Blue Light at Night Circadian Hazard
Blue Light at Night Circadian Hazard
Undoctored (page 305),
on the topic of sleep, advises in an inset:
“If possible, it is helpful to avoid electronic
screens for at least 1 hour before bedtime. The screens of
modern smartphones, tablets, computers, and TVs are dominated
by light in the blue wavelength, which can ‘reset’
human circadian rhythm, causing your brain to act as if it were
bright daylight even late at night…”
Here’s more detail on this issue.
We have yet another growing post-1980
era problem with a significant adverse health effect:
artificial light at night, with a substantially
increased intensity at blue wavelengths that suppress
melatonin production. This blue light at night hazard,
by the way, is entirely distinct from eye damage caused
by the “blue light hazard”.
All of the following info is on the web in various
places, but I wasn’t able to find any single page
that pulled it all together. This paper, however, is
a great overview of the problem, and has a long list of cites:
light-at-night — a novel lifestyle risk factor for metabolic disorder and cancer morbidity
This blue light at night phenomenon is a measurable
effect. How significant is it? If you have trouble
sleeping, or sleep poorly, and/or get sleepy during
the day despite a low carb diet, you already know.
NASA is taking it seriously: Solid
State Lighting Module, SDTO 15008U (SSLM)
Don’t think you’re affected?
Consider this: despite having equal exposure to
carcinogens and toxic diets, blind women have
one half the breast cancer rate
of sighted women. Now why would that be?
Light makes you/keeps you fat?
Calories Proper: Artificial
light regulates fat mass: no bueno.
In this study, adding 4 hours to the usual
12 hours of light slammed the autonomic nervous
system, disrupting sympathetic input into brown adipose
leading to a significant increase in body fat
“despite not eating more or
2016 report in Bipolar Disorders: Blue-blocking
glasses as additive treatment for mania: a
randomized placebo-controlled trial.
Material improvements were achieved.
This is a physiological response, and there probably
are no exceptions for genotype, age or gender.
You, the reader, are affected.
Anyone who is tweaking their diet to get more
antioxidants (just to mention one trendy fad),
but doing nothing about circadian
rhythm, is fooling themselves on reducing the
risks of cancer and other ailments.
Where this ends up at is that you actually need
to do at least one of the following:
Plan A: get blue blockers and use them (trivial), or
Plan B: control all your light sources at night
(surprisingly non-trivial), or
Plan C: take melatonin (tricky), or
Plan D: revert to a total paleo lifestyle
Humans are adapted primarily to a sunlight cycle that
is red-ish at sunrise, blue sky most of the day
(weather permitting), red-ish at sundown, then
dark, with a night sky containing very low average
light levels, and no peaks in the blue. Both the
direct mid-day sun, and the sky itself
(the “north light”) have a peak in
the blue at wavelengths around 470nm. We respond
to this light cycle.
are genetic variations which affect light
sensitivity. Those who are CC or CT for SNP
rs1079610 may have heightened sensitivity at normal
indoor lighting levels. That SNP may only show up
in V3 Chip raw data.
What’s the mechanism?
Your eye has receptors other than rods and cones -
called “intrinsically photosensitive retinal
ganglion cells” (ipRGCs). They contain
melanopsin, a photopigment with a sensitivity peak
in the 460 to 484nm range (blue, matching skylight
This is a recent [re]discovery (circa 1990s, with
data back to the 1920s). One thing these ipRGCs do
is tell the pineal gland to suppress secretion of
melatonin. Low melatonin results in delayed or poor
sleep, plus other health hazards still being tallied.
Humans are adapted to some amount of light at night.
Prehistorically, this was limited to starlight and
a monthly lunar cycle. Although reported to be the
same spectral power distribution (SPD) as sunlight,
at a substantially reduced intensity, the one SPD
plot I found for starlight and moonlight was
relatively flat down into the blues (no blue peak).
We are more recently presumably somewhat adapted
to thousands of years of firelight, which is
yellowish and usually contains little or no blue light.
Later domestic lighting remained low-threat:
candles, gaslights and the ubiquitous, but now
“environmentally incorrect” Edison
tungsten filament incandescent light bulbs.
These produce little blue. The common amber
high pressure sodium street lights are also
seriously non-blue (but they are being replaced
by LEDs which, so far, are not so harmless).
Other early/mid-20th century lighting technologies
(fluorescent, mercury vapor) did indeed have more
blue light content, typically with a nasty
peak at the melanopsin wavelengths, but these were
rarely used in living spaces (other than the kitchen).
Until compact fluorescent lamps (CFLs) arrived, the
biggest blue light threat in the home was from the
blue phosphors of your color television’s relatively
small CRT, and perhaps a lone GE Circline fluorescent
over the sink.
The development of CFLs and then light emitting
diode lamps (LEDs) has changed that, and
living/sleeping space lighting with significant
blue content is rising ominously.
Both CFL (as CCFL, cold cathode fluorescent lamp)
and so-called “white light” LEDs are
also used as backlights for liquid crystal
displays (LCDs). This includes desktop monitors,
LCD TVs, “LED” TVs, laptops, many
tablets, many phones, car console displays and
some appliances. These displays leak blue even
when you think the blue is off.
It’s worth mentioning that OLED (AMOLED, PMOLED)
displays can go fully blue-dark. They are direct
surface emitters, and are increasingly used on
mobile devices. They generate red, green and blue
independently, which is both potentially the same
blue hazard, but is also a hazard more easily
mitigated than with any of the above.
Be careful of what you wish for.
For most of the 20th century, the holy grail for
architectural (and photographic) lighting was Color
Rendering Index (CRI); basically getting the
artificial light to mimic daylight.
We got there.
Yup, this is yet another novel pre-millennial
era health threat, just like the rise of hybrid
semi-dwarf wheat, HFCS, Omega 6 PUFA
industrial oils, low fat diets, etc. It’s a
threat we see rather than eat.
So what to do? Simple:
Reduce your retinal exposure to blue
light starting at sundown, or two hours prior to
planned sleep time.
Plan A: Blue Blockers
The most effective way to reduce your retinal
exposure to blue light is right at the eyeballs.
Get blue-blocking eyewear and don it at dusk or
-2 hours, continuously until bedtime (and for
any later bathroom trips).
of short wavelengths alters sleep and the ipRGC pupil response
Your best bet here is probably the Honeywell
UVEX brand, which can be had inexpensively from
Amazon and similar resellers. If you don’t wear
glasses, consider the S9133X.
If you do wear glasses, get the S0360X.
These work. At this household, the blue light
sources either become green or go completely
dark when viewed through these glasses (I was
wearing a pair when this article was first authored).
Chris Masterjohn favors Solar Shield brand in Amber,
which are even lower cost.
Not Just a Central Vision Issue
It’s important to make sure that no blue light
sneaks around the frames. The ipRGCs sites are
apparently all over the retina, including in the
peripheral vision. Your eyes are responding to
skylight even when you are looking mostly at the
ground. If necessary, wear a hat to close off
the high sneak path.
Be cautious about buying random brand blu-blockers
or just sunglasses with amber lenses. They may or
may not effectively block the crucial wavelengths.
Plan B: Curtail the Blue Emitters
Do Plan A immediately, and work on Plan B as
opportunities develop. It may turn out that there
is some transdermal (skin) response to blue light,
in which case eyewear isn’t a 100% solution. One
trial suggested a transdermal effect, but wasn’t
Do not rely on your eyes to determine the blue
content of white or non-blue sources. Research
the SPD for the lamps where possible. Due to
the complexity of human color vision (principally
metamerism), trust the SPD chart, and not your
own lyin’ eyes. A pink light source, for example,
usually contains a lot of blue, greens can,
violets are certain to. I can think of a way a
digital camera (capable of custom white balance)
might be used to home test light sources for blue,
but won’t get into it in the present article in
the interest of space.
Get rid of blue light sources in your bedroom.
Do it now. This includes clocks, night lights,
weather radios, etc. For anything you can’t
get rid of right away (smoke detector), cover
the blue or green lamp or display with some amber,
orange or red filter material. Roscoe medium amber
filter gels do the trick. It doesn’t take much
blue light (time or intensity) to disrupt sleep.
Get rid of other blue night lights around the home.
Switch any electroluminescents from blue to green
if you can. Put in red or amber bulbs where that’s
an option. Put filter material over all the other
myriad blue light sources (appliances, digital
clocks, home entertainment gear, etc.)
You are probably already in the process of
replacing your incandescent bulbs as they die off.
Go directly to LED,
and stick with quality LEDs for which SPD charts
can be found that confirm low blue, or at least
non-peaky blue. Most of the LED bulbs presently
on the market have nasty blue light peaks.
Desirable products include Cree brand
soft whites, which do
not have a blue peak. Cree consumer products
are at the moment a Home Depot exclusive in the
US, but can be mail-ordered. If price is no
object, consider the Philips Hue technology,
which have discrete, variable, red, green and
blue sources inside. You can program up to 50
Hue bulbs from one hub using your smart phone,
and actually set them to routinely ramp down
the blue at night. GE is now entering this
market with their Align branding, and it is
promoted as a circadian aid.
Yard lights: have the power company replace that
mercury vapor eyesore with high pressure sodium,
or something else with no blue (or just get rid
of it - your local astronomers will thank you).
Dealing with the Digital Display Blues
For a very few screen types (CRT, DLP,
“laser”, plasma, OLED,
and LCD with local-dimming discrete RGB LEDs),
fully turning down the average blue level in
the signal actually does fully extinguish the
blue light emitted from the screen. An app
can be useful with these display types. Getting F.lux
into the signal path to your conventional TV screen
could be a challenge, however. Getting F.lux for an
Iphone is a problem at the moment, but one site
speculates that Apple
may be working on a standard solution.
F.lux still won’t fix blue indicator LEDs on your
reader/tablet/phone. Put a filter or black tape on
those, wear blu-blockers, or just stop using them
at night. Consider the blue light issue when
choosing a replacement device. See footnote
on this mess.
A bigger problem is that the display you are
probably reading this on is NOT one of the above
technologies with fully dimmable blue. It’s
probably an LCD screen with CCFL or “white
LED” backlight unit (called a BLU, no irony
intended). Your so-called “LED TV” is
merely an LCD TV with an LED BLU. This was
originally written on an older LCD with a CCFL BLU.
Typical BLUs are most likely at full intensity
(including full blue) at all times. When you dial
down the blue, you are at the mercy of LCD
technology limitations, which cannot reach
full black (terrible black level is one reason why
videophiles are lamenting the passing of
plasma TV). An LCD pixel is actually a triad
of pels (picture elements) consisting of R,G
and B micro color-filters, and three liquid
crystal light valves, which use variable
polarization to vary the light allowed
through. Fully polarized is as dark as a
pel gets, and it’s not sufficiently black
to block enough of the blue. TN vs. IPS
vs. VA doesn’t matter much here. LCD just doesn’t
get truly black (that’s why higher end sets
have “local dimming” or “full
Monitor makers are introducing “low
blue” models with a blue wavelength
shifted toward green, but if you care about
daytime color accuracy, this is not a wise
choice. You won’t even get 100% sRGB gamut,
much less 100% NTSC or 100% AdobeRGB, and
forget completely about BT.2020.
As we wrap up plan B, for extra credit:
when you redecorate your home, don’t paint
the bedroom blue.
Plan C: Take Melatonin and Tryptophan
Although you can (in most places) buy
melatonin supplements, it presents challenges,
not to mention risks. Like most dietary
supplements, it’s not standardized or
routinely tested (other than by consumerlab.com).
Dialing in an exact dosage is tricky. There are
potential side effects. Safety has not been
determined for various cohorts like children,
pregnant women, people with kidney/liver disease,
As a medical non-professional, my prescription
would be: don’t go there unless you have no
other choice, or controlling blue light at
night doesn’t help. Undoctored
has some advice starting on 306. The need for
arise in cases of irregular shift work, or
trying to switch between day sleep on workdays
and night sleep on weekends. Melatonin is not
in my view a complete fix for the circadian
hazards of shift work or jet lag. Shift work,
by the way, is classified by IARC as a group 2A
carcinogen all by itself. The shift premium is
no prize. Own
Out of curiosity, I took a look at what
consumerlab.com (subscription site) had to say
about the supplement
approach for circadian problems,
specifically taking melatonin and tryptophan.
Melatonin is available in credible formulations,
and CL has tested them. Dosage requires some
dialing-in. There is a tangential concern about
BSE risk if animal-sourced (no reported incidents).
Tryptophan is another matter. CL hasn’t even
published test results for products due to an
inability to be confident about potential
contaminants. They recommend instead that
you get it from foods, and listed quite a
number. Eggs top that list, but it takes a
kilogram of egg to get the typical 1 gram
dose of tryptophan.
Plan D: Paleo Time Machine
Revert to a simpler pre-tech existence.
A time before Facebook (and its extra-toxic
blue color scheme).
So Simple a Caveman Could Do It.
Yeah, that would work.
Dark Means Dark
It doesn’t take much light pollution in the
sleeping space to provoke measurable
neurological responses. Here’s a recent trial
showing performance impairment beginning
somewhere between 5 and 10 lux:
in fMRI brain activation
during working memory performed
after sleeping under 10 lux light
For some perspective, that would be like having
5 to 10 candles burning in the room.
Ancestrally, sleeping under a clear sky
overhead full moon is a mere 1 lux.
Conversely, get light exposure in the morning.
Sunlight on the face is ideal (without eyewear).
Don’t look directly at the sun, of course, but allow
some direct sunlight to fall on the periphery
of the retina. If you can’t get sun, investigate
artificial light sources that mimic solar SPD,
including the blues and possibly into the uV.
|Footnote: Why are
blue LEDs indicators so pervasive?
they were rare and expensive, and became
trendy as they became practical. Designers
think they look “cool”, and indeed
they do, in the color rendering sense. Early
(1960s+) commercial LEDs were able to emit
infrared through green relatively easily.
Blue took much longer (1994). Why? LED
fundamentally depends on quantum mechanics.
The light arises from electrons falling
between energy levels at a P-N (diode)
junction of semiconductor materials, which
materials are ideally transparent. The
specific band gap, limited to discrete
values (quanta), gives rise to the wavelength.
The available energy quanta choices are
limited by the periodic table, the expense of
some rare earth materials, and processing costs
today most “white” LEDs are not
discrete R-G-B LEDs mixing to white, but
single-color (perhaps even UV) LEDs, which
then filter, reflect or re-emit the generated light
via a phosphor. Most white LEDs are little
more than fluorescent lights based on a
phosphor excited by an LED rather than by
ionized mercury vapor. Blue LEDs may or
may not be actual magnesium-doped gallium
nitride pure blue LEDs. However the blue
is generated, faddish gadget makers need
to get a message: stop using blue LEDs,
and stop using white LEDs with more than
trivial blue content.
Return to article