Re: How much laser power is required to kill a spider?

I think the answer is a little over 2 Watts per square centimeter.  That 
number comes from the safety manual in my laser lab, specifiying the power 
that can potentially set objects on fire.  However, being a safety manual, 
I'm sure that number is a little low.  To get the exact number, for any 
particular spider, there's a relatively simple experiment you can do.

Take a spide outside on a sunny day.  Spiders can survive just fine in 
bright sunlight (even if they don't like it), and sunlight has a peak power 
of 0.4 Watts per square centimeter.  Once you start playing around with a 
magnifying glass, I know you can turn ants to toast (from personal 
experience), and I assume the same can be done with most spiders (although 
maybe giant tarantulas would be tougher?).

Bring the magnifying glass down on the spider slowly, until the focused 
light becomes just bright enough to have the desired effect on the spider.  
(PETA members, consider this a thought experiment.)  Then, keeping the 
magnifying glass in the same place, measure the diameter of the focused 
spot of sunlight.  Call this number A.  Measure the diameter of the 
magnifying glass; call this B.

The answer to your question is then:

0.4 (B/A)^2 Watts per square centimeter.

As for whether a laser would do the job better than sunlight...  I doubt 
it.  Both lasers and sunlight are very directed, so the only difference is 
the coherence of the light.  A laser might BLIND a spider more easily than 
sunlight, depending on the wavelength, but in terms of whole-body damage 
the two forms of light should require roughly the same power.

An additional answer was received from Steve Guch:


How much laser power does it take to kill a spider?

The most direct answer to the question is that probably nobody knows, sinc=
e
there is not much in the laser and electro-optics technical literature abo=
ut
spiders.

Absent hard data, the best bet is to try to break down the problem into a
number of parts and see if a reasonable answer (or answers!) can be deduce=
d
(or guessed, more likely).

Several issues that need to be addressed are the following:
  1. What kind of spider is to be killed?
  2. What is the postulated "kill mechanism"?
  3.  What kind of laser is available?
  4.  Last, but not least, a question that many physicists don=92t like =
=96 Does
it make sense to
       to try to kill spiders with lasers?

1.  What kind of spider is to be killed?

While it=92s probably not very important to determine the species of arach=
nid to
be offed, it is important (for some kill mechanisms) to determine the geom=
etry
and composition of the spider.  The range of possibilities is vast, from t=
eeny
little submicroscopic guys (the smallest are probably in the 1/10 millimet=
er
diameter range) to the giant guys (like tarantulas, something like 3 or 4 =
cm
in body dimension with legs extending out another 3 or 4 cm on both sides)=
.

Because I=92m a physicist, the approach we=92ll follow is to postulate a g=
eneric
spider and make some rough calculations, then it will be left to you to mo=
dify
the calculations to fit the particular kind of spider that you are dealing
with=85  This approach is referred to in an old physics joke (the only jok=
e many
of us remember) called "=85 the spherical chicken in a vacuum=85" approach=
 to
making approximations needed for solving a problem, but that=92s for anoth=
er
time.

For our purposes, we=92ll assume an annoying but not terrifying spider wit=
h a
spherical body that=92s 2 mm in diameter, with 8 legs each 4 mm long and 0=
.5 mm
in diameter.  We=92ll also assume that the spider has a thin exoskeleton t=
hat=92s
roughly equivalent in properties to the plastic tape used in video cassett=
e
players, over internal organs that are pretty much all water.  The rough
volume of the spider can be calculated by assuming it=92s spherical in sha=
pe, so
volume =3D (4/3) * pi (3.14159) * the square of the radius, or just about =
.0042
cubic centimeters. 


2.  What is the postulated kill mechanism?

This is probably the hardest part, because of the variety of possibilities=
 and
lack of detailed lethality information from the literature.  A couple of
possibilities are summarized below.

A.  Instant Disassembly.  Hit it with enough energy instantly to vaporize
relatively large amounts of the body and disassemble the poor creature.
Anecdotal information from laser fusion experiments indicates that the rap=
id
deposition of something like 10,000 joules of energy per cubic centimeter =
of
anything in a period shorter than it takes heat to be removed easily =96 s=
horter
than 1 millisecond, for example =96 will pretty well cause anything to fly=
 apart
into unrecognizable bits.

B.  Steam Explosion.  Boil some of the water in the spider and cause a ste=
am
explosion that will rupture his body and cause relatively rapid death.

C.  Bleed To Death.  Puncture the exoskeleton and allow the fluids and org=
ans
to drain out, causing a slow and agonizing death.


D.  Mobility Kill.  Hit enough limbs with enough energy in a succession of
pulses to excise them at which point where the spider loses its ability to
hunt and expires.  This will probably result from a number of Instant
Disassembly events, with energy requirements per A. above.


3.  What Kind Of Laser Is Available?

For A. and D. kill mechanisms, the most likely types of lasers are pulsed
Nd:YAG and CO2 lasers.  These brutes typically put out pulses 0.1 to 1 jou=
le,
so for convenience we=92ll assume that we have a 0.3 joule pulse laser
available.  The individual pulse energy density from such a laser would be=
 0.3
joule per .004 cubic centimeters, or only about 75 joules per cubic centim=
eter
if you focused the beam to expose the whole spider =96 clear not enough to
produce Instant Disassembly of the whole spider from a single pulse. On th=
e
other hand, if you focused on one of the legs the volume exposed to the be=
am
would be a cylinder 0.5 mm long and of cross sectional area with 0.5 mm
diameter =96 which is a volume of  (0.05 cm length) * ( pi * (0.05 cm /2)
squared) or about .0001 square centimeters.  If the leg is exposed to the
beam, the pulse energy density would be 0.3 joule per .0001 square centime=
ter,
or about 1000 joules per square centimeter=85  probably enough to do some
damage, but probably not enough to reliably blow the limb off.  The best b=
et
for a Mobility Kill might be to try to fire multiple pulses at a leg =96 3
seconds at 30 pulses per second might be feasible =96 but my guess is that=
 it
might be very hard to keep the beam on the spider after it=92s had the inc=
entive
of being hit by fairly jarring blasts from sources unknown.  The conclusio=
n to
be derived from this is that using a laser to produce either an Instantane=
ous
Disassembly or Mobility Kill would be so hard as not to be too useful,
particularly since it=92s unlikely that 100% of the beam would be absorbed=
 on
every pulse=85 10-30% is probably a more reasonable number, so it=92s pret=
ty
apparent that this is not a particularly effective way to kill a spider.

For mechanism B, it might be possible to use either a Nd:YAG or CO2 laser,=
 or
even a high powered laser diode.  The first two can easily put out more th=
an
10 watts of average power in either pulse or continuous operation, and the
most capable laser can produce a few watts of output from a very small pac=
kage
diodes (these are NOT the kind in laser pointers, which only produce .001
watts output or so).  To figure out how much power might be needed to init=
iate
boiling of the spider=92s internal fluids, we can assume that the internal=
s are
all water and that it takes 1 calorie of energy to raise the temperature b=
y 1
degree C.  Since water has a mass of about 1 gram per cc, the total mass o=
f
water in the spider is about .004 grams.  The amount of energy needed to r=
aise
the temperature by 1 degree C can be estimated to be about .004 calories, =
or
about 0.016 joules.  To get the spider from room temperature =96 about 25 =
C =96 to
100 C takes about 1.6 joules, based on this logic.  To boil water requires
about 539 calories or about 2300 joules per cubic centimeter =96 for our t=
iny
spider, the amount of energy to boil all his innards corresponds to about =
9
joules.  Stopping for a moment, 2 things are apparent:  it takes a lot mor=
e
energy to boil water than to move its temperature around, even a lot, and =
it
takes a lot less energy to boil something that=92s mostly water than to
completely disassemble it (into a gaseous plasma of miscellaneous ions and
electrons and things).  But, back to the spider, we need about 10 joules t=
o
boil all of its innards and assure that it=92s thoroughly cooked dead.  My
guess, however, is that one of two things would make a thermal (heat kill)
easier than this:  (1) if the spider got to 100 C, it would be dead just
because it=92s biological processes can=92t stand the heat (think about ho=
w you
would feel if your body were brought to the temperature of boiling water!)=
;
and, (2) you probably would only have to boil a little bit of the innards =
to
cause the steam to increase the pressure inside the animals exoskeleton to
rupture the exoskeleton and kill it.  So, if we believe that 1 joule will =
be
needed, it appears that this should be feasible within 0.1 to 1 seconds, e=
ven
if we allow for modest absorption of the beam by the spider.  It=92s not c=
lear
that the spider would sit still for this insult, but it does look like the=
se
types of laser might work for the Steam Explosion mechanism at powers of
order3 to 10 watts (if the beam is properly focused, of course).

For kill mechanism C, it=92s only necessary to poke a small hole in the sp=
ider
and let the oozing begin.  In this case, it=92s probably reasonable to ass=
ume
that we should poke a hole that=92s much bigger in size than particles of =
some
reasonable size which might plug it up and stop the process =96 it=92s unl=
ikely
that anybody knows an exact number for this, but it=92s probably reasonabl=
e to
assume that if you drilled a hole 1/10th of the spider=92s size, it would =
have a
hard time stopping the bleeding =96 think how you=92d react to a hold 1/10=
th your
height!  The hole needed can be estimated to be 0.02 centimeters in diamet=
er,
by this reasoning.  If we assume the spider=92s skin is only 1/100th of it=
=92s
body diameter (again, think about your skin=85 so this might be reasonable=
, if
not precise), or about 0.002 centimeters in thickness.  The volume of the =
skin
that would have to be removed can be calculated to be (very roughly) .02 *=
 .02
* .002 cubic centimeters, or about 0.000001 cubic centimeters.  If we take=
 a
pulsed laser of .3 joules, the energy density is about 100,000 joules per
cubic centimeter =96 clearly enough to pop the skin in a single pulse.  If=
 we
consider a continuously running laser of 3 watts output, it would only tak=
e
about 1/3 of a second to penetrate the skin (since 3 watts is 3 joules per
second, and the spider wouldn=92t have enough time to remove heat in 1/3 s=
econd
=96 although it might well move!).  

The bottom line is that it=92s probably more reasonable, from the standpoi=
nt of
minimizing the energy involved, to use a pulsed laser to poke a hole throu=
gh
the spider=92s exoskeleton than to use any other mechanism.  A single puls=
e
could do the job, avoiding any possibility of a need to track the little b=
ug
and steer the beam accurately to keep it on the same spot.

Nothing=92s simple, however, since pulsed lasers of this type typically co=
st
about $30-$50K, and it would probably take another few $K to build the opt=
ics
to focus the beam =96 even if you were going to aim at a single spot and w=
ait
for the spider to wander through.  To build a beam-steering and tracking s=
et
would cost $100=92s of $K more =96 even more impractically expensive than =
the cost
of the laser.  The advantage of the beam-steerer/tracker is, of course, th=
at
you could use a cheap and easy high power laser diode =96 which would only=
 set
you back $3-10K.  From an economics standpoint, then, it would cost somewh=
ere
between $30K and $100K to set up to do the job from scratch =96 ignoring y=
our
salary, if any.


4.  Does It Make Sense?

Absolutely NOT.  Even if you had all the equipment together for other reas=
ons,
it probably doesn=92t make sense to use lasers to kill spiders.  As a lot =
of
research has pretty well established over the past couple of decades, use =
of
lasers to deposit energy at a remote location is feasible but dreadfully
expensive.  Generally, more cost effective approaches involve the applicat=
ion
of kinetic energy (i.e., flyswatters, shoes, newspapers) provided by cheap
chemical energy (i.e., digestion of food driving muscles to propel the abo=
ve
objects) or use of chemicals directly (e.g., Raid or other insecticides). =
 All
of these involve relatively crude application of overkill phenomenom, but =
are
exceptionally cheap because of the relatively low-tech delivery and aiming
systems involved.

All of this points out very clearly what a lot of laser people have known =
for
ages=85  Lasers are great for precision tasks that involve alignment of th=
ings
to very high accuracy (interferometers to measure or position things to
millionths of an inch) or delivery of tiny amounts of energy to locations =
that
can=92t be accessed any other way (surgical instruments to weld detached r=
etinas
within the eye=85  but their use to deliver significant amounts of energy
requires a large laser and an expensive beam handling delivery system that
makes them uneconomical except in exceptional circumstances.  And that doe=
sn=92t
seem to apply in the case of a spider.