Effects of Green Plants on the Indoor Environment and Wellbeing in Classrooms—A Case Study in a Swedish School

Abstract

Many schools in Sweden lack a proper indoor environment due to, e.g., poor thermal-envelope properties, overcrowded classes, poor visual appearance and insufficient ventilation. This study aims to explore the integration of a large number of indoor green plants into classrooms’ environments. This case study consists of three parts: measurements of the indoor environment including a final energy model, a questionnaire to the pupils with questions about their well-being and qualitative interviews with teachers. The case was two classrooms in a secondary education facility in central Sweden with an average annual temperature of 3 °C and a long and dark winter period with snow. The results showed 10% lower CO₂ and slightly higher and more stable temperatures due to the green plants. Worries about climate change and war among the pupils decreased after several months with the plants and worry about infectious disease increased. The teachers experienced fresher air from the plants and used the plant stands for a flexible classroom design. The conclusion is that indoor plants have the potential to contribute to a better indoor environment, but due to the high number of uncontrolled variables (including the effect of COVID-19) in measurements of real-life conditions, more studies are needed.

1. Introduction

This article takes a holistic approach to sustainability in the school indoor environment, including environmental, economic and social sustainability. Improving the indoor environment in schools is in line with the following Sustainable Development Goals: good health and well-being, quality education and sustainable cities and communities.

A large part of Sweden’s building stock was built during the 1960s and 1970s, which includes preschools and elementary school buildings. At that time, energy efficiency and the indoor environment were not such high priorities, but this view has changed due to the climate effect and since people are spending more than 90% of their time indoors [1]. The schools that were built at that time (among other buildings) were later renovated, insulated and air-tightened, which was the reason for the “sick houses” phenomena. According to a study conducted in 2003 [2], seven out of ten municipalities in Sweden have had problems with sick house syndrome in one or more of their elementary schools and the problem seems to increase over time. The causes of the problems are inadequate ventilation, high humidity and the construction of the buildings. That the buildings are used incorrectly, e.g., with too many students in the classrooms, is also believed to have an impact [2].

The public health agency of Sweden (Folkhälsomyndigheten) stated that carbon dioxide above the level of 1000 parts per million (ppm) should be seen as an indication of poor ventilation [3]. According to a study on the indoor environment in Swedish schools in 2009 [4], 16% of 118 studied schools in Sweden reported higher values [5]. Similar results were reported in schools in Poland, Spain and Greece [6,7]. This is seen as an indication that the ventilation is not satisfactory. A survey that was conducted in Sweden in 2014 showed that 85% of the preschools to high schools had approved obligatory ventilation controls. Still, over 40 percent of the schools have complaints about ventilation. From the comments in the survey responses, it can be read that this does not only apply to the air quality on the premises but also to other aspects that link to the ventilation such as noise, drafts and temperature [8].

Previous studies have shown that indoor plants can have an impact on the indoor climate and peoples’ health and well-being [9]. Besides the home, schools are the place where children spend most of their time. They are also more vulnerable to airborne pollutants compared to adults [10]. The presence of plants in the school environment is often low or non-existent. Introducing plants into the school environment may have higher significance in temperate climates (or colder) such as in northern Sweden due to its long cold season with its snow cover and short daytime, especially during the school year (mid-August to mid-June).

The National Aeronautics and Space Administration (NASA) started a project in the 1970s to address and prevent the problems of polluted indoor air with the help of plants. Twelve indoor plants were studied for their ability to purify the air from pollutants such as benzene, trichloroethylene and formaldehyde. The study showed that the plants were specialized in purifying the air from different substances as well as transforming carbon dioxide to oxygen, thus improving the indoor environment [5]. Wolverton [11], who was the leader of that project, stated that plants appear to be able to challenge and supplement technical ventilation solutions. This would mean that the indoor air in homes, schools, and other buildings can be improved simply by providing the rooms with well-chosen plants and thus helping to solve health problems related to poor air quality.

Several experimental studies also showed that contact with nature can promote human well-being, both in the form of restorative effects, as well as the promotion of cognitive functions, productivity, mental mood and the reduction of stress and anger [12,13]. Indoor plants have been shown to improve performance in an office environment [14,15], both the well-being and performance of people in school environments as well as quicker recovery in hospitals [16,17,18,19]. A follow up by Han and Ruan [9] of Bringslimark, Hartig and Patil [16], with the inclusion of Chinese studies, showed that the most important effects of indoor plants on participants were increasing positive emotions and reducing negative feelings, followed by reducing physical discomfort. They mean that even a couple of plants within three meters give an effect [9].

However, studies in this area need to consider the complexity of the context. For example, when school students are asked about what they consider to be important factors for a good work environment at school, the physical environment is often mentioned [20]. When visual methods have been used to let students describe important factors for health and learning, students have taken pictures of the physical environment and discussed, among other things, the importance of colours on the walls and the presence of paintings [21]. In the same study, students also photographed indoor plants as a suggestion of a better learning environment (Unpublished results, Warne). The psychological benefits of indoor plants on students in classrooms have been reported in junior high schools [22,23]. The view of six indoor plants resulted in stronger feelings of preference, comfort and friendliness as compared to the control group. Furthermore, the experimental group had significantly fewer hours of sick leave and punishment records due to misbehaviour than the control group [22].

To our knowledge, there is a lack of studies that look at both the physical and psychosocial environments in a classroom, both of which are important factors for the SDG goals. The purpose of this study was to explore the effects of integrating green plants into classroom environments and whether such an integration could improve the indoor climate by removing carbon dioxide during photosynthesis, reducing airborne particles and increasing humidification. The study also investigated if green plants could contribute to lower ventilation airflow and thus lower energy use. Furthermore, the study aimed to study the effects of the green plants on student well-being, the teachers’ experiences of the project and its effects on the working environment and pedagogy.

2. Materials and Methods

The case study consists of three parts: measurements of the indoor environment including a final energy model, a questionnaire to the students with questions about their well-being and qualitative interviews with teachers. The process and different parts of the study are described in more detail below.

2.1. Setting

The case was a secondary education facility in the municipality of Östersund that is located in the middle of Sweden with a latitude of 63.2° N (see Figure 1). It has a temperate climate, which includes an average annual temperature of 3 °C and a long and dark winter period with snow cover from November–December until March–April; about half of the academic year. The school itself is located in one of the villages outside the city centre within the municipality. Forests and freshwater bodies surround the village, and the amount of traffic in the area is relatively low.

The school includes several buildings that provide education for children from pre-school to the 9th grade. The focus of this study was on the two 8th grade classes, with 25 and 26 pupils each in the age range of 14–15 years old. Each of the 8th grade classes has its own classroom; there, the pupils spend most of their education time. The educational day starts at 08:25 and ends at 15:00 from Monday to Friday. However, some activities such as sports and meals take place in other parts of the school. During these times, the classrooms are left unoccupied. The two classrooms have a similar size and are located in a common corridor, as shown in Figure 2. The corridor also includes two other classrooms belonging to the 9th grade pupils. One is similar in size to the 8th grade classrooms, and the second is a bit larger.

In this study, green plants were installed in the two 8th grade classrooms. These classrooms were used to study the effect of green plants in the education environment, which hereafter will be referred to as Green classroom A and Green classroom B, as illustrated in Figure 2. The 9th grade classroom with a similar size was used as a Control classroom in this experiment and will hereafter be referred to as Control classroom. All the classrooms are connected to the same corridor and share the same heating system and the same ventilation system with a similar ventilation channel. The classrooms have similar windows facing the same direction and receive a similar amount of solar radiation. The windows remain closed at all times and the classrooms rely on the ventilation system for fresh air. The Green classroom A is the only one located at the corner of the building with two facades facing the outdoor environment. The Control classroom is adjacent to a staircase in the North direction.

All the classrooms include tables, chairs, a teacher’s stand, whiteboards, curtains for shading from solar radiation and some cabins next to the walls, as illustrated in Figure 3a,b. All the classrooms have large window façades facing the west and artificial light formed of sets of fluorescent lamps in the ceiling, as illustrated in Figure 2.

2.2. The Design Process

The green plants were installed in four different plant stands (see Figure 4a,d); two in each of the 8th grade classrooms. The design process of the plant stands was carried out in a project work course on the Industrial Design program at Mid Sweden University by third-year students during the autumn semester of 2019. The students worked in four groups. Each group was given the task of designing a working plant stand prototype according to the following design criteria:

• It should be a robust design and be safe to use in classrooms by teenagers.
• It should be a watertight design, eliminating any possible leakage of water into the classrooms.
• It should be designed to be low maintenance with an automatic irrigation system and a water tank that can accommodate at least one week of water usage.
• Materials should be chosen to be as sustainable as possible.
• The cost of the design should be as low as possible since it is intended to be used in school environments, which often have a limited budget.

The results of the design process were four unique prototypes with the following characteristics:

• Waterfall: This design is made of metal sheets and includes three trails of water at three height levels that are slightly tilted. The green plants are located on the trails, just as they were purchased. No soil is needed. Water is pumped from a water tank at the bottom of the stand to the top trail. Then, by gravitational force, the water flows to the other side of the trail and then falls down to the trail below. The water flow continues, passing through all three trails and then back to the water tank. When passing through the trails, the water flows through all plants and is absorbed from the bottom of the plants by capillarity forces. A timer connected to the water pump regulates the duration of the water flow.
• Green Working Space: This stand is made of wood and includes five levels of wooden trays: four on each side and a long one on the top. Each wooden tray is sealed with a plastic sheet and filled with soil. The plants are planted in these different wooden trays. In the centre of the stand, there is a seating place, and a wooden board above that if folded down creates a working place that includes a table. Water is pumped from a water tank at the bottom of the stand to the top tray. Then, by gravitational force, the water flows through diffused pipes along the trays to the bottom tray, and by that irrigates all the plants. A timer connected to the water pump regulates the duration of the water flow.
• Plants Pillars: The plants are planted in soil, each in its own sealed plastic sewage pipe. A water tank is located at the bottom of the stand. Water is lifted to the top of the stand using a water pump and then runs down due to gravity and fills plastic bottles located within the stand at different levels. The water is transported by capillarity forces from the plastic bottles to the soil of each sewage pipe via a cotton rope. The pillars are made of Cardboard and wood and stand on four small wheels. They can be moved around the class and are connected to the electricity via electric contacts sitting along with the fluorescent lamps (see Figure 3). A timer connected to the water pump regulates the amount of water provided.
• Plants Picture: This design resembles a wall picture. It is made of wood and includes LED lamps around the frame for creating a cosy atmosphere around the stand. The plants are planted in soil within long plastic plant boxes. The plant boxes are located at five height levels, with two boxes at each level. The plant boxes are slightly tilted toward the viewer so the plant can stick out from the stand and conceal the boxes above them. Water is pumped from a water tank at the bottom of the stand to its top. Then, by gravitational force, flow down in drip irrigation pipes from the top plant box to the lowest and by that provide water to the plants in all the boxes. A timer connected to the water pump regulates the amount of water provided.

2.3. The Green Plants

In total, 310 green plants were installed, which include: 90 spider plants, 70 peace lilies, 60 Saint George’s swords and 90 Philodendrons (see Figure 5). The plants were mixed in all four designed stands. Peace lilies and Saint George’s swords are plants that grow upward. Spider plants grow downward, and Philodendrons can climb up. These different characteristics were used to create a vivid design. The selection of the green plants was done using the following criteria:

• They should be part of the NASA list of air purification plants [5].
• They should be part of the list of allergy-free indoor plants [24].
• They should have air-purifying abilities.
• It was important to use plants that are robust and not sensitive to changes in conditions, e.g., soil moisture level.
• They should be available in the local market.

2.4. Measurements of the Indoor Environment

Monitoring of the indoor environment was done by high-quality off-the-shelf sensors that were assembled into one measurement device. Three such measurement devices were built, one for each classroom. They were installed in the following classrooms: Green classroom A, Green classroom B and the Control classroom, as illustrated in Figure 2. The measurement devices were installed at about 2.5 m above the floor to be out of reach of the pupils. Each measurement device includes:

• A robust IP67 box with 3D printed casings. The IP67 box shields the sensors from any external interference but at the same time allows them to collect the environmental information through its shell.
• An SPS30 optical particle sensor that uses laser scattering and Sensirion’s contamination resistance technology. It measures both PM_2.5 and PM₁₀.
• A Sensair S8 carbon dioxide sensor (CO₂) based on infrared technology.
• A BME680 air quality sensor that provides measurements of ambient air temperature, relative humidity and barometric pressure.
• A VEML6030 ambient light sensor with a digital 16-bit resolution.
• A Raspberry Pi 3 A+ as the main hardware (small computer). It has built-in 802.11b/g/n capabilities and can utilize and read a wide range of different sensor hardware.

A set of values from all the sensors’ measurements were imported every minute with a simple python program running on the cloud system, which required a virtual private network (VPN) connection to the building web portal API. The cloud system was realized and implemented as a single server running InfluxDB time-series database for storing and organizing the sensor values, as well as a basic website for visualization using the open observability platform Grafana. The database receives data via a simple password-protected representational state transfer (REST) interface, which both the python program for the existing building sensor and the complementary sensor boxes utilize.

2.5. Final Energy Modelling

The VIP-Energy software was used to model the final energy use of the entire classroom floor, as illustrated in Figure 2.

Table 1. Values used for modelling the classrooms.

	U Value			School Time	Night Time
External walls	0.265	W/(m2 K)	Ventilation flow	7 L/(s person)	0.1 L/(s m2)
Windows	1.2	W/(m2 K)	Body heat	90 W/person	0 W/m2
External roof	0.179	W/(m2 K)	Electricity use	9.4 W/m2	0 W/m2

lists the thermal conductivity values of the building’s different elements. Since the classrooms were on the second floor, similar temperatures on both floors were assumed and therefore no heat transfer between the floors. Different values were used between school time and night time for ventilation flow, metabolic heat from the pupils and electricity use. The final energy use was modelled with and without the effect of the green plants.

The potential for reducing ventilation flow ${\dot{V}}_{Air\_reduced}$ due to the integration of green plants was calculated using a CO₂ mass balance in Equations (1)–(3). Equation (1) represents the classroom without green plants with ventilation air flow ${\dot{V}}_{Air}$ and the concentration of CO₂ in the inflow of fresh air $PP{M}_{In}$ (atmospheric concentration) and outflow air $PP{M}_{Out}$. C is a constant conversion factor and G represents the CO₂ generation rate of molecules by the pupils and teachers in the classroom.

$${\dot{V}}_{Air}·\left(C·PP{M}_{In}-C·PP{M}_{Out}\right)+G=0$$

(1)

Equation (2) represents the classrooms with the green plants. R is the removal rate of CO₂ molecules by the green plants and $PP{M}_{Out\_GP}$ is the concentration of CO₂ of the outflow air.

$${\dot{V}}_{Air}·\left(C·PP{M}_{In}-C·PP{M}_{Out\_GP}\right)+G-R=0$$

(2)

Equation (3) represent the classrooms with the green plants but with a reduced ventilation air flow ${\dot{V}}_{Air\_reduced}$ and with similar CO₂ concentrations as the classrooms without green plants.

$${\dot{V}}_{Air\_reduced}·\left(C·PP{M}_{In}-C·PP{M}_{Out}\right)+G-R=0$$

(3)

Solving Equations (1)–(3) provides a ratio between the reduced ventilation rate with green plants to the ventilation rate without green plants, as shown in Equation (4). Using this ratio, it is possible to model the potential reduction of ventilation heat loss with the integration of green plants.

$$\frac{{\dot{V}}_{Air\_reduced}}{{\dot{V}}_{Air}}=\frac{PP{M}_{In}-PP{M}_{Out\_GP}}{PP{M}_{In}-PP{M}_{Out}}$$

(4)

2.6. Questionnaire

Pupils in both Green classroom A and Green classroom B were asked to answer a questionnaire on four different occasions during the school year of 2019/2020, after contact with the head of the school and the teacher responsible for each class. One of the researchers distributed the questionnaire at the first and second measurements and was available to answer any questions from the pupils. At the third and fourth measurements, a teacher distributed the questionnaire since the COVID-19 pandemic restrictions prohibited the researchers from accessing the school premises. In total, there were 51 pupils. In the respective classrooms, there were 25 (13 girls, 12 boys) and 26 (13 girls, 13 boys). Most of the pupils were 14 years old and two of them were 15 years old. The measurements were conducted on 22 November 2019 (49 pupils participated), 11 February 2020 (47 pupils participated), 2 April 2020 (46 pupils participated) and 2 June 2020 (45 pupils participated).

The questionnaire included the following scales and items: Psychosomatic problems (PSP) was measured using the 8 questions previously used in the health behaviour in school-aged children study (HBSC) [25]. The response alternatives were on a Likert scale: 1 = always, 2 = often, 3 = sometimes, 4 = rarely and 5 = never, and the resulting scale ranged from 8 to 40 with a higher score indicative of less PSP. The Positive Health Scale (PHS) [26], validated for use with teenagers 12–16 years old by Warne, Snyder and Gillander Gådin [21] was also used. The scale measures positive mental and cognitive aspects of health: being alert, happy, relaxed, creative, decisive, finding it easy to concentrate, feeling well, having energy and being social. The questions are worded “In the past 6 months, how often have you…” and continue with different aspects. The response scale ranges from always to never. The scale ranges from 0 to 36, with a higher score indicative of more positive health.

School stress was measured with the question “If you think about how it has been in recent months, have you felt stressed about schoolwork?” This was a modification of the question used in the HBSC study [27] (p. 43). Worries about climate change, war, infectious disease and unemployment were measured with the questions: “Do you usually feel worried about the following things: (a) Climate change in the world (b) That there will be war (c) Infectious diseases, which spread rapidly (d) Unemployment in society”. The response options for school stress and worry were on a 5-point Likert scale: 1 = always, 2 = often, 3 = sometimes, 4 = rarely, 5 = never. The scale was reversed in the analysis so that a higher value indicates more stress and worry. Sick absence was measured with the question “How often have you been away from school in recent months because you were ill or feeling unwell?” and truancy with the question “How many times have you played truant in the last few months?” The response options for sick absence and truancy were on a 6-point Likert scale: 1 = never, 2 = occasionally, 3 = once a month, 4 = two to three times per month, 5 = once a week, 6 = several times per week. The questions about worry, sick absence and truancy have been used previously by Gillander Gådin (unpublished) in a study about health among school-aged children aged 12 to 16 years old and have been approved by the Ethical Review Board in Umeå, Sweden (Dnr 09-179M).

Statistical analysis was performed in SPSS version 27, with a significance level at <0.05. Because of the relatively small sample size and non-normally distributed data, non-parametric tests were used. Descriptive statistics were presented as median and interquartile ranges. Differences in each variable between the different measurements were analysed with the Friedman test. Data from the first measurement were recorded on the group level only; therefore, comparisons between the first and second measurements were done using the one-sample Wilcoxon test.

Informed consent was obtained from parents for the pupils that were under 15 years of age. Pupils older than 15 years provided informed consent themselves. The pupils were informed that their participation was voluntary even though their parents had given consent. The project adhered to ethical standards and national law and obtained ethical approval from the Swedish Ethical Review Authority (Dnr: 2019–04009).

2.7. Interviews with Teachers

Qualitative interviews were conducted to explore teachers’ experiences of how the green plants affected the classroom environment physically and psychosocially. In total, eight teachers worked in the classrooms with the green plans, and of these five participated in individual interviews. Two of the participating teachers were mentors for the two 8th grade classes and had about nine classes a week with 60 min each in the classrooms with the green plants. The other teachers that were interviewed had lessons in the same classrooms on some occasion every week. Both male and female teachers were interviewed, and they had been working in the case study school for a period of a few years to 23 years.

First, the two mentors were invited and interviewed, then they passed on the invitation to their colleagues, and three more teachers were interviewed. The interviews were held by two of the authors using a videoconference that was recorded. The participants were initially informed about the voluntariness of their participation and that they could withdraw at any time. They were also informed that the material would be handled confidentially and that it would not be possible to identify individual statements in the presentation of the results. The interviews lasted for 27–59 min. An interview guide was used that centred around questions about the experience of the project, the indoor climate and the work environment. One of the researchers conducted the interviews while the other took notes and asked follow-up questions and then the roles were switched.

The material from the interviews was transcribed and analysed with thematic content analysis inspired by Braun and Clarke [28] that identifies, analyses and reports patterns in data. Two of the authors read the transcriptions individually and created preliminary themes. The preliminary themes were then compared and found to be relatively similar. A comparative analysis was conducted between themes and text, and subthemes were created in dialogue. The analysis resulted in three themes.

3. Results

3.1. Indoor Environment

3.1.1. Hourly Variations in Indoor Temperature and CO₂ Levels

Figure 6 illustrates the CO₂ levels and indoor temperatures in a classroom over one day. The high levels of CO₂ correspond to times when the pupils are in the classrooms, while the low levels correspond to the times of the day that the classroom is empty. The indoor temperature varies in a similar pattern to CO₂. The ventilation flow is reduced from 17:00 to reduce heat loss. At 06:00 the ventilation flow rate increases, which results in a lower indoor temperature due to a higher inflow of cold outdoor air. The indoor temperature increases again when the pupils arrive at school at 08:25. For both CO₂ and indoor temperature, the number of peaks, their height, width and their times over the day depend on the activities in the classrooms and can vary for each day.

3.1.2. Indoor Temperature

Figure 7 illustrates the indoor temperature in histogram form, from the lowest to the highest values for the two classrooms with the green plants and the Control classroom. The figures include only data measured during school time (08:25–09:00) and exclude weekends and vacations. A comparison among all the classrooms shows that the plants seem to increase the indoor temperature by 0.8 °C on average, which is illustrated by the displacement of the peaks in the histogram. The measurements in the classrooms with the green plants also had a lower temperature span of 2 °C in comparison to the Control classroom without the green plants with 4 °C (Figure 7).

3.1.3. CO₂ Levels

Figure 8 illustrates the carbon dioxide levels during school time in histogram form for all three classrooms. The figures include only the data measured during school time (08:25–09:00) and exclude weekends and vacations. Two peaks were measured: one with an atmospheric level of CO₂ (of about 410 ppm), which probably corresponds to periods with unoccupied classrooms, e.g., during lunch and sports activities. The second and higher peak represents the time when pupils were present in the classrooms. The CO₂ levels between the two peaks could correspond to periods where there were only a few pupils in the classrooms, or, alternatively, transit periods when pupils enter and leave the classrooms. The comparison between the classrooms with the green plants and the Control classroom shows that the presence of the green plants can reduce the carbon dioxide level by about 70 ppm on average when the pupils are in the classrooms, which corresponds to 10% of the peak value. The Control classroom had more events in which the CO₂ levels exceeded the recommended 1000 ppm level on a few occasions.

3.1.4. Humidity, Light Level and Airborne Particles

Figure 9 illustrates the average values of humidity, light level and airborne particles during school time and during night time for all three classrooms. No significant effects were found during school time, but during the night when the ventilation flow is low, there was higher humidity in the classrooms with the green plants and higher levels of airborne particles in all the classrooms. The level of airborne particles was low in all the classrooms, probably due to the school’s location in the countryside. The reason for the differences in light level is not clear, but may be due to the use of curtains to block sunlight to a greater extent in the classrooms with green plants.

3.2. Final Energy Modelling

Figure 10 illustrates the modelled specific final energy supply of the classrooms in three scenarios: without green plants, with green plants and with green plants and reduced ventilation flow due to the uptake of CO₂ by the green plants. The results show that the plants could reduce the heat supply from 132 kWh/(m² year) to 125.5 kWh/(m² year) due to the increased temperature illustrated in Figure 10, which is an energy reduction of 5% of the total heating demand (corresponding to 430 kWh on average per classroom and year). If the ventilation flow was also reduced to achieve a similar level of carbon dioxide to that which existed before the plants were installed, the heat supply would be further reduced to 109 kWh/(m² year), which is an 18% lower energy use of the total heating demand (equivalent to 1556 kWh on average per classroom and year). The amount of solar energy, personal heat and electricity is constant and not dependent on the green plants.

3.3. Plant Stand Design Performance

The performances of each of the four plant stand designs differ in terms of efficiency of irrigation, aesthetics and functionality. These differences are listed in

Table 2. The advantages and disadvantages according to the teachers’ experiences with each of the four plant stand designs.

Planstand	Advantages	Disadvantages
Waterfall	This plant stand has an outstanding function. It is easy to replace plants because they remain in their pots. The plant container is “clean” (simple, clean and functional) throughout its construction.	There are fingerprints on the sheet metal sides. From the beginning, there was a filter (coffee filter), but it is gone, and the soil washed into the tank. Sometimes a leaf settles over the drain hole and then there can be water flooding.
Green Working Space	It is experienced as very beautiful. People who enter the classroom react. The standing table and seat are used all the time and are much appreciated by the pupils. They may take turns using them. Convenient with only one power connection.	The water supply was low in the three lowest plants levels. On these, only the Saint George’s sword survived. It would have been better if the drip irrigation was not buried in the soil so that it would have been possible to see if they were providing enough water.
Plants Pillars	Everyone thinks it is a nice design.	In the beginning, the pumps did not work due to a gap in the transformer. The plants began to wither immediately. After the electricity was moved from the floor to the top of the plant container, the irrigation seemed to work. The plant stand is large for the space, which makes it difficult to furnish well.
Plants Picture	The drip pipes are located on top of the soil and water better than in the large Green Working Space plant stand design.	The long plastic plant pots are hung in metal hooks that have damaged the pots. The design of the plant stand makes it difficult to use, partly due to having two water tanks, and also because the water tank and electricity, located under the bottom compartment, are difficult to access.

as they were perceived by one of the mentor teachers.

3.4. Student Well-Being

The effects of the introduction of plants into the classrooms were analysed on a group level between measurements 1 and 2. The results showed that truancy (p < 0.001) and worry about unemployment (p = 0.009) were significantly lower at the second compared to the first measurement (results not shown).

Table 3. Median and interquartile range (IQR) for the variables at four different measurements and results of the Friedman tests for measurements 2–4, presented with n (df), chi-square value and p-value.

		Friedman Test
	Median (IQR)	n (df)	Chi-Square	p-Value
Psychosomatic problems (PSP) 1		39 (2)	7.09	0.03
Measurement 1	20 (16–25)
Measurement 2	19 (16–24)
Measurement 3	17 (14–21)
Measurement 4	19 (15–23)
Positive health 2		39 (2)	0.83	0.66
Measurement 1	30 (26–34)
Measurement 2	31 (27–35)
Measurement 3	31 (28–34)
Measurement 4	31 (27–33)
School stress 3		39 (2)	8.82	0.01
Measurement 1	3 (2–3)
Measurement 2	3 (3–4)
Measurement 3	3 (2–3)
Measurement 4	3 (2–4)
Worry about climate change 3		39 (2)	11.74	0.003
Measurement 1	2 (2–3)
Measurement 2	2 (1–3)
Measurement 3	2 (1–2)
Measurement 4	1 (1–2)
Worry about infectious disease 3		39 (2)	14.4	0.001
Measurement 1	2 (1–3)
Measurement 2	2 (1–3)
Measurement 3	3 (2–4)
Measurement 4	3 (2–4)
Worry about unemployment 3		39 (2)	2.74	0.26
Measurement 1	2 (1–2)
Measurement 2	1 (1–2)
Measurement 3	1 (1–2)
Measurement4	1 (1–3)
Worry about war 3		39 (2)	6.32	0.04
Measurement 1	2 (1–3)
Measurement 2	2 (1–2)
Measurement 3	2 (1–2)
Measurement 4	1 (1–2)
Sick absence 4		39 (2)	3.8	0.15
Measurement 1	2 (2–2)
Measurement 2	2 (1–2)
Measurement 3	2 (2–2)
Measurement 4	2 (1–2)
Truancy 4		39 (2)	5.35	0.07
Measurement 1	1 (1–3)
Measurement 2	1 (1–2)
Measurement 3	1 (1–1)
Measurement 4	1 (1–2)

¹ The scale ranged from 8 to 40 with higher values indicative of less PSP. ² The scale ranged from 0 to 36 with higher values indicative of more positive health. ³ 1 = never, 2 = rarely, 3 = sometimes, 4 = often, 5 = always. ⁴ 1 = never, 2 = occasionally, 3 = once a month, 4 = two to three times per month, 5 = once a week, 6 = several times per week.

shows the students’ responses to the questionnaire at the different measurement points. The median values for truancy were “never” and for sick absence “occasionally”, at all measurements. Median values for worry were most often “rarely” or “never”; however, worry about infectious disease had the median response “sometimes” at measurements 3 and 4. The median value for school stress was “sometimes” at all measurements. The pupils responded 30–31 out of 36 on the scale for positive health, indicating high positive health. The values of psychosomatic problems (PSP) were between 17–20 (Table 3). Changes on the individual level were analysed from measurements 2 to 4. There were significant differences between the measurements for PSP (p = 0.03), school stress (p = 0.01) and worry about climate change (p = 0.003), war (p = 0.04) and infectious disease (p = 0.001). PSP and school stress were lower at measurement 3 compared to measurement 2, but somewhat higher again at measurement 4 (Table 3). Worry about climate change and war were lowest at measurement 4; however, worry about infectious disease was highest at measurements 3 and 4 (Table 3). Positive health, truancy, sick absence and worry about unemployment showed no differences between measurements 2–4.

3.5. Teachers’ Experiences

Three themes emerged from the analysis of the interviews with five teachers: a feeling of freshness and cosiness, a sustainable design for plants and students and involvement in taking care of the plants.

3.5.1. A Feeling of Freshness and Cosiness

The teachers said that the air felt fresh when they entered the classroom with indoor plants. They felt that the air was also cooler, but they added that they did not know if this was because of their expectations on the project or if the air really was better. The teachers appreciated the look of all the greens and the change from a poor indoor environment to a cosier one. They said that the students were enthusiastic from the beginning and that they felt involved. After a while, the feeling was a bit normalized; it became like every day. The teachers questioned whether there was a real effect of the plants or whether it was more about them wanting to experience an improvement in air quality. One teacher answered the question of if they had noticed any effects from the indoor plants:

[I] think if I were to ask the students, they would probably say no […] I may have felt it a little at first, but it was very difficult to know what is what, if you found it cosier, if that made it feel like it was a little better or if it really was the air that got better. It’s a little hard to tell the difference, somehow.

Additionally, the teachers reported that the students had discussed if what they felt related to the plants was real and would be shown in the measurements or whether it was only their imagination. The students were critical and questioned the study, like youths use to do, said one teacher:

And like I said, they are teenagers so they are a little sceptical at times. Just like I said earlier, ‘But can this really be something that can be measured?’ and ‘What do they see in this?’ and ‘We’ll read out letters, can it really do anything?’ [laugh]. Yes. It is in itself, I think, a little healthy that they are a bit questioning.

3.5.2. A Sustainable Design for Plants and Students

The results from the interviews with the teachers, related to the design, were both about the functionality of the plant containers to provide the plants with water and light, and how the plant containers contributed to creating a pleasant classroom environment where new study environments emerged.

Some of the containers did not work as intended. They gave the plants too little water. In others, there were a lot of flies. This resulted in several corrective actions from one of the researchers and the problems also affected the teachers’ and students’ experiences of the plants. The flies were annoying, and the students worried about the withering of the plants. The teachers discussed how the problems should be handled.

It has messed a lot with the irrigation systems, and so on. Because it was part of the project [to test different systems], that it [the plants] would be self-watering, you would only need to fill the large containers at regular intervals and so on. […] They were like the pillars of our classroom […] And it has been a lot of trouble, so [names the researcher] has been here quite a lot in the evenings and fixed [laughs] to make it work. So it turned out that it… That [responsibility] has probably been a little more on him [laughs] than I thought at the beginning, that the students could fix with the plants.

By furnishing with the plant installations, new study environments could be created. Several of the teachers highlighted the flexibility that the plant containers provided. The new environment contributed to flexibility for those who needed to go out or stand up for a while or take a rest on the couch.

… And in the lessons, there are some like, ‘yes, but I have to stand for a while, I’m so restless’. Yes, but go and stand there at that standing table [beside the plant pillars] and work. Bring your Ipad or your math books and stand there and work.’ It has become a natural workplace that you would like to have more of.

The installations gave new opportunities to create privacy and opportunities for concentration for the students, but not all the teachers used them.

3.5.3. Involvement in Taking Care of the Plants

This theme describes the ambivalence of being involved or not in the caretaking of the plants and teachers’ thoughts about using plats in their pedagogy. The fact that the plants were not doing well during some periods worried some students and the teachers said that the students experienced the flower flies as disturbing. Several of the teachers thought about whether the students should be more involved, but the uncertainty in the research approach hindered that to some extent. One of the teachers who has been most involved thought a lot about the methods in the research project and was somewhat worried to do something that could destroy the experiment.

The problems with the flower flies are something that is mentioned at the beginning of the spring term, while the problems with plants that may get too little water or light remained longer. Some teachers have taken more active responsibility and, together with the students, sought the answer to certain problems:

“Because everything has worked so well, the students have had a commitment spontaneously, ’but that plant doesn’t seem to be doing well, should we not do something?’ But, they are a bit like, ‘yes, but we’ll go and see what it could be due to’.”

The researchers asked about the teachers’ views on the opportunity to use the plants in different subjects, but they said that this had not happened in this project to the extent they might have thought or hoped for. It was partly because there was uncertainty about what to do and not to do so as not to disrupt the research, but some also expressed that it just did not happen.

4. Discussion

The overall aim of this study was to explore the effects of integrating green plants into schools’ indoor environments. The effects were studied regarding (a) if green plants integrated into such environments could improve the indoor climate by removing carbon dioxide during photosynthesis, reducing airborne particles and increasing humidification; (b) if green plants could contribute to lower ventilation air flow and thus lower energy use; (c) if green plants can affect student well-being, and (d) the teachers’ experiences of the project and its effects on their working environment and pedagogy.

4.1. Results Discussion

The results of the different parts are discussed, followed by a summary/concluding discussion.

(a) Measurements of the indoor temperature showed that the classrooms with the green plants had slightly higher temperatures with lower variations over time, which is in contradiction to the view that plants provide a cooling effect by evapotranspiration [29]. The reason for the slightly higher measured temperature (0.8 °C) is not clear and more measurements are needed to determine the mechanism, but it could be: (i) the additional thermal mass in the classrooms due to the plants, soil and stands; (ii) higher absorption of light energy by the plants, soil and stand; (iii) due to exothermic reactions in the plants themselves. The lower variations and higher temperature can improve the level of thermal comfort or alternatively reduce heating demand.

(b) The green plants in this case study did not seem to have lower levels of air-born particles in comparison to the Control classroom. However, the levels of air-born particles were low in all the classrooms both for PM₁₀ and PM_2.5 particles. These levels did increase with the lower ventilation airflow after school hours. The school in this study is located in a rural area surrounded by forests and rivers. According to the World Health Organisation (WHO) [30], PM₁₀ levels in northern Europe countries are low and not exceeding 20–30 μg/m³ even in urban areas, but it would still be interesting to perform a similar study in schools located in urban centre areas. A study that was done in a school in Taborstrasse, Vienna [19], with much higher particulate concentrations indoors had varying results depending on the indoor plants that were used. A combination of plants, as in this report, resulted in lower particulate matter and lower CO₂ levels.

The public health agency of Sweden (Folkhälsomyndigheten) recommends that the ventilation airflow in schools and childcare facilities should not be less than about 7 litre/s per person [3]. In addition, they also state that if the carbon dioxide content of a room during normal use regularly exceeds 1000 parts per million (ppm), this should be seen as an indication that the ventilation is not satisfactory [3]. CO₂ levels in the classrooms with the green plants were found to be 10% lower in comparison to the Control classroom, which supports the findings from related studies [19,31]. Classrooms have relatively high ventilation airflows due to the high density of pupils and teachers during education times. The lower CO₂ level indicates that green plants could be used to counteract the lack of ventilation airflow, or alternatively reduce airflow, and thus decrease heat energy losses from the ventilation system. An energy model of this specific case study predicts 18% lower heating needs if green plants were used to reduce ventilation flow without exceeding the recommended level of CO₂. This assumes that the level of air-born particles also remains low.

(c) The results from the questionnaire showed some changes in pupils’ health and well-being after the introduction of green plants in the classrooms (less truancy and worry about unemployment). For about seven months with the plants, there were changes in PSP, school stress and worry about war, climate change and infectious disease. For PSP and school stress, one could see a certain decrease when measured after five months with green plants in the classrooms. However, this positive effect of plants in the school environment disappeared when the last measurement was performed two months later at the end of the spring semester. Worry about climate change and war were lowest at the end of the spring semester; however, worry about infectious disease was highest during the spring. These are conflicting results with what, e.g., Bringslimark et al. [16] and Han and Ruan [9] concluded in their literature reviews, that indoor plants increase positive emotions and reduce negative feelings, followed by reducing physical discomfort. However, the articles included in these reviews represented mostly university students and not only pupils/junior high school students. On the other hand, Fjeld [32] compared the effects of indoor plants on the reduction on health and discomfort symptoms from office personnel, radiology department employees and school children and found similar effects in all three study groups. The school children reported 21% fewer problems compared to the Control classroom without indoor plants; this figure for office personnel was also 21% and, for radiology department employees, 25%. This means that the effects on children and grown-ups can be quite similar. Another explanation of the differences can be related to the sample size. Only two of the 50 studies in Han and Ruan’s review [9] had fewer than 50 participants. The differences between our study and previous research are not easy to explain. Studying the health effects of a classroom intervention among students is complex because many factors affect them simultaneously. Studies included in the review by Han and Ruan [9] showed that the effects of plants differed by the type of plants (colour, scent) as well as the number of plants and placement and distance to the plants. For example, lower anxiety was found among students who were seated close to the plants compared to the classmates who were further away from the plants [9]. In our study, worry about climate change and war were lower after several months with the plants at the end of the study. Since there were many plants in the classrooms, all the students could see them, which may have contributed to this positive effect. However, no lasting effects were found on school stress in our study, which conflicts with the findings by Han [33], who showed that the stress-reducing effect of plants increased when students at a school were responsible for the care of the plants compared to those who only watched them. This highlights the importance of the degree of involvement during the implementation process; for example, the degree of involvement of students and school staff needs to be investigated when studying the introduction of plants into the school environment. In the present study, the teachers were unsure about how they and the students could be involved in taking care of the plants and therefore they did not use the plants in their education, which may explain the absent effect on school stress. The fact that school stress, truancy, worry about unemployment and PSP showed signs of improvement after the introduction or after a few months with the plants but not at the end of the study could possibly be attributed to the COVID-19 pandemic. When this survey was carried out, Sweden was shut down due to the worldwide COVID-19 pandemic. This likely dominated the students’ everyday lives to such an extent that the stress-reducing effects of the plants no longer appeared. Most probably, the pandemic had an impact on the worry about rapidly spreading infectious diseases that increased during the spring.

Regarding the choice of plants in classroom environments, it is important to consider the risk of allergy or other unwanted reactions to the plants. Therefore, plants with a scent or colourful plants may not be a good choice even though these can be perceived as positive attributes by some. The design should be carefully considered to create good conditions for the students as well as the plants. In our study, one design worked better than the others, which made the plants in the less appropriate designs fade and attract flies. This may have had an effect on the results of our study. For the introduction of green plants in classrooms to work over time, it is important to consider the systems for providing water and light to the plants.

(d) The teachers found that the plants made the air feel fresher and the previously poor indoor environment cosier. The plant stands contributed to a flexible design and new environments could be created to promote concentration and privacy for the students. However, some plants faded and flies were annoying. The physical classroom environment is one important domain that can facilitate or impede the performance of students and teachers [34,35]. Classroom relaxation stations are suggested by Maich et al. [36], who defines these as safe places in the classroom where the pupils can relax when they need. The authors describe it as a cosy corner, accessible but with easy access and also shielded. Such a place can facilitate self-regulation and thereby reduce student stress [37]. Plant installations might be used as such relaxation corners. In her study, Fjeld [32] also found that high school students reported the classroom with plants as more spacious than the control room.

As mentioned above, the teachers did not use the plants in their teaching. Based on one goal in the SDGs, (4.7.1) education for sustainable development should be mainstreamed in the curricula. Indoor plants in the classroom might be a way to learn about indoor climate, sustainable lifestyles, energy use and climate change.

4.2. Methodological Considerations

4.2.1. Indoor Measurements

There are a few aspects regarding the indoor measurements that need to be clarified and might have an effect on the results. First, the measurement box was located about 2.5 m above the floor. It would have been better to be located at 1.5 m at the pupils’ height. The higher level was chosen to be out of reach of curious pupils and reduce the chance of interference.

Second, pre-measurements of the indoor environment before the plants were installed were not possible. Such measurements could be compared with the post-measurements and provide additional information. However, the period of the measurement could also have an effect on the results, due to seasons and changes in the classrooms such as the number of pupils.

Third, this study did not measure the effect of the number of plants. In addition, there was no use of additional light for the plants. Therefore, more research is needed to find the optimal conditions that provide the highest benefit.

The results of the psychological and cognitive impacts of plants from the two classrooms were merged to get a big enough research group. That may even out the effect since the problems with the plant health and flies were only in one of the classrooms.

4.2.2. The Questionnaire Study

Bringslimark, Hartig and Patil [16] pointed out that there is reason to critically consider results that point to the effect of green plants on humans indoors due to disparate methods that are difficult to compare. They [16] also discuss that the possible psychological benefits attributable to green plants in indoor environments are more complex phenomena that are difficult to capture in context. As regards the effects of green plants on health and well-being, this study was planned with an experimental approach with repeated measurements in the form of a questionnaire. However, the experimental approach of the survey was mixed with the participation and activity of both pupils and teachers around the installation of the plants, which make it increasingly complex to try to determine what results can really be attributed to the plants themselves.

A shortcoming of the study is the lack of a control group. It was estimated that pupils in 8th grade, ages 14 to 15, could constitute an appropriate selection. After finding a suitable school for collaboration, it turned out that the school’s two grades in 8th grade, a total of 51 pupils, were needed for a sufficiently large experimental group. It would take some resources and time to seek collaboration with yet another school for 8th grade control groups where no plants had been installed. Using 9th grade pupils at the school in question as a control group for the study was deemed insufficient as 9th grade pupils have achieved a higher general maturity that would not be equivalent to 8th grade pupils. Therefore, this study should be considered a pilot study.

Naturally, well-being among pupils is affected by numerous factors on different levels and relating to different areas in their lives (family, friends, media, etc.), and it was not possible to include all the contributing factors in this study. The variables in the questionnaire were not controlled for other factors that may have influenced the results other than the green plants. Therefore, the changes in the outcome variables cannot be interpreted as caused by the plants only.

The results from the questionnaire study rely on pupils’ self-reports about questions such as truancy and sick absence. It would have been more objective to obtain the school’s records for these variables, but, unfortunately, this was not possible. Self-reported data can be affected by social desirability bias, which is another limitation of this study.

Furthermore, it must be bore in mind that studies in school classes are not based on a random sample but rather a set of groups where all the pupils in a classroom influence each other, which means that individual behaviour or performance cannot be considered independent [38,39]. Thus, when studying school classes, the conditions for independent observations are not fulfilled [40], which means that it may be more difficult to detect the experimental effects of green plants as in this study. In addition, it is more difficult to control all the factors that may affect the results when conducting studies in the real world, so-called field studies, compared to conducting studies in a laboratory environment, as Thatcher et al. [41] showed. When they replicated laboratory studies showing positive psychological effects of indoor plants in field studies, these effects were lost [41]. However, an advantage of conducting field studies is that they provide good ecological validity [40].

One of the classrooms had problems with withering plants and flies while the other classroom did not. This may have caused the results of the questionnaire study in the two classrooms to differ. The number of pupils in each classroom was not enough to conduct separate analyses with sufficient statistical power. Furthermore, measurement 1 was performed at the group level and measurements 2–4 at the individual level, which gave less power to the analysis of measurement 1.

4.2.3. The Qualitative Study

The majority (n = 5) of the teachers who taught in the classrooms with the plants participated in the interviews. The participants had a different number of teaching hours in the classrooms with the plants; some were there most of the teaching time and some were a few hours a week. Interviewing teachers with different experiences in classrooms increases the credibility of the study by taking into account different perspectives [42]. Two researchers conducted the interviews and analysed the material. It was relatively straightforward to reconcile the preliminary themes developed in the individual analyses, and there was a consensus on which content was relevant to the purpose of the study, which also strengthens the credibility. All the interviews were conducted at the end of the study in the spring of 2020 when the teachers had been in the classrooms with the plants for a long time and had the opportunity to reflect on the experiences they had during the study.

4.2.4. Impact of the COVID-19 Pandemic

During the spring semester of 2020, the COVID-19 pandemic broke out in Sweden. The study had been going on for several months at the time but had not yet been completed, and the last measurements of the pupils’ health and well-being as well as the interviews with teachers had not been carried out. It is unclear how the pandemic has affected the pupils who participated in the study, but they have likely been affected in different ways. In what way the pandemic has affected pupils’ well-being and how they responded to the questionnaire, we can only speculate. Interestingly, one of the questions queried about the worry of “communicable diseases, which are spreading rapidly”. This was one of the few variables that were significantly higher at the measurements during spring compared to the previous measurement occasions. The fact that several variables did not show any significant differences over time can be attributed to a lack of effect of the plants, but it is also conceivable that the pandemic affected health for the worse, which overshadowed any positive effects of the plants. The pandemic also affected the ability to visit the school for data collection, which is why the third and fourth measurements (in April and June 2020, respectively) were carried out by teachers instead of the researcher (KW) who conducted measurements 1 and 2.

5. Conclusions

In this interdisciplinary case study, we have included several aspects important for sustainable development and in line with the SDGs. The school is a workplace for both children and adults. How the indoor environment is designed is important for their health, especially for children, who are extra sensitive to air pollution and in need of a good learning environment, but also from an environmental perspective concerning the use of energy resources. Our results indicate that indoor plants have the potential to contribute to a better indoor environment, but due to the high number of uncontrolled variables (including the effect of COVID-19) in measurements of real-life conditions, more studies are needed. The following are the main conclusions from this study:

• Ventilation flow in buildings could be reduced with the use of indoor green plants, and thus reduce ventilation heat losses, or, alternatively, green plants could be used to improve air quality in buildings with poor ventilation.
• It is important that the plant stand provides enough water, fertilizer and light to keep the plants healthy since withered and dead plants may have a worse impact than no plants.
• The plant stands should be designed to reduce maintenance work, particularly irrigation. The dimensioning of the irrigation system is crucial for the survival of indoor plants.
• Plants and plant stands can be used to create spaces for different activities in the classroom, such as relaxing corners. Future projects could involve pupils and teachers in the design from this perspective.
• Pupils should preferably be involved in taking care of the plants. Researchers must be clear about what the teachers and pupils are allowed to do and encourage the use of the plants in education, for example, to use indoor plants as a way to work with the SDGs.
• Plants might contribute to a more pleasant classroom environment, better indoor climate and increased well-being, but more research with a long follow-up time is needed.