Effects of Red Light on Circadian Rhythm: A Comparison Among Lamps With Similar Correlated Color Temperatures Yet Distinct Spectrums

Abstract

Blue crest between the wavelength of 460 nm and 480 nm was reported to present melatonin suppression effects, whereas effects of red light on circadian rhythm regulation remain unclear. Spectrum plays an important role in circadian rhythm regulation, yet a lot of researches focused on the correlated color temperature, although a correlated color temperature value corresponds to various possible spectrums. Here, we performed human factor experiments with 3 lamps on 17 participants, comprising 9 males and 8 females. Our results showed that spectrums with high blue intensity tended to cause abnormal regulations of melatonin and cortisol, while the abnormalities were likely to be compensated by the 606-635-nm red light, which was indispensable for the photo-biological effects concerning circadian rhythm regulation. Abnormal circadian rhythm regulation was also found to be influenced by the illuminance, as abnormalities were significant in 500 lux whereas they were likely to disappear in 250 lux, implying the existence of threshold doses to trigger abnormities concerning circadian rhythm regulation. Furthermore, circadian rhythm responses were distinct between males and females. Our work may have implications for the development of light source, as we suggest that lighting source designers should increase the 606-635-nm intensity for bed room luminaires to decrease melatonin suppression effects.

Full text

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 1
Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000.
Digital Object Identifier 10.1109/ACCESS.2017.Doi Number
Effects of red light on circadian rhythm: a
comparison among lamps with similar
correlated color temperatures yet distinct
spectrums
Jianqi Cai1,2,*, Wentao Hao3, Shanshan Zeng2, Xiangyu Qu2, Ya Guo2, Shanshan Tang1, Xin
An1, Aiqin Luo1,*
1 School of Life Science, Beijing Institute of Technology, Beijing, 100081, China.
2 China National Institute of Standardization, Beijing, 100191, China.
3 Beijing YangMing ZhiDao Photoelectric Science and Technology Co., Ltd, Beijing 100102, China.
*Corresponding author: Jianqi Cai (e-mail: caijq@cnis.ac.cn); Aiqin Luo (e-mail: bitluo@bit.edu.cn)
This work was supported in part by the National Key Research and Development Plan under Grant 2017YFB0403700, and in part by the
Fundamental Research Fund of China National Institute of Standardization under Grant 512019Y-6685.
ABSTRACT Blue crest between the wavelength of 460 nm and 480 nm was reported to present melatonin
suppression effects, whereas effects of red light on circadian rhythm regulation remain unclear. Spectrum
plays an important role in circadian rhythm regulation, yet a lot of researches focused on the correlated
color temperature, although a correlated color temperature value corresponds to various possible spectrums.
Here, we performed human factor experiments with 3 lamps on 17 participants, comprising 9 males and 8
females. Our results showed that spectrums with high blue intensity tended to cause abnormal regulations
of melatonin and cortisol, while the abnormalities were likely to be compensated by the 606-635-nm red
light, which was indispensable for the photo-biological effects concerning circadian rhythm regulation.
Abnormal circadian rhythm regulation was also found to be influenced by the illuminance, as abnormalities
were significant in 500 lux whereas they were likely to disappear in 250 lux, implying the existence of
threshold doses to trigger abnormities concerning circadian rhythm regulation. Furthermore, circadian
rhythm responses were distinct between males and females. Our work may have implications for the
development of light source, as we suggest that lighting source designers should increase the 606-635-nm
intensity for bed room luminaires to decrease melatonin suppression effects.
INDEX TERMS Spectral Power Distribution, Circadian Rhythm Regulation, Narrow Blue Crest, Gender
Difference, Long-wavelength Red Light
I. INTRODUCTION
Ambient light affects visual comfort [1-2] and ocular
safety [3-4] through a series of neural and chemical
processes. The characterization of ambient light is
described by a set of photometric parameters, which
correspond to the featured photo-biological (visual and non-
visual) effects [5-7]. Concerning the health-related issues
caused by the ambient light, standards have listed detailed
requirements on the photometric parameters [8]. Light
source generally comprises LED and phosphor materials,
and the quality and arrangement of the light source
determines the optical performance of lamp and display [9-
11]. Light source development is accelerated by the
abundant knowledge of photo-biological effects [12-14],
therefore studies on related issues are necessary and
pressing.
Light intensity is a significant photometric parameter due
to the relation to power, and it is generally quantified as
illuminance or brightness [15-16], or the sum of photon
energy [16]. Light color is determined by the wavelength
distribution (also known as the spectral power distribution)
and also related to the visual and non-visual effects [17-18].
Visual effects are generally reflected by ocular structural
variations [19], while non-visual effects are usually

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 9
characterized by neural and hormone responses [20].
Melatonin is an extensively used indicator for non-visual
effects regarding circadian rhythm regulation, as the
melatonin content presents periodic fluctuation with clock
time [21-23]. Normal melatonin content is necessary in
sleeping, immune function and cardiovascular activities
[22-23]. Researches on light-intervened melatonin emission
have reported the suppression effects of the high-blue-
intensity spectrum [24-26].
Photo-biological effects have been reported to be
affected by correlated color temperature (CCT) in a number
of studies [27-30], whereas the fact that a CCT corresponds
to multiple SPDs tend to be ignored in previous researches
[31-32], resulting to the lack of requirements regarding
spectrum in available standards for lamps and displays. In
the fabrication of light source, the mixture of red, green and
blue phosphors contributes to the final color [33-35],
therefore a certain color corresponds to various possible
phosphor proportions and photoluminescence spectrums
[31]. Lacking in the knowledge on the photo-biological
effects of spectrum has limited the health-related
development of lighting and displaying.
In the current study, we investigated 3 lamps with similar
power and CCTs (around 5000 K) yet distinct spectrums,
by human factor experiments on 17 participants. We
analyzed the characterizations of melatonin and cortisol
emission following the use of each lamp. We were sought
to clarify the effects of similar-CCTs-distinct-SPDs light on
circadian rhythm regulation, especially the non-visual
effects of red light on circadian rhythm regulation.
II. TESTING AND MEASUREMENT METHOD
A.
AMBIENT LIGHT
The light sources used in our experiments comprised a
dorm lamp (4000 K) and 3 experimenting lamps (around
5000 K). The dorm lamp was installed as the default lighting
device of the student dorm, and participants had been
accustomed to the dorm lamp as the daily lighting. The dorm
lamp was used as the comparison baseline, while the 3
experimenting lamps were used in our experiments for
comparison. The lamps presented distinct spectrums (Fig.1),
although other optical performances were similar (Table 1).
Here, the CCT value of each experimenting lamp was around
5000 K, as suggested in standards and previous literatures
[8,36].
(A)
(B)
FIGURE 1. Light environments of the experiments: (a) SPD characterization
of each lamp, with the total intensity (representing the power) of each lamp
normalized to the same value; (b) lamp distribution.
TABLE I
OPTICAL PERFORMANCE OF EACH LAMP
Lamp ID
CCT (K)
Blue Peak
Ra
Dorm Lamp
4000
450
>80
Lamp 1
5150
453
Lamp 2
5000
467
Lamp 3
5050
453 & 467
B.
SUBJECTS
In the present study (including the experimental protocols),
all of the 17 participants (university students, including 9
males and 8 females, with the age from 21 to 30 years)
provided written informed consent, and all of the methods
used were performed in accordance with the relevant
guidelines and regulations. Fundus inspection and blood test
were performed on all the subjects to confirm they were

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 9
appropriate in our experiments. No one had the habit of
alcohol, sleeping pills, caffeine or hormone-related drugs.
Each participant was required to adapt the sleeping schedule
to 22:30-6:30 around 7 days ahead of the experiments with
the lighting of dorm lamp.
C.
EXPERIMENTS
Our study lasted from May to July and comprised 2 parts:
the visual task part (19:00-22:30, with each participant seated
by the desk for printed-materials reading; any screen
watching was not allowed) and the sleeping part (22:30-6:30,
with all lamps turned off to keep dark). Participants finished
supper before 17:30 and then brushed teeth without
toothpaste. Eating was not allowed after 18:00. From 18:40
each participant entered his/her own apartment dorm for the
corresponding experiment. The visual task part was over at
22:00, and the corresponding lamp was turned off at 22:30.
Participants were required to go to bed prior to 23:00.
During the experiments, the ambient light for each
participant was constructed in his/her own room in the
apartment dorm. All window curtains were closed to screen
out the light outside the room. The lamp was placed 225 cm
above the desk, which was 75 cm in height. The lamp power
was set to have the illuminance of 500 ± 50 lux on the desk
center. Each experimenting lamp was employed for 4
consecutive evenings from Monday to Thursday, while it
was replaced by the dorm lamp in the evenings from Friday
to Sunday for the recovery of circadian rhythm regulation
which was affected by the experimenting lamp. It was
considered that 3 days was competent to eliminate the
residual effects of the previous experimenting lamp. All the 3
lamps were employed with the desk illuminance of 500 ± 50
lux, and additional experiments using Lamp 1 and Lamp 3
were performed with the desk illuminance of 250 ± 50 lux as
they presented abnormal effects. The illuminance 250 ± 50
lux is close to 300 lux, which is the threshold illuminance for
reading in hospital ward (GB/T 26189-2010) and is
considered to be harmless to participants’ visual health
within the duration of the visual task. The final rounds of
experiments on different lamps were as follows: Lamp 1: 500
± 50 lux, 250 ± 50 lux; Lamp 2: 500 ± 50 lux; Lamp 3: 500 ±
50 lux, 250 ± 50 lux.
Saliva was collected at 19:00, 22:00, 23:00 and 7:00 every
day. During the extraction of melatonin and cortisol from
saliva, each participant gargled with pure water and then kept
a piece of cotton under the tongue. After 1 min the cotton
was extracted from the mouth with tweezers and stored at -20
oC with the SARSTEDT Salivette tube (Germany) for
Enzyme-Linked ImmunoSorbent Assay (ELISA) analysis
that was used to analyze the levels of melatonin and cortisol.
For the ELISA analysis, coefficient of variation inside
BRK1519H1/H2 kit was below 10%, and the measurement
was performed using the VersaMax microplate reader (450
nm).
III. RESULTS AND DISCUSSION
Circadian rhythm regulation is affected by the ambient
light due to the non-visual effects [24]. The biological
oscillating clock in mammal generates and regulates the
physiological rhythms, which oscillates with a period close to
24 hours, due to the function of the suprachiasmatic nuclei
(SCN) locating at the brain’s hypothalamus region [24-25].
Melatonin and cortisol are significant indicators for circadian
rhythm, and their content variation with clock time was
generally employed to reflect the non-visual effects of
ambient light.
Prior to experiments with the 3 experimenting lamps, each
participant spent 7 days to adapt to the 22:30-6:30-sleeping
schedule. The dorm lamp, which was installed in the ceiling
of apartment dorm for daily lighting, was used for the
lighting during the 7-day’s adaptation as the circadian rhythm
baseline. Saliva was collected in the 7th evening from 19:00
to 7:00. Melatonin variations presented the rising trend
(females: t=5.493, p=0.001; males: t=3.594, p=0.007) from
19:00 to 23:00, and the decreasing trend (females: t=7.42,
p=0; males: t=7.055, p=0) from 23:00 to 7:00. Cortisol
variations presented the decreasing trend (females: t=12.492,
p=0; males: t=7.15, p=0) from 19:00 to 23:00, and the rising
trend (females: t=9.376, p=0; males: t=9.953, p=0) from
23:00 to 7:00 (Fig.2). Melatonin difference between 22:00
and 23:00 was insignificant (females: t=0.538, p=0.607;
males: t=1.11, p=0.299). Cortisol difference between 22:00
and 23:00 was insignificant for females (t=0.863, p=0.417)
yet significant for males (t=2.175, p=0.061).
(A)

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 9
(B)
(C)
(D)
FIGURE 2. Variations of chemical indicators contents: melatonin contents
of (a) males and (b) females, and cortisol contents of (c) males and (d)
females, following the 7-day’s schedule adaptation.
Experiments with the 3 lamps were performed following
the 7-day’s adaptation. Participants presented similar
melatonin-variation trends characterized by the rising-to-
decreasing features during 19:00-7:00 from Monday to
Thursday, except for those using Lamp 1 and Lamp 3 on
Wednesday and Thursday (Fig.3). For participants using
Lamp 1 and Lamp 3 with the illuminance of 500 lux,
melatonin suppression began to appear in the 3rd evening at
22:00 although the melatonin content soared to high level at
23:00. In the 4th evening, the suppression was more
significant as the melatonin content at 23:00 was still
abnormally low.
With the illuminance decreased from 500 lux to 250 lux,
abnormalities regarding melatonin emission disappeared
among females using Lamp 1 and Lamp 3, as well as males
using Lamp 1. However, males employing Lamp 3 still
presented melatonin suppression at 22:00 and 23:00, and
melatonin enhancement at 7:00. It was suggested that males
were likely to be more sensitive to Lamp 3 in low
illuminance compared with females.
(A)
(B)

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 9
(C)
(D)
(E)
(F)
(G)
(H)
FIGURE 3. Melatonin content variations of females (A, C, E, G) and males
(B, D, F, H) using the 4 lamps. Lamp 1 (L) and Lamp 3 (L) represents Lamp 1
and Lamp 3 with low intensity (corresponding to the desk illuminance of 250
lux).
Cortisol regulation is related to melatonin content, therefore
the cortisol variation with clock time is generally on the
contrary to melatonin variation except for the case of
continuous sleeping deprivation. In the current study,
participants using the 3 lamps presented descending-to-
rising trends of cortisol variations during 19:00-7:00 from

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 9
Monday to Thursday. Similarly, cortisol variations were
abnormal for those who used Lamp 1 and Lamp 3 on
Wednesday and Thursday (Fig.4).
For participants using Lamp 1 and Lamp 3 in 500 lux,
cortisol enhancement began to appear at 22:00 in the 3rd
evening whereas the cortisol content dropped to low level at
23:00. In the 4th evening, the cortisol enhancement
continued at 23:00. With the illuminance decreased from
500 lux to 250 lux, abnormalities regarding cortisol
variation still existed for males using Lamp 3, while all
abnormalities disappeared among males using Lamp 1 and
females using Lamp 1 and Lamp 3.
(A)
(B)
(C)
(D)
(E)

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 9
(F)
(G)
(H)
FIGURE 4. Cortisol content variations of females (A, C, E, G) and males (B,
D, F, H) using the 4 lamps. Lamp 1 (L) and Lamp 3 (L) represents Lamp 1
and Lamp 3 with low intensity (corresponding to the desk illuminance of 250
lux).
As the baseline for melatonin and cortisol variations in our
experiments, the dorm lamp was characterized by the
spectrum comprising the blue crest at the wavelength of 450
nm, and the correlated color temperature of around 4000 K.
Participants using Lamp 2 presented similar circadian rhythm
regulation compared to those using the dorm lamp. Lamp 1
and Lamp 3 were special as they caused abnormalities
regarding melatonin and cortisol variations with the
illuminance of 500 lux. In general, the melatonin contents at
22:00 and 23:00 should be the higher than those at 19:00 and
7:00; correspondingly, the cortisol contents at 22:00 and
23:00 ought to be lower. However, participants employing
Lamp 1 and Lamp 3 in 500 lux exhibited abnormally low
melatonin level and high cortisol level at 22:00 and 23:00.
Effects of Lamp 1 were especially significant even with the
decreased illuminance to 250 lux for males.
Each experimenting lamp was turned off at 22:30 in the
evening, and participants prepared to go to bed prior to 23:00.
There should be few differences in melatonin and cortisol
variations between 22:00 and 23:00. However, participants
using Lamp 1 and Lamp 3 in 500 lux in the 3rd evening
presented lower melatonin contents at 22:00 than those at
23:00 (females: t=6.625 and p=0 for Lamp 1, while t=4.712
and p= 0.002 for Lamp 3; males: t=5.595 and p=0.001 for
Lamp 1, while t=6.218 and p=p for Lamp 3), and higher
cortisol contents at 22:00 than those at 23:00 (females:
t=10.192 and p=0 for Lamp 1, while t=2.979 and p= 0.021
for Lamp 3; males: t=11.706 and p=0.001 for Lamp 1, while
t=6.856 and p=p for Lamp 3).
The 3 lamps used in this study presented similar correlated
color temperature yet distinct spectrums in the blue and red
region. Lamp 1 presented the blue crest at 467 nm, while the
blue crest of Lamp 2 was at 453 nm. Lamp 3 exhibited two
blue crests at 453 nm and 467 nm respectively. The
abnormities of circadian rhythm regulation caused by Lamp
1 and Lamp 3 were likely to be related to the wavelength
distribution, as their spectrums presented high blue
proportions, which had been reported to cause melatonin
suppression.
Melatonin suppression sensitivity was reported to
distribute in the 420-600-nm range [39-40]. Here we
calculated the total suppression of each lamp, using the
previously reported wavelength-related suppression
sensitivity [41]. Each lamp’s SPD data was normalized to
100. Single suppression at each wavelength was calculated
by the multiplication of suppression sensitivity and
normalized SPD, and the total suppression as the 420-600-
nm integral was calculated as the sum of the suppression at
each wavelength. Finally we obtained the total suppression of
the dorm lamp (with the value 29.72), Lamp 1 (with the
value 38.87), Lamp 2 (with the value 31.18), and Lamp 3
(with the value 36.04). For comparison between lamps,
ANOVA analysis was performed by comparing the
suppression effect at each wavelength using SPSS 20.0. The
suppression caused by Lamp 1 was significantly higher than
that caused by the dorm lamp (t=3.127, p=0.002) and Lamp 2
(t=2.979, p=0.003). Lamp 3 presented higher suppression
compared with the dorm lamp (t=5.134, p=0) and Lamp 2

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 9
(t=5.371, p=0). Melatonin suppression caused by the dorm
lamp was slightly lower than that caused by Lamp 2 (t=-
2.666, p=0.008). Lamp 1 and Lamp 3 presented similar
effects regarding melatonin suppression without significant
difference (t=1.614, p=0.108).
Differences still existed in the spectrum characterization
between Lamp 1 and Lamp 3, as the blue crest of Lamp 1
was at 467 nm while the blue crest of Lamp 3 was separated
at 453 nm and 467 nm respectively. Correspondingly, Lamp
1 had more significant effects on melatonin suppression
compared with Lamp 3. Males presented abnormal melatonin
and cortisol variations when they used Lamp 1 and Lamp 3
in 250 lux, while the abnormalities disappeared when they
used Lamp 3 in 250 lux. Both females and males using Lamp
1 and Lamp 3 began to present abnormalities in the 3rd
evening, whereas there seemed to be gender difference
regarding sensitivities to the non-visual effects, although we
could hardly exclude the possibility of the lacking in dose
accumulation that resulted in the insensitive responses of
females. Distinctions in effects on circadian rhythm
regulation in different illuminances implied the existence of
the threshold dose that triggered non-visual effects.
Our results showed that Lamp 3 presented higher
suppression even with the illuminance that was reduced by
half, inferring that the final suppression was affected by the
effects of both blue suppression and red compensation.
According to previous researches, the equations are likely to
be as follows [39-41]:
  
(1)
  
(2)
  
(3)
In the equations above, E represents the final suppression
effects, with S representing blue suppression, and C
representing red compensation; Iλ represents the wavelength-
dependent intensity which is normalized to constant, with Nλ
representing the wavelength-dependent sensitivity of
suppression, and Mλ representing the wavelength-dependent
sensitivity of compensation; k1 and k2 are constants. The
value of Mλ was unknown due to the lack of sufficient data,
whereas the wavelength in the range of 606-636 nm seemed
to be significant due to the spectral difference between Lamp
1 and Lamp 3 (Fig.5). For simplicity, the function Mλ within
the range of 606-635 nm was treated as M620, and the
constants k1 and k2 were set 1. As the suppression effect of
Lamp 3 was higher than that of Lamp 1, M620 was estimated
as follows according to Eq.1-Eq.3:
 
(4)
As the suppression effect of Lamp 1 was higher than that
of Lamp 2, we have:
    
(5)
The value range of M620 was estimated as 2.63<M620
<2.67. Further researches were needed to narrow this value
range and expand the data range of Mλ.
(A)
(B)
FIGURE 5. (A) Spectrums of Lamp 1 and Lamp 3 in the range of 606-635
nm; (B) Estimated range of M620 (solid spheres represent the blue
suppression sensitivity reported in previous literatures [39-41], while hollow
bars represent the range of M620).
For participants using Lamp 1 and Lamp 3, melatonin
suppression presented phase shifts as the content peak and
content trough was delayed in the 4th evening. The phase
shift was comprehensible as participants suffering from the
lack of sleep were more likely to be sleepy. In the first 2
evenings, Lamp 3 caused normal circadian rhythm
regulation due to the cumulative amount of lighting. From
the 3rd evening, abnormalities appeared as the lighting
effects got enough accumulation. Lamp 1 presented more
significant melatonin suppression due to its high intensity
in the 460-480 nm range, whereas the suppression effects
decayed with sleeping as the ambient light was sheltered by

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 9
closing eyelids. At 22:30 melatonin suppression was
weakened with the experimenting lamp turned off, and
melatonin emission was likely to increase without
continuous accumulation of lighting in the 3rd evening.
In this study, none of the participants suffered from
depression or any mental disease. Cortisol abnormities
caused by Lamp 1 and Lamp 3 were likely to be related to
the affected emotion originating from interfered sleep, as
several participants expressed the perception of drowsiness
but insomnia. For participants using Lamp 1 and Lamp 3 in
500 lux, melatonin suppression started at or before 22:00,
inferring that cortisol reduction in participants using Lamp
3 was likely to be caused following melatonin suppression.
IV. CONCLUSION
In the present study, we analyzed the effects of 3
experimenting lamps on circadian rhythm regulations of 17
participants. The 3 lamps were characterized by similar
correlated color temperature of 5000K yet distinct
spectrums: Lamp 1 with 467-nm blue crest, Lamp 2 with
453-nm blue crest, and Lamp 3 with double blue crests at
453 nm and 467 nm. Following the lighting of Lamp 1 and
Lamp 3, significant abnormalities on circadian rhythm
regulation appeared on participants with the illuminance of
500 lux, while in 250 lux abnormalities disappeared for
females yet still existed for males. Lamp 1 and Lamp 3
were special due to the narrow blue crest around 460 nm,
which had been reported to cause melatonin suppression,
while difference existed between Lamp 1 and Lamp 3 as the
red proportions presented distinct in the 606-635-nm range.
We suggest that lighting source designers should increase
the 606-635-nm intensity for bed room luminaires to
decrease melatonin suppression effects.
Our findings are concluded as follows: (1) Red light in
the range of 606-635 nm is likely to compensate for the
melatonin suppression caused by the high-blue-intensity
spectrum, and the relative compensation sensitivity was
estimated in the range of 2.63-2.67; (2) Light-induced
abnormity in circadian rhythm regulation requires enough
dose, as there seems to be a dose threshold to trigger the
abnormity; (3) Circadian rhythm reaction presents gender-
related difference, as males tend to be more sensitive to
incidence stimulus compared with females.
REFERENCES
[1] Ueno, T. et al. Approximated super multi-view head-
mounted display to reduce visual fatigue. Opt. Express.
28(9), 14134-14150 (2020).
[2] Hirota, M. et al. Comparison of visual fatigue caused by
head-mounted display for virtual reality and two-
dimensional display using objective and subjective
evaluation. Ergonomics. 62(6), 759-766 (2019).
[3] Chaopu, Y. et al. Change of blue light hazard and
circadian effect of LED backlight displayer with color
temperature and age. Opt. Express. 26(21), 27021-27032
(2018).
[4] Nie, Y. et al. Flexible Thin Film and Bulk Switchable
Relaxor Coexisting Most Optimal 473 nm Blue Light
without Blue-Light Hazard/Visual Injury. J. Phys. Chem.
C. 123(46), 28385-28391 (2019).
[5] Price, L.L.A. et al. Linking the non-visual effects of
light exposure with occupational health, Int. J.
Epidemiol. 48(5), 1393-1397 (2019).
[6] Ortega, J.C.G. et al. Negative effect of turbidity on prey
capture for both visual and non-visual aquatic predators,
J. Anim. Ecol. 89(11), 2427-2439 (2020).
[7] Manson, G.A. et al. Rapid online corrections for upper
limb reaches to perturbed somatosensory targets:
evidence for non-visual sensorimotor transformation
processes, Exp. Brain Res. 237, 839-853(2019).
[8] Lighting of indoor workplaces, ISO-8995-1-2002.
[9] Padmasali, A.N. et al. A Generalized Methodology for
Predicting the Lifetime Performance of LED Luminaire.
IEEE Trans. Electron Devices. 67(7), 2831-2836 (2020).
[10] Zhang, H. et al. Thermal and mechanical properties of
micro Cu doped Sn58Bi solder paste for attaching LED
lamps, J. Mater. Sci. 30 (2019) 340-347.
[11] Kim, S. et al. Optical properties of phosphor-in-glass
through modification of pore properties for LED
packaging. Opt. Mater. 75, 814-820 (2018).
[12] Touitou, Y. et al. Effects and Mechanisms of Action of
Light-Emitting Diodes on the Human Retina and
Internal Clock. Environ. Res. 190, 109942 (2020).
[13] Lanza, S.M. et al. Effect of a Simulated Sunset versus
Typical Indoor Lighting on Evening Melatonin Levels.
Sleep. 43(Suppl. 1), A12 (2020).
[14] Lin, S. et al. Are classroom thermal conditions, lighting,
and acoustics related to teacher health symptoms?
Indoor Air. 30(3), 544-552 (2020).
[15] Griepentrog, J.E. et al. Bright environmental light
improves the sleepiness of nightshift ICU nurses. Crit.
Care. 22(1), 295 (2018).
[16] Cai, J. et al. A better photometric index of photo-
biological effect on visual function of human eye:
illuminance or luminance? IEEE Access. 7, 165919-
165927 (2019).
[17] El-Ghoroury, H.S. et al. Color temperature tunable white
light based on monolithic color-tunable light emitting
diodes. Opt. Express. 28(2), 1206-1215 (2020).
[18] Wang, Q. et al. Bi3+ and Sm3+ co-doped La2MgGeO6:
A novel color-temperature indicator based on different
heat quenching behavior from different luminescent
centers. J. Lumin. 206, 462-468 (2019).
[19] Cai, J. et al. Imaging Quality and Fatigue Quantification
of Ocular Optical System. IEEE Access. 8, 25159-25169
(2020).
[20] Brouwer, G.J. et al. Cross-orientation suppression in
human visual cortex. J. neurophys. 106(5), 2108-2119
(2011).
[21] Coleman, M.Y. et al. Advanced melatonin onset relative
to sleep in women with unmedicated major depressive
disorder. Chronobiol. Int. 36(10), 1373-1383 (2019).
[22] Mortezaee, K. et al. Melatonin application in targeting
oxidative-induced liver injuries: A review. J. Cell.
Physiol. 233(5), 4015-4032 (2018).
[23] Slominski, A.T. et al. Melatonin: A Cutaneous
Perspective on its Production, Metabolism, and
Functions. J. Invest. Dermatol. 138(3), 490-499 (2018).
[24] Figueiro, M.G. et al. Non-visual effects of light: How to
use light to promote circadian entrainment and elicit
alertness. Lighting Res & Tech. 50(1), 38-62 (2018).
[25] Xiao, H. et al. Non-visual effects of indoor light
environment on humans: A review. Physiol. & Behav.
228(1), 113195 (2021).
[26] Levinton, J.S. et al. Temperature-related heart rate in
water and air and a comparison to other temperature-
related measures of performance in the fiddler crab
Leptuca pugilator (Bosc 1802). J. Therm. Biol. 88,
102502 (2019).

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 9
[27] Shih, S.H. et al. High efficiency color-temperature
tunable organic light-emitting diode. J. Mater. Chem. C.
48, 15322-15334 (2019).
[28] Lasauskaite, R. et al. Influence of lighting color
temperature on effort-related cardiac response. Biol.
Psychol. 132, 64-70 (2018).
[29] Veitz, S. et al. Effects of lighting with continuously
changing color temperature and illuminance on
subjective sleepiness and melatonin profiles. J. Sleep
Res. 27(5), 234 (2018).
[30] Zhang, L. et al. Generating Digital Painting Lighting
Effects via RGB-space Geometry. ACM Trans. Graph.
39(2), 1-13 (2020).
[31] Hartstein, L.E. et al. Do Dynamic Changes in Light
Level and Spectral Power Distribution Improve Acute
Alertness During Daytime? Neuropsychobiology. 16(4),
289-301 (2020).
[32] Roinila, T. et al. Hardware-in-the-Loop Methods for
Real-Time Frequency-Response Measurements of on-
Board Power Distribution Systems. IEEE T. Ind.
Electron. 66(7), 5769-5777 (2018).
[33] Baur, F. et al. Warm-white LED with ultra high
luminous efficacy due to sensitisation of Eu3+
photoluminescence by the uranyl moiety in
K4(UO2)Eu2(Ge2O7)2. J. Mater. Chem. C. 26, 6966-6974
(2018).
[34] Sreena, T.S. et al. Strong Narrow Red Emission in a
Perturbed Fergusonite System: Y3Mg2Nb3O14:Eu3+ for
White LED Applications. J. Electro. Mater. 49(4) (2020)
2332-2342.
[35] Moore-Ede, M. et al. Circadian Potency Spectrum with
Extended Exposure to Polychromatic White LED Light
under Workplace Conditions. J. Biol. Rhythm. 35(4),
405-415 (2020).
[36] Cai, J. et al. Influence of LED Correlated Color
Temperature on Ocular Physiological Function and
Subjective Perception of Discomfort. IEEE Access. 6,
25209-25213 (2018).
[37] Wullink, B. et al. Type VII Collagen in the Human
Accommodation System: Expression in Ciliary Body,
Zonules, and Lens Capsule. Invest. Ophthalmol. Vis. Sci.
59(2), 1075-1083 (2018).
[38] May, P.J. et al. Central mesencephalic reticular
formation control of the near response: lens
accommodation circuits. J. neurophysiol. 121(5), 1692-
1703 (2019).
[39] Brainard, G.C. et al. Action spectrum for melatonin
regulation in humans: evidence for a novel circadian
photoreceptor. J. Neurosci. 21(16), 6405–6412 (2001).
[40] Brainard, G.C. et al. Sensitivity of the human circadian
system to short-wavelength (420-nm) light. J. Biol.
Rhythm. 23(5), 379–386 (2008).
[41] Rea, M.S. et al. Modeling the spectral sensitivity of the
human circadian system. Lighting Res. Technol. 0, 1–12
(2011).

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2021.3073102, IEEE Access
VOLUME XX, 2017 9
Jianqi Cai acted as the director of the Visual
Health and Safety Protection Laboratory in China
National Institute of Standardization from 2010.
He has devoted to the visual human factor and
lighting health mechanism research of glasses,
lighting and displaying field. He was elected as the
Chinese deputy of the Sixth Part (Photo-biology
and photo-chemistry) of CIE, and has acted as the
leader of the Lighting Health Group of ISA. He has
also served as the chief secretary of the Medical and Health Committee of
CSA.
Mr Cai has been the principal investigator of one National Key
Technology Research and Development Program of the Ministry of
Science and Technology of China, one National Key Research and
Development Plan, and one Global Environment Fund Program.
Wentao Hao received the M.S. degree from
University of Chinese Academy of Sciences in 2014.
Now he is working in the Visual Health and Safety
Protection Laboratory as well as the Human Safety
Protection and Risk Assessment Laboratory, China
National Institute of Standardization.
His research has been focused on the measurement
and quantitative evaluation of visual fatigue of
human eyes, and the health direction of research and development of LED
luminaires.
Shanshan Zeng received her M.S. degree from
Beijing Normal University. She is currently working
with the Visual Health and Safety Protection
Laboratory, China National Institute of
Standardization.
Her main research interests include visual health and
safety protection standardization, photo- biological
effect of artificial luminaire, as well as the human
factor experimental measurements. She is
undertaking the National Key Research and
Development Plan as well as the Fundamental Research Fund of the China
National Institute of Standardization.
Xiangyu Qu received her M.S. degree from Beijing
Normal University. She is currently working with the
Visual Health and Safety Protection Laboratory,
China National Institute of Standardization.
Her main research interests include visual health and
safety protection standardization, photo- biological
effect of artificial luminaire, as well as the human
factor experimental measurements. She is
undertaking the National Key Research and
Development Plan as well as the Fundamental Research Fund of the China
National Institute of Standardization.
Ya Guo received the B.S. degree from Changchun
University of Science and Technology, China, and
the M.S. degree from Beijing University of
Technology. Now she is working in the Visual Health
and Safety Protection Laboratory of China National
Institute of Standardization.
Her main research interests include visual health
and safety protection standardization, photo-
biological effect of artificial luminaire, as well as
human factor experimental measurements. She is undertaking National
Key Research and Development Plan as well as the Fundamental Research
Fund of China National Institute of Standardization.
Shanshan Tang is pursuing her doctor’s degree at
School of life science, Beijing Institute of Technology,
Beijing, China.
Her main research interests include visual health
and safety protection standardization, photo-
biological effect of artificial luminaire, as well as
human factor experimental measurements. She is
undertaking National Key Research and
Development Plan as well as the Fundamental
Research Fund of China National Institute of Standardization.
Xin An is pursuing her master’s degree at School of
life science, Beijing Institute of Technology, Beijing,
China.
Her main research interests include visual health
and safety protection standardization, photo-
biological effect of artificial luminaire, as well as
human factor experimental measurements. She is
undertaking National Key Research and
Development Plan as well as the Fundamental
Research Fund of China National Institute of Standardization..
Aiqin Luo is the professor of Beijing Institute of
Technology, and she is the member of Teaching and
Instruction Commission of bio-engineering and bio-
technology of Education Ministry.
Ms Luo focuses on the researches on chiral
separation, molecular blotting, and separation
technologies of amino acid and protein. She also
devotes to the field of photo-biological effects of
light source on ocular physiological functions and
circadian rhythm regulation.