Site:
Topographical overview map of the Cao Phong research site, Hoa Binh Province, Vietnam

Cao Phong

Province:

Hòa Bình

GPS:

20.756065, 105.249208

Soil Type:

Orthic Acrisols

Indigenous Group:

Mường

Prototype Status:

Testing Phase - Completed

"Can an earthen block remember?" Cao Phong lies in Hòa Bình, within a region with a long history of Mường settlement, where patterns of habitation and land use have changed over time. The project is situated on two adjacent hilltops with distinct geological formations, within a tectonic fracture zone. Although the original commission called for a hotel and apartment complex, the team proposed dedicating part of the site to a geological and Mường cultural museum. This reframes the project beyond hospitality and acknowledges the land’s natural history and the knowledge, practices, and histories of Mường communities.

Location Images

No items found.
Cmd / Ctrl + Scroll to Zoom
Satellite view
Soil types ▶
Ao — Orthic Acrisols
Af — Ferric Acrisols
Ag — Gleyic Acrisols
Fa — Acric Ferralsols
Fo — Orthic Ferralsols
Fr — Rhodic Ferralsols
Ge — Eutric Gleysols
I — Lithosols
Je — Eutric Fluvisols
Jt — Thionic Fluvisols
Lc — Chromic Luvisols
Od — Dystric Histosols
Re — Eutric Regosols
Vp — Pellic Vertisols
Ethnic groups ▶
Austro-Asiatic
Vietnamese (Kinh), Muong
Khmers, Banar
Sedang (incl. Rongao)
Sino-Tibetan
Chinese (Han)
Hani (Akha, Uni…)
Hmong-Mien
Miao
Yao
Tai-Kadai
Thai, Tay (Tho)
Lao, Lu
Austronesian
Cham
Jarai, Ede

Fieldnotes

No items found.

Property History

The site is located on the side of a hill that was recently flattened for construction, while the existing surface vegetation has been preserved.

Soil Layers

No clear distinction between soil layers was observed in the 1 m-deep soil pit. Soil samples were collected at a depth of approximately 30 cm below the ground surface to minimize the influence of surface organic matter.

Sensory Tests

The soil was moist and water-saturated following recent rainfall. It felt sticky, with a noticeable gravel-like texture, and was easily moldable.

‍

Physical

Physical

Field Tests
field-tests
Lab Tests: Grain Size Distribution
lab-tests-grain-size-distribution-conductivity
Single and Mixed Fiber Comparison
single-combined-1-fiber-comparisons
Combining Fibers and Tung Oil (≤1%, 3%, 5%, 7%)
ceb-experiment-fiber-resin-mix-block-production
Mechanical Testing Methods
mechanical-testing-method
Water Resistance — Immersion, Suction, Erosion & Absorption
block-testing-water-resistance----immersion-erosion-absorption

Turning Cao Phong soil into compressed earth blocks involved more than finding a suitable mixture. The soil had to be prepared, the fibers dispersed, and the blocks pressed, removed from their molds, dried, and tested. The research moved back and forth between material formulations, fabrication methods, testing, and design.

Working with traditional natural building techniques as a starting point, the experiments investigated how natural additives could improve the mechanical strength and water resistance of compressed earth blocks under Cao Phong's tropical monsoon climate. The research used fibers and tung oil, and progressed from testing individual additives to combining them in hybrid mixtures. Block production, molds, and wall assemblies were developed.

In total, the team produced approximately 1,600 blocks and tested 426 of them, using around 15 tons of soil, 50 kg of fibers, and 15 liters of tung oil.

Field Tests

Field tests were conducted to assess the suitability of the local soil from Cao Phong as a base material for compressed earth blocks. They provided an initial reading of the soil’s potential and limitations before more controlled recipes were developed. In this way, field testing became the first step in translating local ground conditions into a material system, allowing the soil’s texture, composition, cohesion, and plasticity to be estimated. Jar test, cookie test, cigar test, pH test, and Munsell Soil Color Chart were used.

Single and Mixed Fiber Comparison

Early experiments used rice husk and sugarcane bagasse. Pineapple leaf fiber was introduced because of its longer, more branched structure and its potential to create additional internal connections within the blocks, improve bonding and reinforcement, and enhance overall mechanical and water-resistance performance. When the two fibers were mixed, visual observation suggested that the pineapple leaf fibers wrapped around the sugarcane bagasse, forming omnidirectional connections within the mixture.

However, without tung oil, the pineapple leaf fiber sample did not outperform the sugarcane bagasse sample in compressive or flexural strength (3.3 vs. 3.4 MPa; 0.5 MPa each). It also showed greater immersion loss (7% vs. 4%) and a greater spray-erosion depth (14 vs. 6 mm/hr).

Overall, among the samples without tung oil, the single-fiber samples achieved higher compressive strength (3.3–3.4 vs. 2.5–3.1 MPa) and flexural strength (0.5 vs. 0.4 MPa) than the mixed-fiber samples. In the spray test, all samples containing pineapple leaf fiber recorded higher erosion rates of 11–16 mm/hr, compared with 7 mm/hr for soil-only blocks. Immersion loss did not follow a clear pattern, and all fiber samples passed the 3-hour suction test, against 0.5 hours for the soil-only sample.

Combining Fibers and Tung Oil (≤1%, 3%, 5%, 7%)

Recipes combining sugarcane bagasse or rice husk with 3%, 5%, and 7% tung oil lost 0% in immersion, against 17% for soil-only block, and generally lowered water absorption. However, the high cost of tung oil raised concerns. Subsequent experiments therefore tested lower concentrations of 0.25%, 0.5%, 0.75%, and 1%, both separately and in combination with fibers.

When used alone, tung oil at these lower concentrations had little effect on mechanical strength, which remained between 2.7 and 3.0 MPa for compression and 0.4–0.5 MPa for flexure, against 2.7 MPa and 0.4 MPa, respectively, for soil-only blocks. Even at 0.25%, oil-only blocks recorded 0% immersion loss and passed the 24-hour suction test, but spray-erosion depth increased from 7 mm/hr for soil-only blocks to 13 mm/hr.

Combining fibers with tung oil produced different results. Without tung oil, the mixture containing 0.35% sugarcane bagasse and 0.15% pineapple leaf fiber reached 2.8 MPa in compression and 0.4 MPa in flexure. With 0.25% tung oil, these values increased to 5.1 MPa and 0.6 MPa, respectively. Immersion loss fell from 3% to 0%, while spray-erosion depth decreased from 11 to 5 mm/hr. These blocks only passed the three-hour suction test, compared with 24 hours for the oil-only blocks. Another tested mixture, containing 0.15% sugarcane bagasse, 0.35% pineapple leaf fiber, and 1% tung oil, achieved 6.0 MPa in compressive strength, 0.7 MPa in flexural strength, 0% immersion loss, and passed the 24-hour suction test.

These results suggested that performance depended on the interaction between fibers and tung oil rather than on any single component alone.

Mechanical Testing Methods

The investigation of compressed earth blocks with natural additives began with the aim of using the blocks as load-bearing elements, which requires adequate compressive strength. Depending on the design scenario, the blocks may also serve as an infill within a steel frame and therefore require adequate flexural strength. These potential applications therefore informed the evaluation of both compressive and flexural strength.

Testing Methods:

Compressive Strength: Testing and assessment of the CEBs are carried out in accordance with DIN 18945: 2024-03

Flexural Strength: Testing and assessment of the CEBs are carried out in accordance with TCVN 6355-3:2009, ARS 1333:2018

Water Resistance — Immersion, Suction, Erosion & Absorption

In northern Vietnam, water resistance is a key concern for earthen construction. Cao Phong has a tropical monsoon climate and the site is exposed to heavy rain, high humidity, strong winds, tornadoes, and landslides. Given the limited availability of comprehensive testing frameworks for compressed earth blocks in tropical climates, the research adapted procedures drawn from relevant national standards: German (DIN 18945:2024), Vietnamese (TCVN 6355:2009), and New Zealand (NZS 4298:2024). Each test exposed the blocks to water in a different way.

The immersion test showed whether the block would disintegrate when partially submerged. The suction test showed whether the block would swell or crack through capillary moisture. The spray-erosion test showed how much surface material would erode under rain-like impact. The absorption test showed how vulnerable the blocks may be to prolonged wet conditions. Water resistance, therefore, could not be expressed as a single number. It appeared as a set of distinct behaviors: absorption, swelling, cracking, erosion, and loss of mass. A block may withstand one type of water exposure and still fail another.

Testing Methods :

Absorption: Testing and assessment of the CEBs are carried out in accordance with TCVN 6355-4:2009


Immersion: Testing and assessment of the CEBs are carried out in accordance with DIN 18945:2024-03

Suction: Testing and assessment of the CEBs are carried out in accordance with DIN 18945:2024-03

Erosion: Testing and assessment of the CEBs are carried out in accordance with NZS 4298:2024, NZS 4299:2024

Socio-
Cultural

Socio-Cultural

Mường House Made in Concrete, Painted to Resemble Wood
muong-house-made-in-concrete-painted-as-wood
Perception of Construction with Nature-based Materials
perception-of-construction-with-nature-based-materials
Mường Vernacular Knowledge and Practices
ethnic-context-vernacular-techniques

Mường House Made in Concrete, Painted to Resemble Wood

Traditional Mường houses were historically built from timber harvested from surrounding forests. As access to timber has become increasingly restricted, many families have rebuilt their homes using concrete. Yet these newer houses often retain the forms, proportions, and details of the former timber structures. In this way, the house can retain the memory of its former material life even as the original practices begin to disappear.

Perception of Construction with Nature-based Materials

An interview with the project developer provided insights into the use of nature-based materials in the Vietnamese context. Natural building materials are perceived through a mix of genuine interest and pragmatic skepticism. It is acknowledged that people could appreciate the aesthetic, cultural, and ecological value of natural materials, yet such materials will remain a "small niche" for the foreseeable future, due to barriers around cost, established construction habits, and technical reliability (particularly load-bearing capacity and resistance to extreme weather). Scientific test data is seen as the decisive factor in dispelling doubts, while demonstrative projects repeatedly come up as the path toward wider adoption of natural materials.

Mường Vernacular Knowledge and Practices

Mường cultural knowledge extends across time-reckoning, agriculture, and ritual. The Đoi calendar is based on observations of the Moon and the Đoi star (the Pleiades) and is recorded on twelve bamboo strips, one for each month. Mường communities have used it to determine auspicious and inauspicious days for production (farming, fishing, hunting), daily life (buying buffalo, building houses), life-cycle rites (weddings, funerals), and community festivals. The quạt ma, used in Mường funerary rites, in which daughters-in-law perform prescribed movements with palm-leaf fans, shows how crafted objects and bodily gestures participate in mourning and obligations toward the deceased. Agricultural tools such as cái nhíp, used to cut individual rice panicles by hand, show how knowledge of cultivation was carried through tools and bodily practice.

Field visits in Mường Village and discussions with cultural researchers were done. Fieldwork informed design through a sequence of translations, from interviews and observations to drawings, studies, block geometries, molds, and wall assemblies. Through this process, the project examined how situated knowledge could alter material and architectural decisions.

Ecology-
Environment

Ecology-Environment

Land Use and Environmental Change
land-use-history-and-environmental-degradation
Ecological Context
ecological-context
No items found.

Land Use and Environmental Change

Before 1980, forest clearance and cultivation on slopes contributed to soil degradation in parts of Cao Phong, leaving grassland and shrubland used for grazing, fuelwood collection, and farming (JICA 2009, 2, 12–14, 24-25). In the wider Đà River basin, the construction of the Hòa Bình Dam (1979–1994) required local communities to resettle, including through di vén, in which households relocated to higher ground around the reservoir (Dao 2010, 324–325, 331). The Cao Phong Reforestation Project later introduced Acacia mangium and Acacia auriculiformis, with a planned 15-year rotation for timber production and land rehabilitation (JICA 2009, 3, 8).

Image 4 incorporates artwork by Đinh Thảo Linh.

‍

Dao, Nga. 2010. “Dam Development in Vietnam: The Evolution of Dam-Induced Resettlement Policy.” Water Alternatives 3 (2): 324–340.

Japan International Cooperation Agency (JICA). 2009. The Study on Capacity Development for AR-CDM Promotion in the Socialist Republic of Vietnam: Project Design Document for a Small-Scale AR-CDM Pilot Project, Version 2.1, 11 November 2008. Prepared by Nippon Koei Co., Ltd. and Sojitz Research Institute, Ltd.

Specific Project Team

MM Lab

Nguyễn Hà
Kim Hojung
Nguyễn Lê Minh Nhựt
Đoàn Duy Thái
Lê Quang Trường Giang
Vũ Tuấn Anh

‍


arb architects

Nguyễn Hà
Phùng Xuyến
Trần Ngọc Quân
Bùi Doãn Huy
Vũ Mai Hồng
Nguyễn Thị Thùy Dương
Trần Tuấn Hưng

‍

‍

Technical Support

Nguyễn Minh Vũ
Nguyễn Anh Quang

‍

‍

Bauhaus Earth

Rosa Hanhausen

Eda Özdemir

‍

‍

‍

‍