostheim, Germany, +49-(0)6027-46620, http://www.lg-lasertechnologies.com/
LIA, Laser Institute of America, Orlando, FL, (800)-345-2737, http://www.laserinstitute.org/
Limo Lissotschenko Mikrooptik GmbH, Dortmund, Germany, +49-(0)231-222410, http://www.limo.de/
LUMERA LASER GmbH, Kaiserslautern, Germany, +49-(0)6301-703-181, http://www.lumera-laser.de/index2.html
Lumics GmbH, Berlin, Germany, +49-(0)30-6780-6760, http://www.lumics.de/
Lumitek Intl. Inc., Ijamsville, MD, (301)-831-1001, http://www.lumitek.com/Sensorcatalog.PDF
Mauna Kea Technologies, Paris, France, +33-(0)1-4824-0345, http://www.maunakeatech.com/
Melles Griot, Laser Div., Carlsbad, CA, (800)-645-2737, http://www.mellesgriot.com/contactus/default.asp
Micro Laser Systems, Garden Grove, CA, (714)-898-6001, http://www.microlaser.com/blueConfocal.html
MWK Industries, Corona, CA, (909)-278-0563, http://www.mwkindustries.com/
Newport Corp., Irvine, CA, (800)-222-6440, http://www.newport.com/
New Focus, Inc. (a division of Bookham), San Jose, CA, (408)-919-1500, http://www.newfocus.com/
Noran Instruments Inc., see Prairie Technologies, Inc.
Novalux Inc., Sunnyvale, CA, (408)-730-3800, http://www.novalux.com/contact.html
Oxford Lasers Inc., Littleton, MA, (978)-742-9000, http://www.oxfordlasers.com/
Optronics Technologies S.A., Athens, Greece, +30-(0)210-983-7121, http://www.optronics.gr/
OZ Optics, Ltd., Carp, Ontario, Canada, (800)-361-5415, http://www.ozoptics.com/
PHC5 28/11/2005 6:15 PM Page 105
Laser Sources for Confocal Microscopy • Chapter 5
105
TABLE 5.3. (Continued)
P.A.L.M Microlaser Technologies AG, Bernried, Germany, +49-(0)8158)-99710, http://www.palm-microlaser.com/
Photonic Products Ltd., Bishops Storford, Hertfordshire, UK, +44-(0)1279-719190, http://www.photonic-products.com/company_ov.htm
Picoquant GmbH, Berlin, Germany, +49-(0)30-6392-6560, http://www.tcspc.com/
POC, Physical Optics Corp., Torrance, CA, (310)-320-3088, http://www.poc.com/
Point Source, Hamble, UK, +44-23-80-744500, http://www.point-source.com/products.asp
Power Technology Inc., Little Rock, AR, (501)-407-0712, http://www.powertechnology.com/pti/index.asp
Prairie Technologies Inc., services Noran microscopes, Middleton, WI, (608)-662-0022, http://www.prairie-technologies.com/index_files/page0005.htm
Quantronix (see Excel/Quantronix)
R&D Ultrafast Lasers Kft., Budapest, Hungary, +36-(0)1-392-2582, http://www.szipocs.com/products.php
Rainbow Photonics AG, Zürich, Switzerland, +41-0)1-445-2030, http://www.rainbowphotonics.com/
Research Electro-Optics, Boulder, CO, (303)-938-1960, http://www.reoinc.com/
Rockwell Laser Industries, Inc., Cincinnati, OH, (800)-945-2737, http://www.rli.com/
Roithner LaserTechnik, Vienna, Austria, +43-(0)1-586-52430, http://www.roithner-laser.com/
Sacher Lasertechnik Group, Marburg, Germany, +49-(0)6421-3040, http://www.sacher-laser.com/
Spectra-Physics Lasers and Photonics , Inc., Mountain View, CA, (800)-775-5273, http://www.spectra-physics.com/ and Darmstadt, Germany, +49-(0)61517080
StockerYale, Inc., Salem, NH, (603)-893-8778, http://www.stockeryale.com/i/lasers/index.htm
Teckhnoscan Laser Systems, Novosibirsk, Russia, +7 (0)3832-397224, http://www.tekhnoscan.com/english/index.htm
Toptica Photonics AG., Martinsried, Germany, +49-(0)8989-99690, http://www.toptica.com/index.php
Torsana Laser Technologies A/S, Skodborg, Denmark, +45-(0)4556-0056, http://www.torsanalaser.com/index.html
Time-Bandwidth Products, Inc. Zürich, Switzerland, +41 (0)44-445-3120, http://www.tbwp.com/Time_Bandwidth/Home/FrameSetHome.htm
TuiLaser AG, Germering/Münich, Germany, +49-(0)89-894070, http://www.tuilaser.com/products/lasers/index.htm
Uvex, Smithfield, RI, (800)-343-3411, http://www.uvex.com/
Visitech International Ltd., Sunderland, UK, +44-(0)191-516-6255, http://www.visitech.co.uk/index.html
Semiconductor or Diode Injection Lasers
Charge carriers in a semiconductor material can be pumped either
optically or by an electric current (Fig. 5.11). Electronbeam–pumped semiconductor lasers can emit 3 to 5 W in the 480
to 650 nm range (Hobbs, 1993), but the most common excitation
method simply starts with a diode not unlike that in a lightemitting diode (LED) and uses an externally applied current in the
forward-biased direction. This process can have an efficiency as
high as 80%. To do so, the electrons must be confined not to a bulk
2s
4s
635.5 nm
636 nm
4f
533.7 nm
537.8 nm
1
2 S
0
543 5 nm
594.1 nm
611.9 nm
632.8 nm
5s
6g
4p
5d
3
2 S
1
5s
1532 nm
325
3p
353.6
1152 nm
441.6 nm
5p
+
HELIUM
NEON
FIGURE 5.9. Schematic energy diagram for a helium–neon laser. Helium
functions as a pump energy transferring buffer for neon.
HELIUM
+
CADMIUM
FIGURE 5.10. Schematic energy diagram for a helium–cadmium laser. Green
and red emission is also possible but not commercialized.
PHC5 28/11/2005 6:15 PM Page 106
106
Chapter 5 • E. Gratton and M.J. vandeVen
CONDUCTION
BAND
BANDGAP, EBG > ª kT
VALENCE
BAND
STIMULATED EMISSION
FIGURE 5.11. Schematic energy diagram for a semiconductor laser. It is
based on laser emission from a forward biased p-n junction when driven by a
well-regulated drive current.
volume (Fig. 5.12), but to a stack of very thin layers forming multiple quantum wells (MQW), or planes of quantum wires or dots
(super lattice). Due to the small confined volume a high radiative
efficiency exists as well as a low lasing threshold. Most devices
still operate in NIR but the trend is to develop diode lasers using
wide bandgap materials that have an output below the red. Blue
diode lasers based either on ZnSe or doubling 860 nm light, emit
at around 430 nm and are about to enter the commercial market.
At higher power levels (Figueroa, 2002), direct frequencydoubling in the diode laser forms an alternative route to the blue
region. For example, the D3™ (Direct Doubled Diode laser, Coherent) delivers 10 mW at 430 nm. When selecting a drive current
source, it is important to select one with a low noise, good stability, and including a temperature controller on the diode laser head.
~ 500 mm
METAL
OXIDE
p-TYPE
lasing channel
n-TYPE
SUBSTRATE
METAL
ELLIPTICAL
OUTPUT
< ~ 1 mm
Diode lasers can change their emission wavelength over a limited
range (10–20 nm) by varying their drive current and junction temperature (Hodgson, 1994). The approximate tuning rate is about
0.1 nm /°C. Cooling brings the lasing wavelength down. Beam
quality suffers from astigmatism that has to be optically corrected
(Snyder and Cable, 1993). Edge emitters now provide high power,
up to several watts CW in NIR.
Semiconductor lasers are very appealing because they are
small [Fig. 5.13(B)], highly efficient, easy to use, and relatively
cheap. Integrated fiber-optic output is another feature available
from many manufacturers [Fig. 5.13(A)]. However, it should be
stressed that this small package comes with important special
requirements. The devices can be rapidly destroyed if both current
transients or nanosecond current spikes at start-up and internal
heating are not kept under control by the power supply electronics. Static discharges (SD) from a person or an ungrounded soldering iron, or the use of solder that is too hot or remains in contact
for too long may instantly destroy the laser. A mechanical shunting device (Unger, 1994) may prevent SD damage during handling. Alternating current (AC) line filters are recommended
(Hodgson, 1994). This market is strongly driven by the digital
video disk (DVD) and audio compact disk (CD) industry where
the goal is to increase information storage densities. Most, if not
all, of the following examples are also wavelength stabilized by
stacks of multi-layer coatings usually deposited at the HR side and
producing bandpass filter reflectivity only for the lasing wavlength
[Fig. 5.14(B,C)]. Combining a proper OC coating with a fiber
pigtail having an inscribed Bragg grating has the same stabilizing
effect.
A
B
LASER DIODE WITH
DBR FIBER OPTIC PIGTAIL
2 mm
LASER DIODE
C
FRONT AREA
~ 100 x 100µm
CLEAVED AND
POLISHED FACES
ACTING AS
RESONATOR MIRRORS
INDEX-GUIDED DOUBLE HETERO-JUNCTION
EDGE EMITTING FABRY-PEROT TYPE
SINGLE-STRIPE DIODE LASER (FRONT VIEW)
FIGURE 5.12. Dimensions and cross-section through a edge emitting diode
laser. The polished sides act as highly reflecting mirror and output coupling
mirror. Thermo-electric cooling to remove heat required.
CIRCULAR BEAM
WITH
ANAMORPHIC OPTICS
ELLIPTICAL, COLLIMATED BEAM
WITH
ASTIGMATISM CORRECTING OPTICS
FIGURE 5.13. Example of (A) optional pig-tail fiber-optics, (B) physical
dimensions of a diode laser, and (C) corrective optics to create a circular beam
profile.
PHC5 28/11/2005 6:15 PM Page 107
Laser Sources for Confocal Microscopy • Chapter 5
A
GRATING
HR
LENS AR
OC
COATING
EXTENDED CAVITY DIODE LASER (ECDL)
B
FREQUENCY SELECTING BRAGG GRATING INTEGRATED IN ACTIVE MEDIUM
DISTRIBUTED FEEDBACK (DFB) DIODE LASER
C
REFLECTOR COUPLER GAIN SECTION
PHASE ADJ. SECTION
DISTRIBUTED BRAGG REFLECTOR (DBR) DIODE LASER
FIGURE 5.14. Diode lasers with wavelength tunability via an angleadjustable grating in an extended cavity configuration (A); improved mechanical and thermal stability as well as wavelength and intensity control via a
bandpass reflective filter formed by a stack of thin coating layers (Bragg
grating). When the grating is incorporated into the gain medium the device is
a distributed feedback (DFB) diode laser (B), otherwise it is a distributed Bragg
reflector (DBR) diode laser. For ultrafast pulsed operation semiconductor saturable absorber (SESAM) layers can be added to the high reflecting mirror
stack or to the saturable output coupler (SOC) to the output coupling face. Edge
emitting diodes require corrective optics because the emission is elliptic.
Violet and Deep Blue Diode Lasers
Because these lasers produce 10 mW of deep blue emission
(395–440 nm) from an input power of less than 10 W, they do not
require water cooling. Compare this with an argon-ion laser producing the same 10 mW from 1 kW of electrical input, gallons of
cooling water per minute, and filling a good part of a laser table.
Most of the diode lasers have a single fixed-wavelength, which may
vary somewhat (see above under FBG Wavelength Stabilization).
Violet 405 nm (Photonic Products Ltd., Table 5.3) lasers efficiently
couple into 3 mm core single-mode fibers. Diode lasers presently
operate at 395 to 440 nm (LG Laser Technologies GmbH,
Coherent; Table 5.3), 415 and 430 nm (Crystal Laser Inc.), 440 nm
(PicoQuant model LDH 440, Power Technology Inc.), 457 nm
(Melles Griot), and 473 nm (Crystal Laser Inc., Power Technology
Inc.). But they are not nearly as cheap as laser-pointers! Prices
range from a few thousand to US$25,000, depending on complexity and features such as thermo-electric cooling.
107
SHG doubling of ~980 nm IR light. Similar specifications are
found for the Starbright 488, 10 mW pumped at 40 W (Torsana
Laser Technologies A/S); the BluePoint 488 at 2 to 5 mW from
Rainbow Photonics AG; and the tunable 488 nm laser from Toptica
Photonics AG. Coherent introduced the Sapphire (NOT to be
confused with a Ti : Sa system). It is a VECSEL GaAs diode laser
consuming 30 to 60 W (mostly in the thermo-electric cooler),
optically pumped at 976 nm, and intra-cavity frequency-doubled to
provide 20 mW of electrical input.
The Novalux Inc. Protera is based on an ECSEL design with
a three-mirror folded cavity for better mode control and uses a
KNbO3 doubling crystal inside the laser cavity.
491 nm Diode Lasers Some designs are based on up-conversion
of a fiber laser to produce 10 mW at 491 nm (Guided Color Technologies and Lumics GmbH). Others use a different principle. The
DPSS Dual Calypso laser (Nordborg and Karlsson, 2004) from
Cobolt AB (Table 5.3) offers 491 nm at 20 mW and 532 nm at
50 mW simultaneously. It uses a periodically poled potassium titanium oxide phosphate (KTP) crystal, partly inside the laser cavity,
for non-linear optical frequency conversion. The 491 nm line stems
from sum frequency mixing of Nd : YVO4 with the 1064 nm line
from the Nd : YAG laser. Four hundred fifty-seven and 1340 nm
lines can be added using similar methods. This laser has a noise
level of <0.3% and can directly replace the argon-ion laser operating at 488 and 514 nm.
606 and 635 nm Diode Lasers The He-Ne 594 nm orange and
the red 632.8 nm lines can be replaced by the 606 nm (orange)
VECSEL Eksel 110 diode laser (Toptica Photonics AG; Häring and
Gerster, 2003) and the ChromaLase 635 25 mW (Blue Sky
Research). The Radius 635 nm CW 25 mW (Coherent) has a
16-h beam pointing stability of about 60 mrad compared with
~160 mrad for a 17 mW He-Ne. A 637 nm red diode laser He-Ne
laser substitute has been incorporated into the Bio-Rad Radiance2100™CLSM (currently Carl Zeiss CellScience Ltd).
685 nm Diode Laser A 685 nm 30 mW CW diode laser is offered
by Lasos Laser Technik GmbH.
Recent Developments in Diode Lasers
In this expanding field, many features are rapidly improving to the
benefit of the user.
Visible and Red Diode Lasers
Emerging Tunability By incorporating a diode laser into an
external cavity equipped with a tunable grating (a Littman-Metcalf
cavity) wavelength tunability from 632.5 to 1630 nm within a
50–150 GHz bandwidth is now available (New Focus; Table 5.3;
Scifres, 1994). These devices are listed as external or extended
cavity diode lasers [ECDL; Fig. 5.14(A)]. A narrow-line width
version has been described in Day and Dessau (1994). Although
for the moment manufacturers seem to focus on single wavelength
devices, mechanical and thermal stability should not be compromised.
488 nm Diode Lasers An alternative to the argon-ion laser has
appeared in the form of the Protera™ 488 system from Novalux
Inc. (Table 5.3). Listed advantages over the argon-ion laser include
twice the life expectancy (20,000 h), 1% of the power consumption, up to 20 mW output power, 2% of the size, 0.2% rms noise
level, and built-in thermo-electric Peltier cooling with conductive
heat removal via the enclosure. The ChromaLase 488™ from Blue
Sky Research has a typical power consumption of 2 W and output
power from 1 to 25 mW output and creates 488 nm laser light by
Beam Quality and Delivery Diode lasers typically require
extensive astigmatic and anamorphic corrective optics to obtain
a circular, parallel beam [Fig. 5.13(C)]. Other methods include
using a spatial filter or a single-mode fiber-optic as well as beam
homogenizers, such as the monolithic V-step design by Ingeneric
GmbH (Table 5.3), arrays of symmetrically arranged lenslets, the
so-called “fly’s-eye” lenses or light integrators (Homburg et al.,
2003), available, for example, from Limo Lissotschenko Mikroop-
PHC5 28/11/2005 6:15 PM Page 108
108
Chapter 5 • E. Gratton and M.J. vandeVen
40 mm
3 mm
0.2 mm
CIRCULAR CROSS-SECTION
OUTPUT
METAL CONTACT
p-TYPE DBR OC
MQW LASING LAYERS
n-TYPE DBR HR
5 mm
SUBSTRATE
METAL CONTACT
MULTI-QUANTUM WELL (MQW)
VERTICAL CAVITY SURFACE EMITTING LASER DIODE (VCSEL)
FIGURE 5.15. Dimensions and layout of a vertical cavity semiconductor
diode laser (VCSEL). Stacks of thin coating layers form the distributed Bragg
reflector (DBR) highly reflective mirror (HR) and the output coupling mirror
(OC). The emission pattern is circular and can be near-Gaussian.
tik GmbH (Table 5.3). Optical elements, called graded index
(GRIN) lenses, have also been introduced to eliminate spherical
aberration in an elegant way (Carts, 1994). The vertical-cavity,
surface-emitting semiconductor laser (VCSEL; Fig. 5.15), a
semiconductor laser with good circular-pattern beam quality
(Cunningham, 1993). Power levels are now reaching 100 mW with
a wall-plug efficiency of 20%, high temperature stability, and good
beam quality.
Output Power and Cooling Output power has been increased
by building diode lasers as banks or arrays but these often require
additional water cooling that adds to their complexity. Quasi-CW
output levels can reach 300 mJ and 1.5 kW peak power. Tens to
100 s of watts of CW power and long lifetimes are becoming
common. Cooling requirements depend on the output power generated and they range from passive air cooling via thermo-electric
(Peltier) cooling to water cooling, the latter often using microchannels for optimum control of the junction temperature.
Temperature Tuning There is no longer a need to have a large
number of different diode lasers nor a large collection of emssion
filters for CLSM. Sivers and colleagues (2004) showed that,
with a 3 mW low-power superlattice (multi-quantum well) 635 nm
diode laser attached to a Bio-Rad MRC600 CLSM, many cell
stains that absorb in a broad range around 640 nm can be reached
by cryogenically cooling the laser down to -196°C. This caused
the emission wavelength to shift linearly ~25 nm down, while the
output power went up about five times. For equal fluorescence
signal, the noise level in the image caused by reflected laser light
decreased five times because it was possible to optimize the laser
wavelength to the optical filters more precisely.
Output Modulation Another important feature of semiconductor lasers is that their output can easily be modulated to well
above 100 MHz and some can reach tens of gigahertz. This makes
them prime candidates for use in frequency-domain fluorescence
lifetime imaging microscopy (FLIM; see Chapter 27, this
volume).
Diode- and Lamp-Pumped Solid-State Lasers
(DPL and DPSS)
A second class of solid-state lasers includes those using
neodymium yttrium aluminum garnet (Nd-YAG) and lasing in the
IR at 1064.2 nm with very high fracture resistivity and good
thermal conductance (Fig. 5.16); Yytrium lithium fluoride (NdYLF) with low thermal lensing (Arrigoni, 2004b); Nd-BaYF with
better subpicosecond generation and amplification properties;
yttrium orthovanadate lasing at 1064.3 nm having a lower lasing
threshold than Nd-YAG (Kaminskii, 1981); GdVO4 lasers and
Nd:LuVO4, also having a lower lasing threshold. All of the materials listed above are pumped by a FBG diode lasers [Fig. 5.14(C)].
Nd-YAG and Nd-YLF provide high overall efficiency and the good
beam quality necessary for frequency doubling to the visible producing lines at 532 and 523 nm. Microchip lasers are formed by
sandwiching laser and frequency doubling crystals into a millimeter size compact structure. Due to their small size they benefit from
ceramic laser materials with high thermal conductivity (Wisdom
and Digonnet, 2004).
A simpler design is possible with ytterbium tungstate
(Yb:KGW) absorbing between 900 and 1000 nm that can be
directly pumped by a red diode laser.
Early models used flash lamp or CW krypton lamp pump
sources but the latest models tend to be equipped with a diode laser
[diode-pumped solid-state lasers (DPSS); Hobbs, 1994]. The
advantages of diode lasers over conventional pump sources, such
as lamps, are reduced cooling requirements because of higher efficiencies due to a collimated and focused output, a perfect match
of pump-source emission wavelength with the absorption spectrum
of the lasing medium, and enhanced frequency stability (Baer,
1986). A typical improvement in electrical-optical conversion efficiency is from 0.5% (TEM00 mode) to 6%.
4
4
S , F
3/2
7/2
4
4
F , H
5/2
9/2
4
F
3/2
730 nm
800 nm
1064 nm
4
I
4
11/2
I
9/2
3+
3+
Nd : YAG and Nd :YVO4
FIGURE 5.16. Schematic energy diagram for a neodymium–yttrium–aluminum garnet (Nd-YAG) laser and a neodymium–yttrium orthoivanadate
(Nd:YVO4) laser.
PHC5 28/11/2005 6:15 PM Page 109
Laser Sources for Confocal Microscopy • Chapter 5
The infrared output at 1064 (Nd-YAG) or 1054 nm (Nd-YLF)
is easily frequency-doubled, tripled, or quadrupled with appropriate non-linear crystals such as lithium triborate (LBO) and beta
barium borate (BBO) to provide radiation at 532 and 527 nm
(SHG), 355 and 349 (THG), 266 and 262 nm [fourth harmonic generation (FHG)], respectively. BBO even allows generation of the
fifth harmonic at 213 nm. The stability of these lasers is very good
and the green emission wavelength is ideal for the excitation of
rhodamine dyes or for pumping the tunable solid-state lasers
described below. Nd-YAG also has lower-gain wavelengths at
1440, 1320, and 946 nm, which when doubled provide 720, 660,
and 473 nm. Power levels sufficient to obtain frequency doubling
create a blue laser suitable to replace the ion laser lines. Argon-ion
514 nm emission can be replaced with a frequency-doubled
532 nm DPSS design when the mode and beam characteristics are
optimized. A 473 nm DPSS laser (CrystaLaser Inc. and National
Laser Corp. with 5–10 mW, Table 5.3) is created by intracavityKNbO3 doubling the 946 nm IR light of a diode-pumped Nd:YAG
laser. Typical cost is about US$8000.
As a pump source, the 514 nm line of a large frame argon-ion
lasers may well be replaced with a frequency doubled VersaDisk515 laser from Electronik Laser Systems GmbH (Table 5.3) emitting, depending on the model, 2.5 to 15 W of 515 nm light in a
small foot print without water cooling, and improving the Ti:Sa
pump efficiency by 20% compared with the more common 532 nm
DPSS pump source. Several companies now offer DPSS systems
in kit form. Examples are the DPSS educational kit from Optronics Technologies S.A. and the Nd:YVO4 and Nd:YAG kits with
Cr:YAG SESAM from ALPHALAS GmbH. These lasers are
extremely compact, stable, and efficient with good beam quality
and offer turn-key operation. Electrical power requirements are
low: tens of watts. Beam quality can be improved with Gaussian
resonator mirrors and phase plates (Casperson, 1994).
Thin Disk Lasers
Very intense lines with much reduced thermal lensing and birefringence can be generated by thin disk lasers. Radial thermal
effects are most severe in the rod-type lasing materials used in the
original Nd-YAG designs. Slab-type designs largely lift this
thermal problem. Scientific designs are profitting from Yb:YAG or
similar based thin disk systems such as the VersaDisc by Electronik Laser Systems GmbH, which can attain tens to hundreds of
watts of output power with minimal heating effects. It is based on
a 100 mm thin disk of Nd:YAG, Yb:YAG, or other material bonded
to a heat sink and optically pumped on the opposite side. Because
the heat gradient is almost perfectly planar (i.e., parallel to the
bonding surface), the thermal lensing strongly present in highpower, rod-type media is drastically reduced along with the
amount of birefringence. Optical aberrations are not introduced.
As the disk is small and thin, a special mechanical mirror arrangement makes the pumplight impinge on it many times. Lasers such
as the VersaDisk currently deliver up to about 100 W CW at
1030 nm and about 15 W at 515 nm (Hitz, 2004a).
Tunable All Solid-State Laser
The release of the CW titanium-sapphire laser was an important
event (Hammerling et al., 1985). These lasers can be pumped
by a CW ion, Nd-YLF, Nd-YAG, or diode laser. Because their
vibrational-electronic levels are spread in a broad band, laser
transitions can take place over a wide range, and, given the right
mirrors, the tuning range extends from 700 to 1000 nm (Fig. 5.17).
Power output is 3.5 W at 790 nm when pumped with an 18 W Nd-
109
2
E
PUMP
ENERGY
660 – 1180 nm
2
T
2
3+
Ti
VIBRONIC LASER
FIGURE 5.17. Schematic energy diagram for the Ti:Sa four-level vibronic
laser.
YVO4 or 20 W all-line argon-ion laser. The Ti:Sa laser has virtually replaced the IR dye lasers; it is much easier to operate, no dyes
have to be changed, and the long-term stability is better than 3%
due to the elimination of flowing dye. The RMS noise between
10 Hz and 2 MHz is less than 2%, a factor of about 10 less than
for dye lasers. The main characteristics of these systems are the
generation of short pulses of about 100 fs (FWHM), and high peak
power for frequency-doubling to the blue and near-UV spectral
range and for use in two-photon CLSM. Several configurations
have appeared resembling either CW ring or standing-wave dye
lasers. To cover the complete spectral range of this laser special
mirror sets are required.
A similar wide-tuning range and high average power is offered
by the Alexandrite laser (Cr3+ in BeAl2O4 host), which covers the
730 to 826 nm region when pumped CW at room temperature. This
laser can also be flash lamp pumped. The Forsterite laser (Cr4+ ions
in a Mg2SiO4 host material) emits between 1167 and 1345 nm and
can be used in the 1200 to 1250 nm tissue penetration window.
Its three main pumping bands are centered around 350 to 550, 600
to 850, and 850 to 1200 nm. Both CW and pulsed models are
available. Its second harmonic is tunable from 600 to 650 nm
(Mortensen, 1994). Liu and colleagues (2001) report multi-photon
femtosecond excitation of plant tissue at about 1250 nm with a
Cr:forsterite laser (see also Chapter 21, this volume). Potential
members of this family of tunable solid-state lasers are the LiSAF
laser (Cr in LiSrAlF6) with a tuning range from 780 to 1060 nm
and a peak emission at 825 nm (Perry et al., 1993), and the LiCAF
laser (Cr in LiCaAlF6) tunable from 720 to 840 nm. Tunable UV
could be created by Ce:LiSAF (Anderson, 1994b).
CW Fiber Lasers and Up-Conversion
Confusingly, fiber-coupled lasers in which a fiber-optic merely
guides the emission are also called fiber lasers. Although also
capable of generating a white continuum, micro-structured photonic crystal fiber (PCF) laser delivery systems should not be classified as fiber lasers. Unlike single-mode fibers, highly multi-mode
PHC5 28/11/2005 6:15 PM Page 110
110
Chapter 5 • E. Gratton and M.J. vandeVen
core fibers are insensitive to twisting and bending but otherwise
behave like single-mode fibers. PCF fibers designed to have better
higher order dispersion compensation include the crystal fiber
large mode area (LMA) PCF.
We will only use the term fiber laser for systems where the
laser gain medium is the fiber itself. IMRA America (Table 5.3)
shows that the design of a fiber laser can be very simple: Two resonator mirrors with the fiber-optic in between (Fig. 5.18). Doubleclad fibers are preferred for fiber lasers because the pump light can
be efficiently coupled into the high NA, ≥0.46, inner cladding
while only a single, transverse-mode-doped core, having a low NA
of about 0.06, is excited. They also possess excellent heat dissipation (Limpert et al., 2004). Furthermore, using LMA doubleclad fiber (for example, 20/400 mm core/outer cladding) reduces
non-linear Brillouin and Raman effects (Galvanauskas and
Samson, 2004). PCF fibers can accommodate high powers
(Folkenberg and Broeng, 2004), with 260 W CW output from a
short piece of 4 mm double-clad fiber, thereby also limiting nonlinear effects.
PUMP
ENERGY 2
LASER
EMISSION
FAST
PUMP
ENERGY 1
ION 1
Up-Conversion Fiber Lasers
Theses devices are based on excited state absoption of rare earth
ions in fluoride “ZBLAN” glass fibers (Fig. 5.18). ZBLAN stands
for zirconium, barium, lanthanum, aluminum, sodium fluoride.
When pumped by a pair of diode lasers, each operating at a different red wavelength, several milliwatts of green light can be
created (Piehler, 1993). In this typical up-conversion scheme, lowenergy photons are sequentially absorbed and create a population
inversion in the excited state (Chenard, 1994). By doping the fiber
with more than one rear-earth material (Fig. 5.19), the spectral
region longer than ~500 nm can be covered. Examples include
doping with Er(bium) for 546 nm, with Pr(aseodymium) to obtain
491, 520, 605, 635, and 720 nm emission at 10 to 40 mW (Guided
Color Technologies, GmbH; Table 5.3), with PrYb for 840 nm,
with T(huliu)m for 482 and 800 nm as well as 470 nm and 1735 ±
1 nm from a single-frequency Tm-doped DFB fiber laser (Hitz,
2004b), with Tm for 980 nm, and with Yb for 1060 nm. One
more example is the 700 blue fiber laser series from LasNova
(Table 5.3).
Lumics GmbH has developed a fluorozirconate (ZBLAN
glass) fiber laser that is a potential replacement for argon-ion
lasers. The spectral linewidth of a fiber laser is typically very small,
about 0.02 nm (Gabler, 2004). One notes that the exact wavelengths at times differ by 1 to 2 nm between various designs, a fact
which may be related to doping and operating conditions.
In the IR, fluoride glass fiber can be doped with rare-earth ions,
such as erbium, which emits at 1540 nm (Hecht, 1993a). The
ION 2
UPCONVERSION LASER
FIGURE 5.19. Schematic diagram for a dual ion doped up-conversion laser.
energy stored in the single-mode laser is about 500 mJ and, in a
multi-mode laser, 10 mJ. The maximum optical peak power
amounts to about 100 kW. Although initial tests were apparently
carried out with an IMRA 775 nm 100 fs laser on a Zeiss microscope, no commercial system has yet been offered by Zeiss.
In a single-longitudinal-mode fiber laser with a fiber core of 2
to 4 mm, the interaction area is very small (a few square micrometers). The interaction length is very long, from millimeters to even
kilometers. This results in a very high single-pass optical gain at
very low threshold (sub-mW) pumping power levels with pumping
efficiencies to 75%. Fibers consist of amorphous material that give
very broad absorption bands (Chenard, 1994). The single-pass gain
is practically unity. Output is always diffraction limited with
excellent beam quality. In combination with an easily attainable
average power level of a few watts this makes them very suitable
for laser-tweezer applications. These advantages of a fiber laser
were recognized as early as 1964. However, at that time there was
little capacity to produce good quality, low-loss, single-axial-mode
fiber.
PULSED LASERS
DIODE
LASER
PUMP
HR
LENS
ZBLAN FIBER
OC
OUTPUT
LENS
FIGURE 5.18. Schematic diagram of a dual ion doped CW ZBLAN upconversion fiber laser. Pumped by a distributed feedback (DFB) diode laser is
emits visible laser light. The connectors contain the high reflecting mirror (HR)
and the output coupling mirror (OC).
Why are pulsed lasers useful? The simple answer is higher peak
power. Sufficiently powerful CW pump sources are often not available. Often the wavelength and intensity offered by available lasers
are not appropriate. Therefore, wavelength conversion schemes
have been developed and many of them are based on non-linear
phenomena that occur only at very high intensity. Because of their
high peak power, pulsed systems provide efficient wavelength conversion and tuning capabilities. In general terms, the peak intensity is greater than the average intensity by a factor that is the
inverse of the duty cycle. The duty sycle is the ratio of pulse width
to pulse interval and can often be 10-5.
It is important to remember that pulsed lasers are only useful
in scanned fluorescence microscopy if the pulses occur at a high
PHC5 28/11/2005 6:15 PM Page 111
Laser Sources for Confocal Microscopy • Chapter 5
enough rate that many pulses occur in each digitizing period and
intensity integration over the pixel period can provide a performance similar to a CW laser. Pulsed lasers are characterized by
the high peak power (see Table 5.2).
Classification of Pulsed Laser Systems
A logical way to classify pulsed laser systems is according to their
temporal pulse behavior. Four methods for pulsed operation exist:
normal mode, Q-switching, mode-locking, and cavity-dumping.
• Normal
•
•
•
mode, free-running: When observed with a fast
diode and an oscilloscope, solid-state lasers generate a continuous train of random spikes with varying intensity caused by
the interaction of various modes in their very small cavities.
This type of operation is not very useful.
Q-switched lasers: The term Q-switched is derived from the
radio and microwave terminology, and it relates to a forced
change in the quality factor, Q, of the resonant cavity (i.e., a
bell is “high-Q,” a lump of wet clay is “low Q”). A low Q resonator is one which hardly supports laser action. What is
needed is a method of changing the gain, or resonance, of the
laser cavity. A simple implementation of a Q-switching device
is the installation of an optical shutter in the resonator. The
sudden opening of the shutter releases all the energy stored in
the excited electrons of the medium in one giant optical pulse.
Active, mode-locked: In a resonant cavity, many simultaneously oscillating optical modes are present at the same time
(Arecchi, 1972; Mollenauer and White, 1987). Usually these
modes interact with each other in a random way and no phase
relationship exists between them. The output power fluctuates
randomly according to a time-dependent interaction between
the various modes. Installation of an intra-cavity modulator
(i.e., a high-speed shutter) with a resonance frequency exactly
matching the round trip time, 2L/c, in the laser resonator will
lock all modes to that particular resonance frequency. A certain
phase relationship is maintained. Throughout the cavity, a
single, intra-cavity pulse is created at a pulse rate of 70 to
100 MHz. Mode-locking frequencies can be changed by using
a crystal with different mode-locking frequency and simultaneously adjusting the laser cavity length. Thermal stability has
to be excellent. Folded compact cavity design, further miniaturization, a one-box design strategy, direct IR diode pumping
using ytterbium tungstate, and standard thermo-electric Peltier
cooling help maintain proper mode-locking. For fiber lasers,
smaller size LiNbO3 Mach-Zehnder modulators are used
because they possess a lower insertion loss and better control
of chirp.
Passively mode-locked lasers: In passive mode-locking a
very high loss is introduced to interrupt CW operation.
Kerr Lens Mode-Locking (KLM)
One way to generate short pulses from a solid-state Ti:Sa laser is
using Kerr lens mode-locking. The Kerr effect refers to the
quadratic is non-linear dependence of the refractive index of a
non-linear material on the applied electrical field. The intensity
dependent refractive index induces a linear frequency sweep on
any short optical pulse leading to spectral broadening. This effect
can be compensated by sending the pulse through a few millimeters of glass or short piece of fiber-optic. High intensity optical
pulses exert strong forces on the valence electrons of the laser
medium. The transverse gradient in a Gaussian beam profile delivers the highest intensity near the center of the beam, less so near
111
the perimeter. Kerr-induced refractive index effects cause the beam
to contract producing self-focusing. This regular focusing pattern
in the gain medium imposes a specific set of modes in the Ti:Sa
laser and eliminates other non-matching modes. An important consequence is that the gain medium works like a master oscillator
with its own set frequency. This method requires an intracavity slit
or pinhole to suppress CW contributions and needs an active starter
such as a noise spike created by tapping a mirror mount and
otherwise perturbing the cavity.
Saturable Bragg Reflector (SBR) and
Semiconductor Saturable Absorber
Mirrors (SESAMs)
Creation of self-starting intense optical pulse trains in, for
example, a Nd-YVO4 or fiber laser running at a high frequency of,
for example, 80 MHz, is implemented by introducing into the
cavity a low-loss, super-lattice, saturable-crystal absorber such as
InGaAs. The material possesses a low reflectance for weak optical
signals (such as noise) and a high reflectance for high power
signals (such as optical pulses) and has a very short recovery time
(saturable absorber mirror, SAM). It is incorporated into a highreflectivity Bragg mirror. Quantum wells can be used in place of
the InGaAs. The wider the spectral bandwidth of the gain medium,
the shorter the optical pulse generated and, in addition, shorter
pulses are created when a shorter rod is used because dispersion
can be better compensated. For normal dispersion, red is ahead of
blue light. For anomalous dispersion the opposite is true. Transparent materials usually possess normal dispersion. Because a
Ti:Sa rod introduces normal dispersion, it must be compensated
by inserting anomalous dispersion components such as prisms,
grating pairs, or chirped mirrors into the cavity. In ultrafast Ti:Sa,
Cr:LiSAF, or fiber lasers, pulse durations of a few tens of femtosecond can readily be created. Mode-locked oscillators such as
an ultrafast Ti:Sa laser create nanojoule pulse energies.
The latest oscillator addition, Yb:KGW (tungstate), is directly
pumped by two fiber-delivered high-power 940 or 980 nm IR diode
pumps (Eclipse, Spectra Physics). No prism pair is required for
compensation of the positive GVD because the cavity contains
dispersion-compensating coatings. In addition, an SBR Bragg high
reflector provides self-starting and passive mode-locking. At 1048
nm, 100 mW average power is generated with a repetition rate
of 80 MHz and a pulse width of 150 fs. Components are readily
available (BATOP GmbH; Table 5.3). Similarly, a microchip
Nd:YAG/Cr:YAG gain medium/absorber can be designed (Kraft,
2004).
Saturable Output Coupler (SOC)
The SAM can be replaced by a saturable output coupler (SOC)
incorporated into the cavity. A passively mode-locked DPSS laser
creates a very simple design because the pump power can enter
through the high reflector into the laser medium.
Cavity-Dumped Lasers
Cavity dumping is a common technique for reducing the laser repetition rate. For example, for frequency-domain fluorescence lifetime imaging microscopy (FLIM) the 80 or 76 MHz laser pulse
repetition rate is too high. The repetition rate can be brought down
to 2 or 4 MHz by applying a momentary radiofrequency (RF)
signal to an appropriate modulator crystal. This allows more harmonics to be used for FLIM imaging and optimizes the system for
the nanosecond or or picosecond lifetimes encountered.
PHC5 28/11/2005 6:15 PM Page 112
112
Chapter 5 • E. Gratton and M.J. vandeVen
Nitrogen Lasers
Among the more common pulsed lasers are nitrogen lasers.
Because of their low pulse repetition rate (20–100 pps), nitrogen
lasers are only used in microscopy for intentionally wounding cells
(Burgess, 2004). The average power is between 1 mW to 1 W,
and the emission wavelength is 337.1 nm (Hecht, 1993b). The
pulse width of these lasers varies between a fraction of a microsecond to less than a nanosecond. The time characteristics of
the pulsed nitrogen laser emission at 337 nm was exploited in fluorescence microscopy by Wang and colleagues (Wang, 1990;
Wang et al., 1990, 1992). Their conclusion was, however, that the
pulse-to-pulse intensity stability of this type of laser (500 ps
FWHM pulse at 25 Hz) is not presently sufficient and long integration times were necessary. For this reason the nitrogen laser is
unlikely to be useful in microscopy, particularly now that deep UV
diode lasers (Crystal Laser Inc.; Table 5.3) possess much more
favorable lasing characteristics. Bioeffects on living cells and
tissues of UV laser irradiation can be severe but for microdissection applications the nitrogen 337 nm line is very suitable (Burgess,
2004).
Excimer Lasers
Although the vacuum-UV and UV output of these excimer lasers
has given them a strong foothold in tissue ablation, this type of
laser is not often used for CLSM. Air-cooled, compact models are
available, such as LEXtra and COMpex models (Lambda Physik;
Table 5.3). This company also introduced a metal–ceramic based
NovaTube™ design, which lasts for 7 ¥ 108 pulses on a single gas
fill.
Metal Vapor Lasers
Metal vapor lasers (Lewis et al., 1988; Hecht, 1993d) are also
extremely powerful and can be operated at much higher repetition
rates than excimer lasers. The average power can exceed 100 W
and the repetition rate can be as high as 20 to 50 kHz, which is still
a bit slow for most microscope applications. Wall plug efficiencies
reach 1%. Their characteristic emission wavelengths are also
reported in Table 5.2.
Dye Lasers
In cancer treatment, high fluency rates are sometimes required.
Real-time CLSM following the effects of high-energy 7.5 J/cm2
585 nm pulsed dye laser irradiation on hyperplasia in vivo was
reported by Aghassi and colleagues (2000). Otherwise, the need
today for stable well-characterized tunable sources is largely filled
by optical parametric oscillators (OPO) and optical parametric
amplifiers (OPA; see below) in combination with solid-state
pumping lasers. Not only is this technology far less messy than
dye laser technology, it also adds IR coverage via difference frequency mixing (Radunsky, 2002).
Dye Laser with Intra-Cavity Absorber
Synchronously-pumped dye lasers equipped with saturable
absorbers have been used extensively in the past to obtain short
pulses to cover the gap between 525 and 700 nm, which exists
for Ti:Sa lasers (see below; Herrmann and Wilhelmi, 1987;
Muckenheim et al., 1988).
Colliding-Pulse Dye Laser
Colliding-pulse mode-locked dye lasers can emit optical pulses
of <100 fs pulse width at repetition rates of 100 MHz. They can
produce an average power of 15 mW at 630 nm when pumped with
at least a 4 W all-line argon-ion pump. Noise is less than 1% but
this figure depends critically on pump laser performance. This type
of laser was used by Denk and colleagues (1990) and Piston and
colleagues (1992, 1994) for their first two-photon experiments. It
has rather limited tunability. Thermal equilibrium is reached within
1 h after turning the system on. Ti:Sa lasers, which produce 100 fs
pulses at 100¥ higher power, have replaced colliding-pulse modelocked dye lasers.
Modulated Diode Lasers
A small current change makes a large change in the output intensity of the diode laser, allowing most CW diode lasers to be operated in pulsed mode. When the internal diode laser processes are
fast enough and stray capacitance from leads, etc., is kept low,
pulse repetition rates of several hundred megahertz can be obtained
as long as the required average output power is moderate. To generate a crisp pulse train, the cut-on and cut-off slopes must be
steep. A recent review by Landgraf (2001, 2003) lists a variety of
diode lasers and their upper frequencies. Selected laser diodes have
reached ~700 MHz, for use in determining fluorescence lifetime.
With a trigger circuit, these devices could act as a poor-man’s Ti:Sa
for FLIM.
DPSS Laser in Pulsed Mode
Doubled or tripled output could be used in tissue ablation but otherwise the pulse repetition rate is too low and the high output
power a waste for CLSM. Used as Ti:Sa pump laser, a dual rod,
tandem cavity DPPS laser can now produce up to 90 W at 527 nm.
An example of these systems producing 1 to 10 kHz optical pulse
trains with a width of a few hundred nanoseconds are the Qswitched Evolution Nd:YLF family from Coherent Inc. (Arrigoni,
2004b) and the 30 W Empower series from Spectra Physics.
Ultrafast DPSS Lasers
Many one-box compact and flexible solutions are appearing.
Examples are the Pallas and Tiger compact picosecond/femtosecond laser systems from Time-Bandwidth (Table 5.3). Direct diodepumped, passively mode-locked DPSS Nd-YLF lasers can make
70 fs pulses (Tiger-picosecond from High Q Laser Production) and
100 ps (Pallas) covering 780 to 860 nm, with a repetition rate of 50
to 200 MHz and an average output power of 500 mW.
Titanium-Sapphire and Related Ultrafast Lasers
This type of laser covers 660 (18 W pump) or 700 to 1100 nm
(Duling, 1995; Fermann et al., 2002). It is an ideal source for twoand multi-photon microscopy. The long wavelength reduces scattering and allows it to penetrate farther into tissues than visible
lasers. The fact that excitation occurs only at the focus plane
improves the signal-to-background ratio and reduces phototoxity.
Extensive overviews of multi-photon fluorescence microscopy and
applications can be found at the excellent Molecular Expressions
site (Florida University) and the Cornell site of DRBIO as well as
in Chapters 28 and 37.
To cover the entire wavelength range, several sets of mirrors
may be necessary. Most Ti:Sa systems are now pumped with low-
PHC5 28/11/2005 6:15 PM Page 113
Laser Sources for Confocal Microscopy • Chapter 5
maintenance, highly efficient 5 to 10 W solid-state diode lasers.
The self mode-locking Kerr effect of the Ti:Sa lasing rod delivers
100 fs pulses at 80 to 100 MHz repetition rate. Modifications in
design may lead to pulses as short as 20 fs or even 5 fs.1 These very
short pulses exhibit extreme pulse broadening as they pass through
dispersive media. Measures to reduce or eliminate this effect are
given above under beam delivery. The high average power combined with the short optical pulse width makes the Ti:Sa laser
almost ideal for frequency doubling and tripling to the visible and
350 to 525 nm spectral regions. Following user demand, completely automated and tunable systems are now offered. These
provide a somewhat limited tuning range to stay away from the
edges of the gain curve. Examples are the Mai-Tai family from
Spectra Physics and the Chameleon products from Coherent
(Table 5.3). Gaussian profile and broadband mirrors are standard
features.
On the other hand, several companies now offer kits to allow
users to build their own ultrafast or CW Ti:Sa laser. These kits are
much less expensive and also less complex. Still further simplification is underway by having all components on a small footprint
in an all-in-one-box design. First, the Nd:YVO4 green pump laser
was upgraded with stronger pump diodes and higher efficiency
doubling crystals. It now emits CW 18 W of 532 nm with low noise
and in TEM00 mode. One pump diode suffices, halving replacement costs. Not only can the 532 nm output pump both the Ti:Sa
oscillator and a regenerative Nd:YLF-based pulse amplifier,
halving the number of green pump sources, but also the edges of
the gain curve tuning range are more easily accessible. Thermal
lensing gets worse with increased pump power but has been overcome by adjusting the focal length of the cavity mirrors. Nevertheless cryogenic cooling may be required if systems are scaled up
further (Arrigoni, 2004b).
Heat dissipation and efficiency are areas where fiber and thin
disk lasers have a clear edge. A further simplification is underway
by using ytterbium oscillators, which can be pumped directly with
IR diode lasers, bypassing the green pump step, but their pulse
width, etc., is still not as short. Table 5.2 lists the specifications of
several commercially-available examples also in kit form. Most,
if not all, use semiconductor saturable-absorber mirrors (SAMs
or SESAMs, see above under mode-locking) to generate a selfstarting, high-repetition-rate femtosecond or picosecond pulse
train (Hogan, 2004).
113
McConnell (2004) on a Bio-Rad 1024ES scan head. Coherent
introduced a CW-pumped, cavity-dumped, regenerative amplifier
with white light continuum generation at repetition rates up to
250 kHz and 100 fs pulses. However, even this much improved
repetition rate is too slow for the usual CLSM applications. This
problem is elegantly solved by a cavity dumped OPO, pumped by
the second harmonic of a Ti:Sa laser. It delivers 30 fs pulses with
a repetition rate of up to 4 MHz between 570 and 660 nm with a
pulse energy of 13 nJ (Potma et al., 1998).
Ultrafast Fiber Lasers
To generate short pulses with high peak power, a ring cavity can
extract more energy from the pump. Bragg feedback (Friebele and
Kersey, 1994) stabilizes the laser wavelength. Mode-locked singlemode fibers produce reliable, high repetition rate, short pulses with
a small wavelength tuning capability (Duling, 1993; Smith and
Lucek, 1993). Diode pumping and proper optics are necessary for
best performance. The pump DFB diode laser, output, and amplifier stages are simply coupled by splicing devices (Fig. 5.20). This
type of laser creates a very compact, efficient, and cheap alternative source for femtosecond pulses at gigahertz repetition rates and
may soon replace the dye laser and some Ti:Sa systems. Hitz
(2004d) reports a diode-laser-pumped Yb ultrafast fiber laser
oscillator-amplifier system equipped with external grating compressor and generating 62 MHz, 100 fs with 25 W average power.
Stability has to be improved against environmental changes caused
by mechanical stress-induced birefringence and temperature
variations. At the moment, a fiber laser is one fifth the size, has the
same initial cost, but only one third the operational costs of a
Nd-YAG or Nd-YVO4 ultrafast system. The FemtoMaster1060, a
passively mode-locked soliton fiber laser is offered by FianiumFILTER
POLARIZATION CONTROL
FIBER LOOPS
OUTPUT
ISOLATOR
White Light Continuum Lasers
A rather inexpensive source of quasi-white light is the CW He-Cd
laser, which seems to the eye to emit white light. Some with a positive column discharge actually emit only three wavelengths simultaneously (325, 354, and 442 nm), while the hollow cathode design
emits from UV to NIR. Selection of five lines creates an RGB
white light laser emission (White Knight RGB Series, The Cooke
Corp.; Table 5.3).
An elegant and wider supercontinuum white light source, with
generation stretching from below 450 nm to above 950 nm, is feasible using a 1 m long photonic crystal fiber (PCF) with a 1.7 mm
core diameter illuminated by the 3 to 5 ps 800 nm pulses from a
5 kW Ti:Sa laser (Espinase, 2004). This is an important step
towards a white light femtosecond light source. A similar PCF
white continuum source with a length of 38 cm was used by
1
One can predict from the relationship Dt.Dn >1 that for a spectral width of
200 nm one obtains a 10 fs pulse at 800 nm.
DFB
WDM
DIODE
COUPLER SESAM
LASER
MODE-LOCKER
PUMP SOURCE
LENSES
TUNING COUPLER
GRATING
ULTRAFAST, MODE-LOCKED FIBER LASER
FIGURE 5.20. Schematic diagram for a femtosecond pulse mode-locked ultrafast diode laser. The emission of the fiber pigtailed distributed feedback (DFB)
diode laser is coupled to the main laser gain medium fiber via a wavelength
division multiplexer (WDM). Laser output exits from the laser resonator cavity
in a similar way. The resonator consists of a lens-coupled high-reflecting mirror
and a semicondcutor saturable absorber mirror (SESAM). The other end of the
cavity contains the grating tuning element. For polarization, pulse width and
dispersion control, extra elements are spliced in. Coupling the SESAM in a different way and lens coupling both gain medium fiber ends to the tuning grating
would create a fiber-ring laser. In a flexible and compact way, diode-laserpumped amplifier stages, pulse-width compressors, and frequency-doubling
stages can be added.
PHC5 28/11/2005 6:15 PM Page 114
114
Chapter 5 • E. Gratton and M.J. vandeVen
New Optics, Ltd., a company that has also developed modelocked, tunable Yb and Nd fiber lasers. Typical specifications are
tuning from 1040 to 1120 nm, 3 ps pulse width and 30 to 100 MHz
repetition rate. With some further development, it may become a
competitor for the Ti:Sa laser (Okhotnikov et al., 2003; Rusu et
al., 2004). Pumped by a pig-tailed diode laser, the mode-locked
ultrafast Er fiber laser FemtoFiber Scientific (FFS) from Toptica
Photonics AG (Table 5.3) produces pulses with a width of <100 fs
at 100 ± 10 MHz, optional tunability over 1150 to 1400 nm and
SHG doubling to 575 to 700 nm, 1550 nm is standard with 200 mW
average and 18 kW peak power; the Femtolite family Er fiber laser
and Yb-fiber based chirped pulsed amplifier FCPA mJewel series
from IMRA America Inc.
WAVELENGTH EXPANSION THROUGH
NON-LINEAR TECHNIQUES
The spectroscopist wishing to excite with laser light finds several
gaps in the table of available wavelengths. Although dye lasers
cover an extended portion of the spectrum, they require too much
maintenance to be useful in standard confocal microscopy. Optically anisotropic crystals provide a good alternative towards
tunable light as long as they are pumped with lasers having high
(pulsed) peak power at a high repetition frequency.
Second and Higher Harmonic Generation:
SHG, THG, FHG Label-Free Microscopy
Non-linear optical (NLO) effects occur in certain classes of optically anisotropic crystals (Lin and Chen, 1987; Tebo, 1988).
Focusing intense laser light into these crystals generates radiation
at half the wavelength or double the frequency. This process, therefore, is known as “frequency doubling” or “second harmonic generation” (e.g., Huth and Kuizenga, 1987; Higgins, 1992). The
intensity of the frequency-doubled light output is proportional to
the square of the incoming light intensity. A Cr:forsterite laser
running at 1230 nm was used to create a very elegant second harmonic generation (SHG) and third harmonic generation (THG)
system operating at 615 and 410 nm (respectively) with a pulse
repetition rate of 110 MHz and a pulse width of 140 ps (Hogan,
2004; see also Chapter 40, this volume). When the incoming beam
is very intense, even third, fourth, or fifth harmonics can be generated. However, the efficiency of harmonic generation decreases
at higher orders. Most doubling crystals can be angle or temperature tuned. THG is very interesting for label-free monitoring of
inhomogeneities at interfaces, for example, cell membranes
(Moreaux et al., 2000). Yelin (2002) and Brocas and colleagues
(2004) compare the THG performance of several systems: the
Spectra Physics Tsunami-OPAL synchronously pumped OPO,
which emits a 80 MHz 150 fs pulsetrain at around 1500 nm with an
average power of 350 mW; and the T-pulse laser (Amplitude
Systèmes) emitting 50 MHz 200 fs pulses at 1030 nm with 1.1 W
of average power. THG reaches 500 and 343 nm, respectively, so
UV transmitting optics may be required. A not-so-pleasant consequence of the doubling process is the doubling of the noise level.
Pump lasers with excellent stability are a must. Higher peak
powers and new, low-threshold materials require a careful reconsideration of design strategies (Beausoleil, 1992). CSK Optronics
(Table 5.3) sells a line of compact doublers and triplers for Ti:Sa
lasers. An important issue in this type of experiment is the amount
of optical damage done to living cells and tissue by these power-
ful femtosecond pulses (Chu et al., 2003; and also see Chapter 38,
this volume). THG peak intensities of 100 to 300 GW/cm2 illumination (i.e., nJ/pulse) are not damaging.
Presently, it is also possible to combine less powerful tunable
dye or semiconductor lasers to cover the wavelength range from
170 nm to 18 mm almost continuously.
Sum or Difference Mixing
Another technique for generating different wavelengths from
certain basic laser wavelengths is sum or difference mixing. When
two laser beams of high and low intensity and of frequency w1 and
w2, respectively, are simultaneously focused into a non-linear
optical crystal, a sum signal is generated. The intensity of the sum
signal, I(w1+w2) is proportional to the product I(w1) ¥ I(w2). The higherintensity w1 laser helps in generating enhanced UV output I (w1 +
w2) with an extended tuning range. In a similar way, difference
mixing, I(w1-w2) leads to a tunable IR laser. (For examples, see
Dunning, 1978; Adhav, 1986; Herrmann and Wilhelmi, 1987;
Kaiser, 1988; Marshall, 1994; Demtröder, 1996)
Optical Parametric Oscillators (OPO)
and Amplifiers (OPA)
The highest repetition rate that can be obtained depends on pump
energy storage. Potma and colleagues (1998) developed a cavitydumped OPO with 13 nJ pulses that reached a repetition rate of
4 MHz. These devices can generate a continuously-tunable output
by non-linear conversion of fixed wavelength highintensity CW or
pulsed laser light (Butcher, 1994; Radunsky, 2002). OPOs are a
reliable and easy-to-use means to cover spectral gaps left by dye
or Ti:Sa lasers. Between the Ti:Sa fundamental, covering 700 to
1000 nm, and the second harmonic, covering 350 to 500 nm, a spectral gap also exists from 250 to 350 nm. This means that neither
tryptophan, tyrosine, coumarine, or naphthalene nor the caged
probes can be covered by a Ti:Sa. OPO efficiencies easily reach
40% to 60%, (Anderson, 1993) and practical systems are available
from many companies (Table 5.2). A continuously tunable output
stretching from UV to IR is generated by non-linear conversion,
that is, optical frequency mixing. Parametric frequency conversion
creates two lower-energy photons (signal and idler) from each
incoming photon. This process is the opposite of sum mixing.
There are two methods to overcome the low efficiency of the
parametric generation: Method 1 — Parametric amplification in a
non-linear BBO crystal pumped by the second harmonic of the
Ti:Sa followed by pulse compression. Parametric down-conversion is incompatible with long pixel integration times, and high
pulse energies may possibly be damaging, and instrumentationally,
it is quite involved. Method 2 — Cavity-dumping a synchronously
pumped OPO can produce pulse repetition rates up to 100 MHz.
This allows FLIM imaging and reduces somewhat the average
power delivered and cell damage. In addition, residual pump
beams at 400 and 800 nm are available. Synchronous pumping
means that the cavity lengths and, therefore, the pulse round-trip
times are equal in both pump source and OPO (Potma et al., 1998).
Non-critical phase-matching (NCPM) in a crystal is preferred
because it allows tight focusing of the pump beam and long interaction lengths, resulting in low OPO thresholds. Efficiency can be
further increased by using periodically poled (PP) waveguides,
which do not diffract a tightly focused beam as does bulk material,
while at the same time they allow long interaction lengths with a
good mode profile. For example, green output from a Q-switched
diode-pumped, frequency-doubled Nd-YAG laser pumps a non-
PHC5 28/11/2005 6:15 PM Page 115
Laser Sources for Confocal Microscopy • Chapter 5
linear crystal creating a broad tuning range and narrow bandwidth
with 15 mW average power at around 670 nm. Typical oscillator and
amplifier crystal combinations cover various spectral regions.
Angle tuning a proper crystal will give a 405 to 2700 nm
tuning range with 45% total (signal + idler) efficiency (QWULasertechnik mbH; Table 5.3). Continuum (Table 5.3) introduced
the Sunlite OPO (Anderson, 1994a), which is currently the Sunlite
EX OPO, with a very broad tuning range stretching from 222.5 to
1750 nm without a gap and providing line widths of 0.075 cm-1. It
is pumped with 100 to 160 mJ pulses at 355 nm from a Q-switched
Nd:YAG. Such a system is a good replacement for pulsed dye lasers.
As mentioned above, there is a trend to create easy-to-operate
automated all-in-one-box solutions. Spectra-Physics provides the
LBO-based femtosecond synchronously pumped OPAL. Near the
peak of the OPO tuning curve, it generates 200 to 250 mW and
covers a wavelength range from 1100 to 2260 nm with pulse widths
<130 fs. By frequency doubling, one covers the 550 nm gap left
by the Ti:Sa systems. A further improvent with higher average
power and repetition rates is obtained with ytterbium tungstate
(Yb:KGW) employed in the Eclipse fs OPA (Spectra-Physics;
Krueger and Féru, 2004).
Several OPO classes exist depending on the bandwidth narrowing optics used. The narrower the bandwidth, the less the gain.
In addition, the number of passes through the crystal influences the
gain. A single-pass device creates little gain for a 5 ns, 355 nm,
100 mJ pulse. Enclosure of the non-linear material in a resonant
cavity gives optical feedback and adding a pulsed pump source
creates extra gain. The complete design forms an optical parametric amplifier (OPA). The frequency stability of an OPO depends
on the pump source stability and, with enough stability, singleresonant OPO (SRO), with feedback for idler or signal, or a doubleresonant OPO (DRO) with feedback for signal and idler, can be built
(Radunsky, 2002). OPOs as well as tunable CW dye lasers are often
used as narrow bandwidth seed sources for OPAs. For example, by
SHG doubling the 820 nm output from a Ti:Sa regenerative amplifier in a BBO crystal, one obtains 410 nm. This 410 nm pumps a
second BBO crystal, the OPA. An 80-fold amplification is obtained
resulting in 100 mJ pulses with narrow bandwidth. Changing the
crystal angle (angle tuning) covers the wavelength range from
450 to 700 nm. These techniques can be adapted and implemented
into fiber-optic OPO and OPA designs as well.
Pulse Length Measurement
Continuous on-line monitoring of the picosecond or femtosecond
pulse shape immediately provides the feedback needed to align the
cavity and perform other adjustments. Thermal drift may require
cavity length retuning twice a day, even in a very large air-conditioned dedicated laboratory space. In a smaller space this may be
required far more often. A pick-off signal from the monitoring
diode is fed into an auto-correlator unit attached to a 400 MHz
digital oscilloscope. Several designs exist but a computer-assisted
background-free design is favored. A website explaining the operation of a background-free correlator is http://linac.ikp.physik.
tu-darmstadt.de/fel/optical_pulse.html. Both time-dependent
amplitude and phase must be known for many experimental procedures. Frequency-resolved optical gating (FROG) creates a twodimensional function of the spectrally resolved autocorrelation at
a range of correlation delays. Second harmonic-FROG in a
non-collinear configuration is background free. Avesta (Del Mar
Ventures; Table 5.3) offers a semiconductor-based design eliminating the non-linear crystal and the PMT detector. Created from
a single pulse, the overlap of two identical, time-delayed pulses
115
with different center frequencies generates a spectral interference
pattern in spectral phase interferometry for direct electric-field
reconstruction (SPIDER). Pulse spectrum and cross-correlations of
a pulse with a pulse copy that has been altered by a dispersive plate
or attenuator in one arm of the correlator are used for phase
and intensity determination from correlation and spectrum only
(PICASO).
MAINTENANCE
On delivery, most laser systems perform well above the guaranteed specifications given by the manufacturer. The consumable
parts (i.e., laser tubes, lamps, dyes, and amplifier optics in solidstate lasers and attached fiber-optics) are usually not covered by
the warranty. Depending on the type of laser used, these costs can
be substantial.
A factor usually forgotten at the time of an initial laser purchase is the need to acquire peripheral equipment for proper maintenance. Extra dyes, filters, power meters, sampling oscilloscopes
and sampling units, fast diodes, infrared viewers, spectrum analyzers, beam-dumps, radiation shields, warning signs, covers, laser
goggles, explosion- and fire-proof cabinets for solvent storage, and
hazardous waste disposal, etc., are often not included.
Maintenance of Active Laser Media
Laser Tubes
The plasma tubes in ion lasers must be replaced when the cathode
sags or the bore of the tungsten disks or BeO tube corrodes to
become too big. The “getter,” always installed in the laser system
to remove impurities in the gas, may become saturated. Cracks that
form in cooling tubes usually lead to its immediate failure. The
introduction of beryllium-oxide (BeO) tubes with superb heat conduction characteristics has significantly reduced the fragility of
these laser systems and increased their reliability. Medium-power
systems should last for 3000 to 5000 h (see Tables 5.1 and 5.2) and
high-power systems for 2000 h.
Most laser systems come with 18-month warranties, but some
self-contained, sealed systems carry a warranty for up to 5 years.
The technology has significantly improved over the last few years
to reduce the deposition of disk material and other contaminants
on the inner surface of the Brewster windows. For the largest ion
lasers, the aluminum oxide ceramic tube technology with brazed
tungsten/copper disks has revolutionized deep UV output power
levels and reliability. For He-Cd lasers, regular tube replacements
are necessary after 1000 h of operation.
Tube Replacement and the Secondhand
Laser Market
Several companies repair and resell worn laser tubes. Always ask
for warranties and try-out. Old BeO tubes are considered hazardous material (HazMat) and can for a nominal amount be
returned to the manufacturer.
What the User Can Do
To prolong the life of a tube, check mirror alignment so that the
stabilized output can be attained with minimum excitation power
to minimize tube component degradation. Check the water flow
around the laser head for bad connections and look for internal
deposits at least once a year. Keep dust out of laser systems; the
laser should be connected to a permanent supply of clean, dry,
room-temperature, oil-free air or nitrogen gas. A slight over-
PHC5 28/11/2005 6:15 PM Page 116
116
Chapter 5 • E. Gratton and M.J. vandeVen
pressure helps to prevent dust from entering the laser head.
Commercial air filters that night introduce dry powders into the
laser system should never be used. Synchronously-pumped
systems perform markedly better and the intensity stability
improves when dust is absent. Do not run at the highest power
unless necessary, keep a user log, and inform users on the cost per
hour to prevent idling.
Dyes
Dye lasers pose a number of hidden costs. Rhodamine 6G is one
of the least expensive dyes, and is the most stable with a lifetime
of 2000 Wh for a typical 1-L reservoir. (i.e., the output power of
the dye is reduced to 50% when it is pumped for 1000 h at 2 W or
400 h when pumped at 5 W). Dye aggregates can be filtered out but
evaporation of the solvents necessitates periodic refilling. When
the dye pump inadvertently starts to suck in air, air bubbles are
generated and cause immediate contamination of the optical mirror
surfaces. This problem can be prevented by careful start-up. Any
faint, sharp clicks at the nozzle position indicate air bubbles
passing the nozzle opening. Operating the dye laser with the wrong
solvent, for example, incorrect viscosity, may cause the dye to
foam, which will quickly clog the system. In order to reduce downtime and contamination of dyes, it is recommended that every dye
is run with its own pump module. This is more expensive, but
results in a much faster hook-up. Used dye and leftover solvent
should be discarded in a manner that does not pose a threat to
humans or the environment. Dye nozzles should be cleaned regularly in a sonicator, especially when the system has not been used
for a while. Non-laminar flow from the jet nozzle is an indication
of nozzle problems.
Gases
Excimer lasers need regular gas changes. Ion lasers refill their gas
from a ballast tank via a microcomputer-controlled valve system.
This reduces the risk of overfilling the plasma tube and degrading
the performance or losing laser power altogether.
Laser and Amplifier Rods
Solid-state media, at least for CW lasers, are quite robust.
However, pulsed systems need a very good Gaussian beam quality,
otherwise the rod coating will be damaged. Often one can hear this
as sharp clicking sounds that indicate that the laser needs immediate attention. Improper Ti:Sa rod cooling leads to water condensation on the pump surface and produces instabilities. In the worst
case, continued operation may damage the bulk of the laser rod
(Soileau, 1987). Degradation of the surface layers of the medium,
caused by the intense illumination from the pumping system and
is referred to as solarization, is less of a problem.
Diode Lasers
These devices are quite sensitive and can easily be damaged or
destroyed. Damaged systems may show very weak output, a shift
in emission wavelength, or a change in divergence or beam shape.
Pig-tail fiber-optics, normally attached in front of the laser diode,
also are suspect when the output power drops significantly but the
drive electronics should be checked as well (as we recently experienced!). The fiber-optics may have been pulled out of its socket
or it may have broken. At high power levels damage to the fiber
surface may also occur, especially with plastic fiber-optics. As discussed above, current transients (spikes) and electric discharges
should be prevented at all times. Forward-biased semiconductor
junctions are inherently thermally unstable; an increase in current
increases the temperature, which, in turn, increases the current
even more. Therefore, current supply design is important and must
be suited to the specific laser.
Maintenance of Pumping Media
For CW or pulsed lasers, which are pumped by arc lamps, the
lamp(s) must be replaced when the output power of the laser
decreases. On the inside of the arc lamp, dark deposits may
develop, which, over time, will cause an increase in absorption and
catastrophic failure due to local heating. Defective lamps usually
fail within the first 10 to 20 h after start-up (Smith, 1986;
Littlechild and Mossler, 1988). Optical amplifier components can
be shielded from damaging UV radiation by doping the quartz
lamp envelope with cerium. Electrode sputtering can be reduced
by using sintered tungsten doped material with a low-work function to create a more uniform cathode current and lower temperatures (Erlandson and Powell, 1991). In addition, longer flashlamp
lifetime can be obtained by using mushroom-shaped cathodes.
Maintenance of the Optical Resonator
Dust is one of the major enemies of the laser and tobacco smoke
should be kept out of any laser laboratory. Dust covers should be
in place at all times and any dust on optical surfaces should be
removed as soon as possible, for example, on wavelength tuning
prisms, birefringent plates, mode locker or Pockels cell surfaces,
or mirrors. As using the wrong solvent to clean optics can destroy
the optical coating, use only the ultra-pure solvents recommended
by the manufacturer and follow the proper procedure. When in
doubt, first try it out on a test surface. Most laser manuals describe
in detail the proper cleaning methods. Never use the same side of
a cleaning wipe for more than one pass. A patterned movement
should be used when cleaning optical surfaces.
The optical coatings on laser cavity mirrors and external
mirrors or fibers cannot withstand prolonged exposure to UV radiation. Colored rings will appear on the coating surfaces or they
may look hazy or foggy. The formation of color centers, which
cause increased absorption, occurs in fused quartz materials:
lenses, optical fibers, etc. Color centers are usually caused by
included impurities. Resulting lattice defects are mostly single
electrons bound to negative-ion vacancy sites: F-centers. Free electrons created by intense radiation become trapped in these sites.
Excimer lasers with their high UV peak powers are particularly
prone to creating color centers. At high power, too tight focusing,
or self-focusing, conditions can produce mechanical damage
(remember, lasers are used for machining!; Günther, 1993).
Damaged parts must be replaced immediately to prevent the deteriorated beam from damaging other components. This is especially
true for pulsed systems. Frequency-doubling elements should be
inspected regularly or whenever laser intensity fluctuations occur.
A microscope may be necessary to see damage to the coating.
Heating of some crystals (e.g., KTP) seems to reduce these effects.
Frequency-doubling and other hydroscopic crystals should be kept
under optimum dry conditions at all times.
Maintenance of Other System Components
Cooling Water
A gradual decrease in laser power can be caused by the growth of
algae in the cooling water. A small addition of sodium azide (NaN3)
prevents a reoccurrence of the growth. An annual check of the
rubber rings in the cooling hoses is recommended. At least once a
year the cooling circuitry should be checked for galvanic and other
PHC5 28/11/2005 6:15 PM Page 117
Laser Sources for Confocal Microscopy • Chapter 5
types of corrosion (Schneider and Williams, 1993). Do not use tap
water for cooling. Many laser and temperature bath manufacturers
now offer closed-loop laser cooling systems. Heat removal with
air-to-liquid cooling is also possible (Goldman, 1993). For proper
operation of a Nd-YAG laser, the resistivity of the cooling water
should be in the range recommended by the manufacturer. When
the resistivity is too low, the lamps will not start. If it is too high,
the plating on the inside of the elliptical resonator will dissolve
and cause a decrease in laser power. Water filters and de-ionizing
filters should be replaced regularly as demanded by the performance of the laser. Stacked diode laser arrays, small in size but
with several tens of watts of optical output power, require liquid
cooling with micro-channel technology or thermo-electric (Peltier)
cooling.
External Optics
All optical surfaces should be kept as clean as possible. Mirror surfaces exposed to UV radiation will have to be replaced regularly,
depending on the impinging power density. Coatings may peel off.
Apertures in spatial filters should be inspected and replaced when
damage (burn) occurs. Fiber-optics, especially the ones exposed to
high-intensity UV light, may develop color centers, which result
in increased absorption and, in the end, failure of the entrance
section. High-intensity visible or infrared light may lead to overheating at the input fiber tip. Fortunately, the length of most fibers
is sufficient to allow them to be cut back, repolished, and reconnectorized, though sometimes at substantial cost and time delays.
If they are bent too tightly, fibers will break, causing a sudden loss
of output power. Pig-tail fibers attached to the main body of a laser
diode may dislodge themselves. Due to the automated fabrication
process, which includes angular positioning to optimize the output
and use the most intense emission spot, manual reattachment
usually does not re-establish the original power level. Laser
damage thresholds for dielectric mirrors and anti-reflective coatings are roughly 250 MW/cm2 at 532 nm and 500 MW/cm2 at 1064
nm, assuming surfaces are spotless and clean (Aubourg, 2002).
TROUBLESHOOTING
A very extensive and instructive body of information for gas, dye,
and diode laser maintenance, including optics cleaning and repair,
exists in Sam’s Laser FAQ (see website listings).
SAFETY PRECAUTIONS
Laser hazards include thermal and fire, acoustic shockwave, and
strongly wavelength-dependent photochemical damage. Types of
beam exposure are direct exposure intrabeam, specular, and diffusion reflection effects. Eyes and skin are most commonly affected.
Non-beam–related hazards include electrical shock, capacitors,
hose leakage, water vapor condensation, air contaminants, fumes,
aerosols with biological agents, cadmium and zinc telluride (that
burns in the presence of high laser intensity and oxygen), radiation damage, fire, compressed gases, gas cylinders, earthquake
damage, excimer gases, and laser dyes.
All lasers are generally divided into four classes:
Class 1. Embedded lasers and laser systems
Laser completely enclosed, radiation not accessible during use.
Class 1M. Lasers and laser systems:
CW = <40 mW blue and = <400mW red.
Very low power: Safe for long-term intrabeam viewing.
117
Class 2. Low power visible lasers and laser systems:
CW = <1mW
Low power level. Safe for brief accidental naked eye direct exposure with blink and aversion response active.
Class 2M. Low-power visible lasers and laser systems.
Low-power visible collimated or divergent large beam diameters. Potential hazard with magnifiers
Class 3R. Visible. Low-power lasers and laser systems.
Accidental exposure usually not hazardous but eye injury possible
upon intentional long-term viewing. Training required. Equivalent
to Class IIIA “danger” [Center for Devices and Radiological
Health (CDRH)] and ANSI 3a (USA).
Class 3R: Invisible. Low-power lasers and laser systems.
Wavelength dependent, limits are five times those of Class 1.
Class 3B. Medium-power lasers and laser systems:
CW = <500mW.
Serious eye injury even for brief accidental exposure to direct
beams possible. Training required.
Class 4. High-power lasers and laser systems.
Even diffuse and certainly direct-eye exposure will lead to serious
eye and skin injury. Poses fire hazard as well. Training required.
with M: magnifying instruments and R: relaxed requirements.
Limiting values for small point-like lasers with angular retinal spot
size smaller than 1.5 mrad.
General safety precautions, that become more stringent with
increasing classification, must be followed when operating a laser
[International Commission of Non-Ionizing Radiation Protection
(ICNIRP) and International Electrotechnical Commission (IEC)
60825-1 guidelines, see revised ANSI-Z-136.3 1996 classification,
safe use of lasers; Tozer, 2001; Schulmeister, 2003; Sliney, 1994;
LIA and Rockwell Laser Industries Inc.; Table 5.3]. Separate photochemical and thermal retinal exposure limits, including limits for
damage from ultrashort, <1 ns pulses, have been added as well as
the *.M classes. Every laboratory should have a Laser Safety
Officer (LSO) who should be consulted when necessary. This
person should be responsible for the proper training of users of
laser-assisted equipment. Safety training videos are available from
LIA, OSHA, and Rockland Laser Industries (Table 5.3) and often
also from your own local university. Outside the laser laboratory,
a “laser in use” warning sign must be posted and a red warning light
should be positioned at the entrance to the laser room.
Inside the laser laboratory, safety precautions are necessary
even for users of enclosed systems attached to CLSMs whenever
the system is opened or when the fiber-optics are aligned and exposure to a laser beam becomes a possibility.
A brief “Do and Don’t” list:
• Inform everybody near the laser set-up to be worked on. Check
the sign-up agenda.
• Check and put all required accessory equipment on the ready.
• Have the entrance warning red light on when any laser is on.
• Remove everybody from the laser site whose presence is not
required.
• Close the access doors.
• Close curtains.
• Remove rings, watches, and ties.
• Put safety goggles on.
• Never look directly into any laser beam.
PHC5 28/11/2005 6:15 PM Page 118
118
Chapter 5 • E. Gratton and M.J. vandeVen
• Align open laser heads with minimum power.
• Check for stray light and reflections with a hole-punched business or IR indicator card.
• Block beams not being attended.
• Disconnect electrical connections.
• Discharge large capacitors with a grounded large screwdriver
with an insulated handle.
1988), cleanliness, and proper handling of chemicals. A sufficient
number of eye protection devices should always be at hand when
laser system components have to be repositioned or realigned.
Naturally, one should reduce the power level as much as possible
during alignment. Periodic training meetings and medical checkups are recommended and can be used as a certificate of good laser
usage conduct.
Beam Stops
Infrared Paper
Beam stops should be made out of anodized, flat black aluminum
and positioned in such a way that no radiation is reflected back
into the room. Stray reflections must be prevented; powerful IR or
visible beams may easily start a fire on electrical cables or other
flammable materials.
To safely find infrared beams, one can rely on infrared viewers but
these are fairly expensive. An alternative is the use of IR indicator cards. These emit a visible luminescence via electron-trapped
up-conversion. Suppliers include Bromba GmbH, Edmund
Scientifics, Electro Optical Components, and Lumitek Intl. Inc.
(Table 5.3).
Curtains
Curtains should be made of special non-combustible material,
preferably black in color. When a spot of 1 mm2 is irradiated with
approximately 1 MW/m2 or more, the black side should emit a nontoxic smoke (that nevertheless may form damaging deposits on
critical optical surfaces!) and may glow. See, for example, the
Rockwell Laser Industries, Inc. and Lasermet Ltd. websites. Curtains are designed for low-power, 100 W/cm2 or up to 1.2 kW/cm2
lasers (MacMullin, 2004). Curtain shielding will prevent the
accidental illumination of other people while aligning a laser or
beam-splitter.
If a visual control is necessary from outside the room, a
radiation-absorbing plastic window should be mounted in the door
or wall. Tape should not be used to hold curtains together because
this could allow the build-up of poisonous gases. Clothing is
another possible source of trouble that is often disregarded.
Famous are the stories of halved neckties.
Laser Goggles
A large range of eye protection devices is available (e.g., from
LFW and Uvex; Table 5.3). A telling Ti:Sa fovea burn track is
shown in Robinson (1998). It is important to realize that a 1 W/cm2
irradiance level at the cornea can become a focused 100 kW/cm2
retina spot of about 20 mm diameter.
CONCLUSION
In the previous sections, we have described the large number of
lasers currently available for CLSM. These lasers vary widely in
characteristic wavelengths, tunability, stability, output power, ease
of use, and price. Rapidly expanding areas such as diode lasers,
optical parametric amplification, and direct-doubling diode lasers
would require an update every three months. Solid-state laser
systems with wide tunability through the incorporation of optical
parametric techniques and fiber lasers are areas showing rapid
development.
Brelje and colleagues (1993) show a figure displaying the excitation wavelengths of various popular fluorophores with matching
ion laser lines. This picture has certainly changed in the past years
with the increase in power of solid-state and fiber lasers and the
expansion of their range of tunability. For the moment we summarize several of the most often used lasers in Table 5.4. It is clear
from this table that the ion lasers are being pushed out of the
door.
• For CLSM, a stable, TEM mode, easy-to-use laser with short
coherence length is imperative.
• Wavelength tunable lasers require pulsed pump lasers with
00
Screens
Where possible, anodized, flat black aluminum pipes should be
used to enclose all laser beams. These not only protect the operator, they also reduce dust in the beam line.
Exposure Effects, Warning Signs, and Interlocks
Each laser system must be equipped with the proper warning signs
and interlocks. The listed references (Sliney and Wolbarsht, 1980;
Rockwell, 1983, 1986; Winburn, 1985; Sliney, 1986) describe in
detail electrical hazards and detrimental reflections from rings,
watches, etc. Biological effects are also extensively covered, that
is, thermal and photochemical effects upon exposure to CW and
pulsed laser radiation, and the maximum permissible exposure to
eyes and skin.
Chemical hazards may be caused by laser dyes, some of which
are mutagenic. Dye solvents and laser gases must be properly
handled. Basic protective measures include laser goggles (Burgin,
•
•
•
high peak power, good mode quality, and the best possible stability to reduce noise in the subsequent wavelength conversion
and amplifier stages, as well as OPO, OPA, or master oscillator parametric amplifier (MOPA) optics. The best source is currently the tunable Ti:Sa laser. Excimer, nitrogen, and similar
pulsed lasers and mode-locked solid-state lasers are still useful
pump sources for tunable dye lasers if pulse repetition rates
are sufficient for your purpose. The goal should be to each have
a US$6000 modular, personal 2-photon source!
Miniaturization and ease-of-operation via efficient, low-cost,
low-noise, tunable semiconductor and fiber lasers is a reality
now and these lasers are replacing big CW and pulsed laser
systems. A trend exists towards modularity, efficiency, and
small footprint.
With the advent of completely automated, self-contained laser
systems (Mai-Tai and Chameleon) the ideal turnkey system is
almost present. Older, bigger, flexible laser systems cannot be
considered turnkey systems and require constant attention
from well-trained, qualified persons.
Laser safety training is a must also for CLSM users and eye
protection and other safety measures should be at hand as
required.
PHC5 28/11/2005 6:15 PM Page 119
Laser Sources for Confocal Microscopy • Chapter 5
119
TABLE 5.4. LASERS OFTEN USED IN LSCM
Manufacturer and CLSM system
Laser Type
Visitec
international
Vt-eye and
Vt-Infinity
Leica
TCP-SP2-AOBS
SP2
Nikon
Digital Eclipse
C1 Plus
Olympus
Fluoview
FV1000 & FV500
Perkin Elmer
Ultraview
WITEC
CRM200
Zeiss
Axiovert 200M
LSM 510 META
Fixed wavelength lasers
HeCd
UV 325 nm
Argon-ion
UV 351 nm
Argon-ion
UV 364 nm
Argon-ion
UV 351, 364 nm
Diode laser
405 nm
10–30 mW
40 mW
>20 mW
50 mW
6 mW
Diode laser
408 nm
Diode laser
DPSS 430 nm
Diode laser
440 nm
HeCd
442 nm
Argon-ion
Multi-line
458, 514 nm
Blue only
488 nm
Laser diode
488 nm
DPSS
532 nm
Green HeNe
543 nm
Yellow DPSS
561 nm
Krypton ion
568 nm
Krypton/Argon
488, 568, 647 nm
Orange HeNe
594 nm
Red HeNe
632.8 nm
Diode laser
630 nm (ext. cavity)
635 nm
638 nm
NIR 750 nm
HeNe
NIR 1152 nm
300 mW
80 mW Enterprise
20 mW
6 mW
(25 mW, opt)
7.5 mW
25 mW
50 mW (510 Live)
0.7 mW
(5 mWopt.)
7 mW
16 mW (510 Live)
6 mW
20 mW
11 mW
488, 514, 568,
7.5/15 mW
each
25 mW @
457–514.1 nm
20 mW (mod.)
8 mW
5 mW (mod.)
20 mW
458, 514 nm,
25 mW
100 mW @ 457,
476, 488,
514 nm 65 mW
40 mW
30 mW
40 mW @ 457,
488, 515 nm
+
10 mW
10 mW
30 mW
458, 477, 488,
514 nm
+
100 mW (510 Live)
75 mW (510 Live)
+
1.5 mW
1.5 mW
1 mW
1.25 mW
1 mW
10 mW (510-META)
25 mW
20 mW
10 mW
+
75 mW
2 mW
2 mW
10 mW
5 mW
2 mW (510 -META)
10 mW
3.5 mW
33 mW
5 mW
35 mW(510 Live)
10 mW (mod.)
640 7.8 mW
2 mW CW
1 mW
PHC5 28/11/2005 6:15 PM Page 120
120
Chapter 5 • E. Gratton and M.J. vandeVen
TABLE 5.4. (Continued)
Manufacturer and CLSM system
Laser Type
Visitec
international
Vt-eye and
Vt-Infinity
Leica
TCP-SP2-AOBS
SP2
Nikon
Digital Eclipse
C1 Plus
Olympus
Fluoview
FV1000 & FV500
Perkin Elmer
WITEC
Ultraview
CRM200
Zeiss
Axiovert 200M
LSM 510 META
Broad-band tunable pulsed lasers
Dye laser
630 nm
Ti:Sa-Chameleon
720–930 nm
Ti:Sa-Mai-Tai
750–850 or 710–920 nm
Ti:Sa-Mira
690–1000 nm
Ti:Sa-Tsunami
680–1050 nm
NIR laser
700–900 nm
3 mW
1000 mW
>1 ps
<1000 mW
>1 ps
100–700 mW
>1 ps
100–700 mW
>1 ps
1000 mW
<100 fs
<1000 mW
<100 fs
100–700 mW
<100 fs
100–700 mW
<100–130 fs
1.2 ps, 1000 mW
720–1000 nm
Mod.: modulated
Pp peak-to-peak
Visitech International, Inc, Sunderland, UK +44 (0)191 516 6255 http://www.visitech.co.uk/
Leica Microsystems, Mannheim, Germany, Ph. +49-(0)621-70280 http://www.leica-microsystems.com/website/lms.nsf
Nikon Inc., Melville, NY. Ph. (800)-526-4566, http://www.nikonusa.com/, http://www.nikon-instruments.jp/eng/
Olympus America, Melville, NY. Ph (631)-844-5000, http://microscope.olympus.com/, http://www.olympusamerica.com/seg_section/seg_home.asp and
Perkin-Elmer, Boston, MA, Ph. (800)-762-4000, http://las.perkinelmer.com/content/livecellimaging/index.asp
Carl Zeiss AG, Oberkochen, Germany, Ph. +49-(0)7364-200 http://www.zeiss.com/
WITec GmbH, Ulm, Germany, Ph. +49-(0)700-948-32366, http://www.WITec.de Confocal Raman microscope
ACKNOWLEDGMENT
This work was supported by National Institutes of Health grant
PHS-1P41-RR03155.
REFERENCES
Adhav, R.S., 1986, Data sheet 714. Sum frequency mixing and second harmonic generation. Quantum Technology, Inc., Lake Mary, FL January:
(407-323-7750).
Aghassi, D., Gonzalez, E., Anderson, R.R., Rajadhyaksha, M., and Gonzalez,
S., 2000, Elucidating the pulsed dye laser treatment of sebaceous hyperplasia in vivo with real-time confocal scanning laser microscopy, J. Am.
Acad. Dermatol. 43:49–53.
Anderson, S.G., 1993, Commercial OPO produces high-energy tunable output,
Laser Focus World April:16–22.
Anderson, S.G., 1994a, Narrow-linewidth OPO uses extraordinary resonance,
Laser Focus World April:15–18.
Anderson, S.G., 1994b, New material promises tunable UV output, Laser
Focus World May:20–23.
ANSI Z-136.3-1993, ANSI Standard for the safe use of lasers. Laser Institute
of America (LIA), Orlando, Florida.
Arecchi, F.T., and Schultz-Dubois, E.O., 1972, Laser Handbook, Vol. 1, NorthHolland Publishing Co., Amsterdam.
Arrigoni, M., 2004a, Femtosecond laser pulses in biology: From microscopy
to ablation and micromanipulation, Biophotonics Intl. June:48–51.
Arrigoni, M., 2004b, The power of the pump. SPIE’s oemagazine
October:29–30.
Austin, L., Scaggs, M., Sowada, U., and Kahlert, H-J., 1989, A UV beamdelivery system designed for excimers, Photonics Spectra May:89–96.
Art, J.J., and Goodman, M.B., 1993, Chapter 2 Rapid Scanning Confocal
Microscopy, In: Cell Biological Applications of Confocal Microscopy.
Methods in Cell Biology, (B. Matsumoto, ed.), Acad. Press, San Diego,
pp. 53–58.
Ashkin, A., and Dziedzic, J.M., 1987, Optical trapping and manipulation of
viruses and bacteria, Science 235:1517.
Aubourg, Ph., 2002, Laser damage: What is it?, Europhotonics
June/July:46–47.
Baer, T.M., 1986, Diode Laser Pumping of Solid State Lasers, Laser
Focus/Electro Optics June:82–92.
Bains, S., 1993, Holographic optics for when less is more, Laser Focus World
April:151–154.
Bass, M., and Stitch, M.L., 1985, Laser Handbook, Vol. 5, North-Holland Publishing Co., Amsterdam.
Beach, R.J., Krupke, W.F., Kanz, V.K., Payne, S.A., Dubinskii, M.A., and
Merkle, L.D., 2004, End-pumped continuous-wave alkali vapor lasers:
experiment, model, and power scaling, JOSA B, 21:2151–2163.
Beausoleil, R.G., 1992, Highly efficient second harmonic generation, Lasers &
Optronics May:17–21.
Bechhoefer, J., and Wilson, S., 2002, Faster, cheaper, safer optical tweezers for
the undergraduate laboratory, Am. J. Phys. 70:393–400.
Bertolotti, M., 1983. Masers and Lasers. An historical Approach, Adam Hilger
Ltd., Bristol.
Bloom. A.L., 1968. Gas lasers, John Wiley and Sons, New York.
Boas, G., 2003, Model pinpoints heating effects in optical tweezer experiments:
Accurate measurements may imprive calibration, Biophotonics Intl.
April:59–60.
Brelje, T.C., Wessendorf, M.W., and Sorenson, R.L., 1993, Chapter 4, Multicolor Laser Scanning Confocal Immunofluorescence Microscopy: Practical Application and Limitations. In: Cell Biological Applications of
Confocal Microscopy. Methods in Cell Biology, (B. Matsumoto, ed.), Academic Press, San Diego, pp. 120–123.
Brocas, A., Canioni, L., and Sarger, L., 2004, Efficient selection of focusing
optics in non linear microscopy design through THG analysis, Opt.
Express, 12:2317–2326.
Brown, D.C., 1981. High-Peak-Power Nd-Glass Laser Systems, In:
Springer Series in Optical Sciences, Springer-Verlag Berlin, Vol. 25 pp.
1–276.
Burgess, D.S., 2004, Laser microdissectionL Making inroads in research, Biophotonics Intl., September:46–49.
PHC5 28/11/2005 6:15 PM Page 121
Laser Sources for Confocal Microscopy • Chapter 5
Butcher, S. 1994, Optical parametric oscillators open new doors to researchers,
Photonics Spectra, May:133–138.
Burgin, C.D., 1988. A guide for eyeware for protection from laser light, LLLTB-87, LLNL, P.O. Box 808, Livermore, California.
Carts, Y.A., 1994, Gradient-index lens tames spherical aberrations, Laser Focus
World January:142–143.
Casperson, L.W., 1994, How phase plates transform and control laser beams,
Laser Focus World May:223–228.
Chenard, F., 1994, New applications abound for rare-earth doped fibers, Photonics Spectra May:124–130.
Chu, S.-W., Liu, T.-M, Sun, C.-K., Lin, C.-Y., and Tsai, H.-J., 2003, Real-time
second-harmonic generation microscopy based on a 2 –GHz repetition rate
Ti:sapphire laser, OpticsExpress 11:932–938.
Cogwell, C.J., Hamilton, D.K., and Sheppard, C.J.R., 1992a, True color confocal reflection microscope: 442 He-Cd, 532 of freq. doubled Nd-YAG
with 633 nm from He-Ne laser, J. Microscopy 165:49–60.
Cogwell, C.J., Hamilton, D.K., and Sheppard, C.J.R., 1992b, Colour confocal
reflection microscopy using red, green and blue lasers, J. Microscopy
165:103–117.
Connor Davenport, C.M., and Gmitro, A.F., 1992, Angioscopic fluorescence
imaging system, in Optical Fibers in Medicine VII, Ed. A. Katzir, SPIE
Proceedings, Los Angeles, California, SPIE 1649:192–202.
Cunningham, R., 1993, Vertical-cavity diode lasers, Lasers & Optronics
December:19–20.
Dantus, M., 2003, Pulse shaping helps achieve selective two-photon excitation,
Biophotonics Intl. September:61–62.
Day, T., and Li Dessau, K.D., 1994, Narrow band tunable external-cavity diode
lasers offer new tools for researchers. Photonics Spectra March:99–103.
Dela Cruz, J.M., Pastrirk, I, Lozovoy, V.V., Walowicz, K.A., and Dantus, M.,
2004, Multiphoton intrapulse interference 3: Probing microscopic chemical environments, J. Phys. Chem. A. 108:53–58.
Demtröder, W., 1996, Laser spectroscopy. Basic concepts and instrumentation.
2nd edition.In: Springer Series in Chemical Physics, Springer-Verlag,
Berlin, Vol, 5, pp. 1–924.
Denk, W., Strickler, J.H., and Webb, W.W., 1990, Two-photon laser scanning
fluorescence microscopy, Science 248:73–76.
Duling III, I.N., 1993, Compact fiber soliton lasers produce ultrashort pulses,
Laser Focus World April:213–220.
Duling III, I.N., 1995, Compact sources of ultrashort pulses. Cambridge University Press, ISBN 0521461928.
Dunning, F.B., 1978, Tunable-ultraviolet generation by sum-frequency mixing,
Laser Focus Magazine May:72–76.
Dunsby, C., Lanigan, P.M.P., McGinty, J., Elson, D.S., Requejo-Isidro, J.,
Munro, I., Galletly, N., McCann, F., Treanor, B., Önfet, B., Davis, D.M.,
Neil, M.A.A., and French, P.M.W., 2004, An electronically tunable ultrafast laser source applied to fluorescence imaging and fluorescence lifetime
imaging microscopy, J. Phys. D: Appl.Phys. 37:3296–3303.
Draaijer, A., and Houpt, P.M., 1988, A standard video-rate confocal laserscanning reflection and fluorescence microscope, Scanning 10:139–145.
Driscoll, W.G., and Vaughan W., 1977, Handbook of Optics, McGraw-Hill
Book Co., New York.
Eden, J.G., 1988, UV and VUV lasers: Prospects and Applications, Optics
News April:14–27.
Ellis, G.W. 1979, A fiber-optic phase-randomizer for microscope illumination
by a laser. J. Cell Biol. 83:303a.
Ellis, G.W., 1988, Scanned aperture light microscopy, in Proceedings of the
46th Annual Meeting of EMSA. San Francisco Press Inc., San Francisco,
pp. 48–49.
Espinase, Ph., 2004, Microstructured Fibers Enable Supercontinuum Generation, SPIE OEMagazine, August:11.
Erlandson, A.C., and Powell, H.T., 1991, Ten thousand flashlamps will drive
the most-powerful laser, Laser Focus World August:95–100.
Fermann, M.E., Galvanauskas, A., and Sucha, G., 2002, Ultrafast lasers: Technology and applications, Marcel Dekker, Inc., New York.
Figueroa, J., 2002, High-power semiconductor lasers. The photonics design and
applications handbook, H-219-H-222, Laurin Publ. Co. Inc., Pittsfield,
Massachusetts.
Folkenberg, J.R., and Broeng, J., 2004, Photonic crystal fibers accommodate
high powers. Photonics Spectra September:82–86.
121
Forrester, S., 1994a, DC/DC converters: Theory of operation, Part 1, Sensors
January:28–35.
Forrester, S., 1994b, DC/DC converters: Theory of operation, Part 2, Sensors
February:64–69.
Fork, R.L., Martinez, O.E., and Gordon, J.P., 1984, Negative dispesion using
pairs of prisms, Opt. Lett. 9:150.
Franceschini, M.A., Fantini, S., and Gratton, E., 1994, LED’s in frequency
domain spectroscopy of tissues (J.R. Lacowicz). SPIE Proceedings. In:
Advances in laser and light spectroscopy to diagnose cancer and other diseases (R.R. Alfano, ed.), Los Angeles, Vol. 2135, pp. 300–306.
Friebele, E.J., and Kersey, A.D., 1994, Fiberoptic sensors measure up for smart
structures, Laser Focus World May:165–171.
Gabler, V., 2004, Miniature fibre lasers aim to modernize visible applications,
Europhotonics August/September:28–29.
Galvanauskas, A., and Samson, B., 2004, Large-mode-area designs enhance
fiber laser performance, SPIE’s oemagazine July:15–17.
Gibson, J., 1988, Laser Cooling Water. The key to Improved Reliability, Photonics Spectra Nov:117–124.
Gibson, J., 1989, Laser water cooling loops deserve attention, Laser Focus
World April:123–129.
Ginouves, P., 2002, Ion lasers, still a practical choice. The photonics design
and applications handbook, H-206–208, Laurin Publ. Co. Inc., Pittsfield,
Massachusetts.
Goldman, R.D., 1993, Air-to-liquid closed-loop cooling system meet the cost
performance goals of today’s laser market, Laser and Optronics February:15–17.
Günther, A.H., 1993, Optics damage constraints laser design and performance,
Laser Focus World February:83–87.
Guy, M., and Painchaud, Y., 2004, Fiber Bragg gratings: A versatile approach
to dispersion compensation, Photonics Spectra August:96–101.
Hammerling, P., Budgor, A.B., and Pinto, A., 1985, Tunable Solid-State Lasers,
In: Proceedings of the First International Conference, La Jolla, California, June:13–15, 1984, Springer-Verlag, Berlin.
Hard, R., Zeh, R.D., and Allen, R.D., 1977, Phase-randomized laser illumination for microscopy, J. Cell Sci. 23:335–343.
Häring, R., and Gerster, E., 2003, Semiconductor laser system fills yelloworange gap, Europhotonics August/September:38–39.
Hecht, E., and Zajac A., 2003, Optics, 4th Ed., Addison-Wesley Publishing Co.,
Reading, Pennsylvania.
Hecht, J., 1992, Ion lasers deliver power at visible and UV wavelengths, Laser
Focus World December:97–105.
Hecht, J., 1993a, Laser action in fibers promises a revolution in communications, Laser Focus World February:75–81.
Hecht, J., 1993b, Nitrogen lasers produce ultraviolet light simply, Laser Focus
World May:87–91.
Hecht, J., 1993c, HeCd lasers offer economical blue and ultraviolet light, Laser
Focus World August:67–71.
Hecht, J., 1993d, Copper-vapor lasers find specialized applications, Laser
Focus World October:99–103.
Hell, S., Witting, S., Schickfus, M.V., Wijnaendts van Resandt, A.W.;
Hunklinger, S., Smolka, E., and Neiger M., 1991, A confocal beam
scanning white-light microscope, J. of Microscopy 163:179–1876.
Herrmann, J., and Wilhelmi, B., 1987, Lasers for Ultrashort Light Pulses.
North-Holland Amsterdam.
Hess, S.T., and Webb, W.W., 2002, Focal volume optics and experimental artifacts in confocal fluorescence correlation spectroscopy, Biophys. J., 60:
2900–2917.
Higgins, T.V.. 1992, Nonlinear crystals: where the colors of the rainbow begin,
Laser Focus World January:125–133.
Hitz, B., 2004a, Something old, something new: German lab marries ring and
disk, Photonics Spectra September:22–23.
Hitz, B., 2004b, Thulium laser enables single-frequency oscillation in new
spectral region, Photonics Spectra September:28–303.
Hitz, B., 2004c, Hollow fiber delivers distortion-free femtosecond pulses, Photonics Spectra August:113–114.
Hitz, B., 2004d, Fiber laser system produces femtosecond pulses at 25 W of
average power, Photonics Spectra November:88–89.
Hobbs, J.R., 1993, Semiconductor lasers diversify, Laser Focus World CLEO
April:116.
PHC5 28/11/2005 6:15 PM Page 122
122
Chapter 5 • E. Gratton and M.J. vandeVen
Hobbs, J.R., 1994, Offset-plane mirrors transform laser beams, Laser Focus
World May:46–50.
Hodgson, D.J., 1994, How power-supply selection can improve laser-diode performance. Laser Focus World January:129–137.
Hodgson, N., and Weber, H., 2005, Laser resonators and beam propagation.
2nd ed., Springer series in optical sciences, Vol. 108.
Hoffmann, A., Meyer zu Hörste, G., Pilarczyk, G., Monajembashi, S., Uhl, V.,
and Gruelich, K.O., 2000, Optical tweezers for confocal microscopy, Appl.
Phys. B. 71:747–753.
Hogan, H., 2004, Imaging with three-part harmony, Biophotonics Intl. September:62–63.
Homburg, O., Finke, B., and Harten, P., 2003, High-precision homogenizers
shape beams, EuroPhotonics August/September:34–35.
Huth, B.G., and Kuizenga, D., 1987, Green light from doubled Nd-YAG lasers,
Lasers & Optronics October:59–61.
Kaiser, W., 1988, Ultrashort laser pulses and applications, Vol. 60, in Topics in
Applied Physics, Springer-Verlag, Berlin.
Kaminskii, A.A., 1981, Laser Crystals Their Physics and Properties (translation; H.F. Ivey ed.), Vol. 14, Springer Series in Optical Sciences, SpringerVerlag Berlin.
Kiat, L.S., Tanaka, K., Tsumanuma, T., and Sanada, K., 1992, Ultrathin singlemode imagefiber for medical usage, In: Optical Fibers in Medicine VII,
(A. Katzir, ed,), SPIE Proceedings, Los Angeles, California. SPIE
1649:208–217.
König, K., Liang, H., Berns, M.W., and Trmberg, B.J., 1996, Cell damage in
near-infrared multimode optical traps as a result of multipoint absorption,
Optics. Lett. 1090–1092.
König, K., Becler, T.W., Fischer, P., Riemann, I., and Halbhuber, K.-J., 1999,
Pulse-length dependence of cellular response to intense near-infrared laser
pulses in multiphoton microscopes, Optics Lett. 24:113–115.
Kraft, T., 2004, Compact, robust lasers suit biotechnology applications, Biophotonics Intl. July: 44–46.
Krueger, A., and Féru, Ph., 2004, Ytterbium Tungstate revolutionizes then field
of high-power ultrafast lasers, Photonics Spectra March:46–51.
Kudryashov, A.V., and Weber, H., 1999, Laser resonators: Novel design and
development, SPIE Monographs PM67, pp. 1–317.
Krupke, W.F., 2004, Diode-pumped visible wavelength alkali laser, US Patent
6693942.
Landgraf, S., 2001, Application of semiconductor light sources for investigations of photochemical reactions, Spectrochim. Acta A, 57:2029–
2048.
Landgraf, S., 2003, Handbook of Luminescence, Display Materials and
Devices, Vol. 3, Chapter 9, pp. 371–398, ISBN:1–58883–032–2., American Scientific Publishers (ASP), Stevenson Ranch, California.
Leitz, G., Fallman, E., Tuck, S., and Axner, O., 2002, Stress Response in
Caenorhabditis elegans Caused by Optical Tweezers: Wavelength, Power,
and Time Dependence. Biophys. J. 82:2224–2231.
Lewis, R.R., Naylor, G.A., and Kearsley, A.J., 1988, Copper Vapor Lasers
Reach High Power, Laser Focus/Electro Optics April;92–96.
Limpert, J., Liem, A., Schreiber, T., Röser, F., Zellmar, H., and Tünnermann,
A., 2004, Scaling single-mode photonic crystal fiber lasers to kilowatts,
Photonics Spectra May:54–63.
Lin, J.T., and Chen, C., 1987, Choosing a Non-linear Crystal, Lasers & Optronics November:59–63,
Liu, T.-M., Chu, S.-W., Sun, C.-K., Lin, B.-L., Cheng, P.-C., and Johnson, I.,
2001. Multiphoton confocal microscopy using a femtosecond Cr:Forsterite
laser, J. Scanning Microscopies 23:249–254.
Littlechild, J., and Mossler, D., 1988, Knowledge of Arc-Lamp Aging and Lifetime Effects Can Help to Avoid Unpleasant Surprises, Laser Focus/Electro
Optics November:67–76.
MacMullin, T., 2004, Safety first, SPIE’s oemagazine June:27–29.
Malzahn, U., 2004, Driving diode lasers is straightforward, Europhotonics
August/September:22–23.
Marshall, L., 1994, Biological monitoring foreseen with ultraviolet light
source, Laser Focus World April:83–87.
McConnell, G., 2004, Confocal laser scanning fluorescence microscopy with
a visible continuum source, OpticsExpress 12:2844–2850.
Miller, P.J., 1991, Taming Laser Noise: Methods and Applications, Photonics
Spectra April:183–187.
Miller, P., and Hoyt, C., 1986, Turning Down Laser Noise with Power Stabilizers, Photonics Spectra June:129–134.
Mollenauer, L.F., and White, J.C., 1987, Tunable Lasers, Vol. 59. Topics in
Applied Physics, Springer-Verlag, Berlin.
Mooradian, A., 1993. External cavity tunable diode lasers, Lasers & Optronics May:35–37.
Moreaux, L., Sandre, O., and Mertz, J. 2000, Membrane imaging by secondharmonic generation microscopy, JOSA B 17:1685–1694.
Mortensen, P., 1994, Solid state lasers: Russians commercialize femtosecond
laser, Laser Focus World April:36–38.
Muckenheim, W., Austin, L., and Basting, D., 1988, The pulsed dye Laser:
Today’s Technology, Today’s Uses, Photonics Spectra June:79–84.
Nan, X., Yang, W.-Y., and Xie, X.S., 2004, CARS microscopy lights up lipids
in living cells, Biophotonics Intl. August:44–47.
Neuman, C., Chadd, E.H., Liou, G.F., Bergman, K., and Block, S.M., 1999,
Characterization of Photodamage to Escherichia coli in Optical Traps,
Biophys. J. 77:2856–2863.
Nordborg, J., and Karlsson, H., 2004, Multi DPSS lasers — a true Ar-ion alternative, Europhotonics June/July:28–29.
Okhotnikov, O.G., Gomes, L.A., Xiang, N., Joukti, T., Chin, A.K., Singh, R.,
and Grudinin, A.B., 2003, 980-nm picosecond fiber laser, IEEE Photonics Technol. Lett. 15:1519–1521.
Perry, M.D., Payne, S.A., Ditmire, T., and Beach, R., 1993, Better materials
trigger Cr:LiSAF laser development, Laser Focus World September:29:85–92.
Peterman, E.J.G., Gittes, F., and Schmidt, C.F., 2003, Laser induced heating in
optical traps, Biophys. J. 84:1308–1316.
Peuse, B., 1988, Active stabilization of ion laser resonators. Active stabilization offers advantages in several areas, Lasers & Optronics November:61–65.
Piehler, D., 1993, Upconversion process creates compact blue/green lasers,
Laser Focus World November:95–102.
Piston, D.W., Sandison, D.R., and Webb, W.W., 1992, Time-resolved fluorescence imaging and background rejection by two-photon excitation in laser
scanning microscopy. In: Time-Resolved Laser Spectroscopy in Biochemistry III (J.R. Lakowicz, ed.), SPIE Proceedings Los Angels, California,
1640:379–389.
Piston, D.W., Kirby, M., Cheng, H., Lederer, W.J., and Webb, W.W., 1994,
Three dimensional imaging of intracellular calcium activity by two-photon
excitation laser scanning microscopy, Appl. Optics. February:33:662–669.
Potma, E.O, de Boeij, W.P., Pshenichnikov, M.S., and Wiersman, D.A., 1998,
A 30-fs cavity-dumped optical parametric oscillator, Opt. Lett.
23:1763–1765.
Radunsky, M.BV., 2002, The OPO: A research tool and more. The photonics
design and applications handbook book 3, H-261–H264, Laurin Publ. Co.
Inc., Pittsfield, Massachusetts.
Rapp, E.W., 1988, Design your cooling system for good laser performance,
Laser Focus/Electro Optics September:65–70.
Reichle, C., Sparbier, K., Müller, T., Schnelle, Th., Walden, P., and Fuhr, G.,
2001, Combined laser tweezers and dielectric field cage for the analysis
of receptor-ligand interactions on single cells, Electrophoresis
22:272–282.
Robinson, K., 1998, Training and equipment help ensure laser safety, Biophotonics Intl. November/December:42–47.
Rockwell Associates Inc., 1983, Laser Safety Training Manual, 6th ed., Rockwell Associates Inc., Cincinnati, Ohio.
Rockwell, R.J. Jr., 1986, An introduction to exposure hazards and the evaluation of nominal hazard zones, Lasers & Applications May:97–103.
Rusu, M., Karirinne, S., Guina, M., Grudinin, A.B., and Okhotnikov, O.G.,
2004, Femtosecond Neodymium-doped fiber laser operating in the
894–909-nm spectral range, IEEE Photonics Technol. Lett. 16:1029–1031.
Schneider, D.J., and Williams, D.C., 1993, Fighting corrosion in laser cooling
systems, Laser Focus World December:110.
Schulmeister, K., 1994, Safety regulations changes, Biophotonics Intl.
August:38–42.
Scifres, D.R., 1994, Diode lasers ride the wave of progress, Photonics Spectra
84–85.
Silfvast, W.T., 2004, Laser Fundamentals, 2nd ed. Cambridge University Press,
New York.
PHC5 28/11/2005 6:15 PM Page 123
Laser Sources for Confocal Microscopy • Chapter 5
123
Sivers, N.L., Van de Workeen, B.C., and Lee, S.M., 2004, Improving fluorescence confocal microscopy with cryogenically-cooled diode lasers, Optics
Express 12:4157–4165.
Sliney, D.H., 1986, Laser Safety. The newest face on an old standard, Photonics Spectra April:83–96.
Sliney, D., and Wolbarsht, M., 1980, Safety with lasers and other optical
sources. A comprehensive handbook, Plenum Press, New York.
Sliney, D.H., 1994, Laser safety concepts are changing, Laser Focus World
May:185–192.
Smith, B., 1986, Lamps for Pumping Solid-state Lasers: Performance and Optimization, Laser Focus/Electro Optics September:58–73.
Smith, K., and Lucek, J.K., 1993, Modelocked fiber lasers promise high-speed
data networks, Laser Focus World October:85–91.
Smith, S.P., Bhalotra, S.R., Brody, A.L., Brown, B.L., Boyda, E.K., and Prentiss, M., 1999, Inexpensive optical tweezers for undergraduate laboratories, Am. J. Phys. 67:26–35.
Snyder, J.J., and Cable, A.E., 1993, Cylindrical microlenses improve laserdiode beams, Laser Focus World February:97–100.
Soileau, M.J., 1987, Laser-Induced Damage, Photonics Spectra November:109–114.
Stry, S., Knispel, R., Hildebrandt, L., and Sacher, J., 2004, Tunable external
cavity diode laser tackles high-power applications, EuroPhotonics February/March:26–27.
Stitch, M.L., 1979, Laser Handbook, Vol. 3, North-Holland Publishing Co.,
Amsterdam.
Szarowski, D.H., Smith, K.L, Herchenroder, A., Matuszek, G., Swann, J.W.,
and Turner, J.N., 1992, Optimized reflection imaging in laser confocal
microscopy and its application to neurobiology, Scanning 14:104–111.
Tai, S.-P., Chan, M.-C., Tsai, T.-H., Guol, S.-H., Chen, L.-J., and Sun, C.-K.,
2004, Two-photon fluorescence microscope with a hollow-core photonic
crystal fiber, Optics Express 12:6122–6128.
Tebo, A.R., 1988, Scientists develop Useful Optical Materials, Laser
Focus/Electro Optics August:103–110.
Tozer, B.A., 2001, Revised safety standards to benefit laser design and use,
Laser Focus World March:81–82.
Unger, B., 1994, Device saves laser diodes from electrostatic-discharge
damage. Laser Focus World May:238–240.
Wang, X.F., Kitajima, S., Uchida, T., Coleman, D.M., and Minami, S., 1990,
Time-resolved fluorescence microscopy using multichannel photon counting, Appl. Spectrosc. 44:25–30.
Wang, X.F., 1990, Fundamental studies on time-resolved fluorescence image
spectroscopy techniques. Dissertation, Osaka University.
Wang, X.F., Periasamy, A., Herman, B., and Coleman, D.M., 1992, Fluorescence lifetime imaging microscopy (FLIM): Instrumentation and applications. Critical Reviews in Analytical Chem. 23:369–395.
Weast, R.C., and Tuve, G.L., 1971, Handbook of lasers with selected data on
optical technology, CRC Press. The Chemical Rubber Co., Cleveland,
Ohio.
Webb, C.E., and Jones, J.D.C., 2004, Handbook of laser technology and applications. Vol.1, Principles, Vol. 2, Laser design and laser optics, Vol. 3,
Applications, IOP Publ., Bristol, UK.
Weber, M.J., 1999, Handbook of laser wavelengths, CRC Press, New York.
Winburn, D.C., 1985, Practical laser safety, Marcel Dekker Inc., New York.
Wisdom, J., and Digonnet, M., 2004, Ceramic lasers: Ready for action, Photonics Spectra February:50–56.
Woods, S., 2003. Fiber lasers offer alternative for analytical applications, Biophotonics Intl. December:48.
Yang, C., and Mertz, J., 2003, Transmission confocal laser scanning
microscopy with a virtual pinhole based on non-linear detection, Optics
Lett. 28:224–226.
Yelin, D., Oron, D., Korkotian, E., Segal, M., and Silberberg, Y., 2002,
Third-harmonic microscopy with a titanium-sapphire laser, Appl. Phys. B.
74:S97.
Zeek, E., Maginnis, K., Backus, S., Russek, U., Murmane, M., and Vdovin, G.,
1999, Pulse compression by use of deformable mirrors, Opt. Lett.
24:493–495.
Zhang, Z.X., Sonek, G.J., Wei, X.B., Sun, C., Berns, M.W., and Tromberg, B.J.,
1999, Cell viability and DNA measurements by two-photon fluorescence
excitation in CW Al:GaAs diode laser optical traps, J. Biomed. Optics
4:256–259.
APPENDIX
More Specific
LIST OF USEFUL WEBSITES AND TUTORIALS
(NOT EXHAUSTIVE)
Samual M. Goldwasser. Laser sales and services, http://www.repairfaq.
org/sam/laserlps.htm#lpslss
Dr. Alexander N. Cartwright, http://www.ece.buffalo.edu/faculty/cartwright/
links.html
Books: Lists of General Reference Books
on Lasers and Applications
http://micro.magnet.fsu.edu/primer/lightandcolor/lasersreferences.html
Spring, K.R., Parry-Hill, M.J., et al., Molecular Expressions, Florida State
University, Talahassee, Florida.
Dr. Rami Arieli, http://www.phys.ksu.edu/perg/vqm/laserweb/
Laser Information: Cleaning of Laser Optics
http://www.repairfaq.org/sam/laseratr.htm#atrcln
General Information with Many Links
http://resonator.physics.sunysb.edu/laser/laserlinks
Buyer’s Guides
Photonics Laser Buyer’s Guide, Laurin Publishing, http://www.photonics.com
Institute of Physics (IOP) Publishing Ltd., Temple Back, Bristol, UK, Ph.
+44-(0)117–9297481- http://optics.org/buyers/
General References: Company Listings
and Buyer’s Guides
General
http://www.photonics.com/directory/bg/XQ/ASP/QX/index.htm
Laurin Publishing, 2002, Photonics Directory, Buyer’s Guide. Brief, instructive chapters on optics, optical components, and comparison of developments for the various laser types in Book 3, The Photonics Design and
Applications Handbook.
History of the Laser
http://www.bell-labs.com/history/laser/
Schawlow and Townes biographies
Titanium–Sapphire Laser
http://www.drbio.cornell.edu/Infrastructure/Apparatus_WWW/Ti%20
Sapphire%20Lasers.html
Evolution of Ti:Sa laser systems
Pulse Length Measurement by Optical Means
http://linac.ikp.physik.tu-darmstadt.de/fel/optical_pulse.html
PHC5 28/11/2005 6:15 PM Page 124
124
Chapter 5 • E. Gratton and M.J. vandeVen
Safety Issues
The health and safety committee at educational institutions should have Standard Operating Procedures (SOP) for laser operation and maintenance
available. As examples may serve:
http://phantom.ehs.uiuc.edu/~rad/laser/
http://www2.umdnj.edu/eohssweb/aiha/technical/lasers.htm
http://web.princeton.edu/sites/ehs/laserguide/appendixC.htm
Laser safety self-audit checklist, http://www.microscopyu.com/articles/
fluorescence/lasersafety.html
Spring, K.R., Parry-Hill, M.J., et al., Molecular Expressions, Florida State University, Talahassee, Florida. Laser classifications, hazards, and bioeffects
of various laser types, a brief and concise overview worth your attention.
http://stwi.weizmann.ac.il/Lasers/laserweb/
http://www.phys.ksu.edu/perg/
vqm/laserweb/Misc/Apindex.htm
Dr. Rami Arieli, The Laser Adventure, Virtual school
Rockwell Laser Industries. Safety materials, regulations: laser safety standards
organizations and products, e.g., Z136.1 ANSI document, laser accident
database, access control measures, and courses, http://www.rli.com/
Rockwell Laser Industries safety issues and links laser bioeffects, manufacturers, standards, and laser user institutes and laboratories,
http://www.rli.com/resources/links.asp?Link=3
http://repairfaq.ece.drexel.edu/sam/CORD/leot/
Laser Electro-Optics Technology (LEOT), Center For Occupational Research
and Development (CORD), Waco, Texas,electrical safety issues.
Standards, hazards, and solutions, http://www.osha.gov/SLTC/laserhazards
Specific Lasers
Argon-Ion Laser
Advanced Level
http://micro.magnet.fsu.edu/primer/techniques/microscopylasers.html
Spring, K.R., Parry-Hill, M.J., et al., Molecular Expressions, Florida State University, Talahassee, Florida. Beam divergence, beam expander, fiber-optic
coupling (*), single and multiple mode fiber-optic, laser types.
Expert Level
Dr. Gavin D. Reid, Leeds University, and Klaas Wynne, University of
Strathclyde, Glasgow, UK, Introduction to ultrafast laser systems and
optical characterization, dispersion and its control, pulse broadening, amplification, optical Kerr effect, white light generation, OPA, auto and cross correlation,http://phys.strath.ac.uk/12-503B2/introduction/introduction.html
Construction of a fs mode-locked laser with a nice Ti:sa energy diagram,
http://www.df.unipi.it/~fisapp/Gruppi/Metrologia/spiegazioni/boris.pdf
Troubleshooting
http://www.repairfaq.org/sam/laseratr.htm#atrcln
Tutorial Sites
Classification of sites:
• Basic level indicates many images, applets but no formulae.
• Advanced level with more thorough description of the issues supported
often with applets.
• Expert level with equations and examples.
Basic Level
Ionized gas spectra, construction characteristics, http://www.lexellaser.com/
techinfo_gas-ion.htm
Laser wavelengths chart, http://www.lexellaser.com/techinfo_wavelengths.htm
John Gormalli, stimulated emission explained, Web Science Resources Net
(WSRNet), http://members.aol.com/WSRnet/tut/t1.htm
Stimulated emission explained, laser theory quiz, http://home.achilles.net/
~ypvsj//quiz/index.html
Laser theory and applet on stimulated emission, http://www.point-source.com/
LaserTheory.asp
Basic atom model, general laser classification and safety issues, http://www.
howstuffworks.com/laser.htm
One page, ruby laser design and applications, http://home.achilles.net/
~ypvsj//glossary/laser.html
Fraunhofer Institute’s laser tutorial, http://www.ilt.fhg.de/ger/lasertutorial.html
Dye Laser
Advanced Level
Commercial laser dye as well as CLSM companies have colorful wall charts
and list dyes and matching laser lightsources on their sites.
http://micro.magnet.fsu.edu/primer/lightandcolor/laserhome.html
Spring, K.R., Parry-Hill, M.J., et al., Molecular Expressions, Florida State University, Talahassee, Florida. Laser fundamentals, safety issues, microscopy
and lasers used, intensity modulation, references, many instructive applets
worthwhile to visit.
http://www.olympusmicro.com/primer/lightandcolor/lasersintro.html
Samual M. Goldwasser’s site, “A Practical Guide to Lasers for Experimenters
and Hobbyists,” Version 6.60, very extensive with safety issues, reference
books and materials, schematics, construction, safety, repair issues,
http://www.repairfaq.org/sam/laserfaq.htm
Dr. Alexander N. Cartwright, University of Buffalo, laser gain and pumping
schemes, photonics educational applets, http://www.ece.buffalo.edu/
faculty/cartwright/photonics/rateequations.html
One page argon-ion laser basic construction and emission properties, one page
overview of laser types and applications, basic level, http://www.rli.com/
resources/argon.asp
Basic argon-ion laser, graph of relative laser line intensities, http://www.rli.
com/resources/argon.asp
Advanced Level
Advanced Level
Dr. R. Alexander, http://members.aol.com/kpublish/Laser/Dye_Laser.html
Dr. Ladic site with excitation and emission spectrum maxima for dyes useful
for CLSM, http://www.cs.ubc.ca/spider/ladic/images/fluoro.gif
Expert Level
Instructive site from LEOT, http://repairfaq.ece.drexel.edu/sam/CORD/leot/
course03_mod10/mod03-10.html
Colliding Pulse Dye Laser
Expert Level
http://utol.ecen.ceat.okstate.edu/thz%20interactive%20tour/cpmlaser.htm
Fiber Lasers Tutorial
Construction, wavelengths, and optical characteristics, http://www.imra.com/
lasers-tech-tut-frame-pdf.html
Titanium-Sapphire Laser
Basic Level
Evolution of Ti:Sa laser systems, http://www.drbio.cornell.edu/Infrastructure/Apparatus_WWW/Ti%20Sapphire%20Lasers.html
Expert Level
Online educational resources for physics teachers
http://www.ba.infn.it/www/didattica.html
Dr. Rami Arieli, The Laser Adventure, virtual school about lasers and their
applications. Optics aspects, basic laser theory, laser cavities, lasing
modes, fluorescence, laser line shape and width, laser beam properties,
power, applets, applications, laser safety issues, questions, reference
books, very extensive and useful also for course development,
http://www.phys.ksu.edu/perg/vqm/laserweb/
Laser Electro-Optics Technology (LEOT), Center For Occupational Research
and Development (CORD), Waco, Texas. Introduction to laser optics, electronics, components, systems, electrical safety issues, http://repairfaq.
ece.drexel.edu/sam/CORD/leot/
PHC5 28/11/2005 6:15 PM Page 125
Laser Sources for Confocal Microscopy • Chapter 5
Dr. P.L. Cross, Optical concepts, laser cavity design, laser operation, database
of spectral properties of lasing materials, absorption and emission spectra,
http://aesd.larc.nasa.gov/GL/tutorial/laser/las_out.htm
List Servers
http://www.msa.microscopy.com/MicroscopyListserver/
MicroscopyArchives.html
Confocal microscopy list server for maintenance and trouble shooting. When
there is a problem often it helps to look for similar experiences and solutions on the web. A good site to start: http://listserv.acsu.buffalo.edu/
archives/confocal.html
Microscopy Laboratory Sites
125
photostability, fluorescence techniques, e.g., FP(photobleaching)R
(ecovery), FCS, SPT, cell viability, resolution with an underfilled objective, real-time image acquisition. A lot of specialized information together
for live cell imaging and spectroscopy, http://www.drbio.cornell.edu/
Infrastructure/Infrastructure%20Index.html
Dr. Bruce Jenks, Nijmegen University, The Netherlands. Laser principle,
optical scanner principle, optical sectioning, single and multi-photon illumination (dis)advantages, http://www.celanphy.sci.kun.nl/Bruce%20web/
scanning%20microscopy.htm
Optics
Ultrafast optics, http://www.newport.com/Support/Tutorials/Optics/o4.asp
Dr. W. Webb, DRBIO, Cornell University, Ithaca, New York. Multi-photon
excitation, physical principles, excitation cross-sections, fluorophore
LIST OF ABBREVIATIONS
AC
ASE
AOTF
AOBS
AOD
AOM
BBO
CCD
CDRH
CLSM
CTA
CW
DC
FBG
FCS
FHG
FITC
FLIM
FLIP
FRAP
FROG
GFP
GreNe
GRIN
GTI
GVD
He-Cd
He-Ne
HR
ICNIRP
IEC
IR
KLM
KTP
LASER
LBO
Alternating Current
Amplified Spontaneous Emission
Acousto-Optic Tuning Filter
Acousto-Optic Beam Splitter
Acousto-Optic Deflector
Acousto-Optic Modulator
Beta Barium Borate
Charge-coupled Device
Center for Devices and Radiological Health
Confocal Laser Scanning Microscopy
Cesium Titanyl Arsenate
Continuous Wave
Direct Current
Fiber Bragg Grating
Fluorescence Correlation Spectroscopy
Fourth Harmonic Generation
Fluorescein IsoThioCyanate
Fluorescence Lifetime Imaging Microscopy
Fluorescence Loss in Photobleaching
Fluorescence Recovery After Photobleaching
Frequency-Resolved Optical Gating
Green Fluorescent Protein
Green Helium-Neon
Graded Index Lens
Gires-Tournois Interferometer
Group Velocity Dispersion
Helium–Cadmium
Helium–Neon
High Reflector Mirror
International Commission on Non-Ionizing Radiation Protection
International Electrotechnical Commission
Infrared
Kerr-Lens Mode-Locking
Potassium Titanium Oxide Phosphate
Light Amplification by Stimulated Emission of
Radiation
Lithium TriBorate
LED
LMA
LPC
MII
MOPA
NA
NCPM
Nd-YAG
Nd-YLF
Nd-YVO4
NLO
NMR
OC
OCT
OD
OPA
OPO
PCF
PICASO
PZT
Q-switch
RF
SAM
SBR
SD
SESAM
SHG
SMD
SOC
SPIDER
SPT
TEC
TEM
THG
TIRFM
UV
VECSEL
VIS
Light-Emitting Diode
Large Mode Area
Laser Power Controller
Multi-Photon Intrapulse Interference
Master Oscillator Parametric Amplifier
Numerical Aperture
Non Critical Phase Matching
Neodymium–Yttrium–Aluminum Garnet
Neodymium–Yttrium Lithium Fluoride
Neodymium–Yttrium Ortho Vanadate
Non-Linear Optical
Nuclear Magnetic Resonance
Output Coupler Mirror
Optical Coherence Tomography
Optical Density
Optical Parametric Amplifier
Optical Paramagnetic Oscillator
Photonic Crystal Fiber
Phase and Intensity from Correlation and Spectrum
Only
Piezo Tuning
Quality (of the laser resonator) switch
Radio Frequency
Saturable Absorber Mirror
Saturable Bragg Reflector
Static Discharge
Semiconductor Saturable Absorber Mirror
Second Harmonic Generation
Single Molecule Detection
Saturable Output Coupler
spectral phase interferometry for direct electric-field
reconstruction
Single Particle Tracking
Thermo-Electrically Cooled
Transverse electromagnetic
Third Harmonic Generation
Total Internal Reflection Fluorescence Microscopy
Ultraviolet
Vertical External Cavity Surface Emitting Laser
Visible
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising