Optimization of D-Mannitol-Modified Magnesium Phosphate Cement Properties Using Response Surface Methodology

doi:10.16552/j.cnki.issn1001-1625.2026.0255

Abstract

Abstract:

Magnesium phosphate cement （MPC） is formed by magnesium-based raw materials （whose main components include magnesium oxide or magnesium hydroxide）， phosphates， and retarders. MPC exhibits excellent performances such as rapid setting and hardening， high early strength， superior bonding capability and good volume stability， and it has been widely applied in road rapid repair， airport runway maintenance and special structural rehabilitation. Nevertheless， the traditional retarder borax causes obvious early strength deterioration of MPC when it is incorporated， and it fails to achieve a good balance between working performance and mechanical strength. Moreover， most new retarders still exert adverse impacts on the early mechanical properties of MPC， so their retarding effect and strength development could not be coordinated well. To address the above deficiency， this study adopted response surface methodology to explore the synergistic influence and regulation mechanism of D-mannitol and borax compound system on MPC performances.

Borax and D-mannitol content were selected as experimental variables， both ranging in mass fraction from 2.50% to 5.01%. A two-factor， five-level central composite design was adopted， and quadratic polynomial regression models were established to correlate borax content and D-mannitol content with setting time， 1 d flexural strength， and 1 d compressive strength. Variance analysis results indicate that all regression models were highly significant， with high coefficients of determination and low coefficients of variation， demonstrating satisfactory fitting accuracy and predictive reliability of the models. Experimental results show that the retarding effect of D-mannitol is more pronounced at a low borax content of 2.50%~3.50% by mass. At a high borax content of 4.50%~5.01% by mass， the early strength initially increases and then decreases with rising D-mannitol content. By contrast， strength declines monotonically as D-mannitol content increases at low borax content. The optimal mix proportion is obtained by model optimization： borax content is 2.50% and D-mannitol content is 4.78%. Under this optimal proportion， the measured setting time is 29.0 min， the 1 d flexural strength is 8.6 MPa， and the 1 d compressive strength is 48.1 MPa. Multiple characterization analyses show that in the optimized composite system， D-mannitol reduces the agglomeration of magnesium oxide particles through its dispersion effect， whereas borax acts as a retarder by forming a protective film on the surfaces of magnesium oxide particles. This dispersion-film formation effect regulates pH evolution， thereby reducing the early hydration rate of MPC. Meanwhile， the optimized composite system accelerates the dissociation of the protective film at specific hydration stages， exposes active sites on magnesium oxide particle surfaces， enhances the generation of hydration products， improves the density of the matrix， and effectively promotes the early strength development of MPC.

This study clarifies the synergistic mechanism of D-mannitol and borax， and provides theoretical support and technical reference for the application of D-mannitol-modified MPC.

Key words: magnesium phosphate cement, retarder, response surface methodology, property optimization, mechanism analysis, setting time

CLC Number:

TQ172

WANG Qiuao, CHEN Feng, PENG Jinying, LI Yue, QIAN Shanshan. Optimization of D-Mannitol-Modified Magnesium Phosphate Cement Properties Using Response Surface Methodology[J]. BULLETIN OF THE CHINESE CERAMIC SOCIETY, 2026, 45(8): 2612-2623.

Figures/Tables 18

Fig.1 XRD pattern of MgO

Table 1 Composition of MgO

Composition	MgO	SiO₂	Al₂O₃	Fe₂O₃	CaO	Other
Mass fraction/%	86.17	5.02	0.64	0.46	1.90	5.81

Table 2 Test factor codes and levels

Design variable	Level
Design variable	-1.414	-1	0	1	1.414
B content （X₁）/%	2.50	2.87	3.76	4.64	5.01
DM content （X₂）/%	2.50	2.87	3.76	4.64	5.01

Table 3 Mix proportion of test

Sample No.	Design variable
Sample No.	B content （X₁）/%	DM content （X₂）/%
1	3.76	2.50
2	4.64	2.87
3	3.76	5.01
4	3.76	3.76
5	2.50	3.76
6	4.64	4.64
7	3.76	3.76
8	3.76	3.76
9	2.87	4.64
10	3.76	3.76
11	3.76	3.76
12	5.01	3.76
13	2.87	2.87

Table 4 Setting time and mechanical properties of each group of specimens

Sample No.	Setting time （Y₁）/min	1 d flexural strength （Y₂）/MPa	1 d compressive strength （Y₃）/MPa
1	16.0	8.9	51.9
2	16.5	8.2	45.6
3	32.5	8.1	43.3
4	18.0	8.7	49.3
5	17.0	9.2	52.8
6	27.0	8.2	44.0
7	17.5	8.7	48.5
8	18.0	8.8	48.8
9	28.0	8.5	47.8
10	18.5	8.6	49.2
11	17.5	8.8	48.6
12	18.5	8.1	43.8
13	13.0	9.3	54.9

Table 5 Analysis of variance for test results

Source of variance	Setting time			1 d flexural strength			1 d compressive strength
Source of variance	Mean square	F-value	P-value	Mean square	F-value	P-value	Mean square	F-value	P-value
Model	75.77	397.20	<0.000 1	0.360	55.30	<0.000 1	29.63	100.94	<0.000 1
X₁	2.67	13.99	0.007 3	1.090	166.16	<0.000 1	83.39	284.05	<0.000 1
X₂	298.10	1 562.69	<0.000 1	0.470	70.95	<0.000 1	54.40	185.32	<0.000 1
X₁X₂	5.06	26.54	0.001 3	0.160	24.35	0.001 7	7.56	25.76	0.001 4
X₁²	0.01	0.07	0.799 2	0.012	1.80	0.221 5	0.47	1.59	0.248 2
X₂²	71.51	374.88	<0.000 1	0.094	14.30	0.006 9	2.58	8.78	0.021 0
	R²=0.997， $R A d j 2$ =0.994 CV=2.20%			R²=0.975， $R A d j 2$ =0.958 CV=0.94%			R²=0.986， $R A d j 2$ =0.977 CV=1.12%

Table 5 Analysis of variance for test results

Source of variance	Setting time			1 d flexural strength			1 d compressive strength
Source of variance	Mean square	F-value	P-value	Mean square	F-value	P-value	Mean square	F-value	P-value
Model	75.77	397.20	<0.000 1	0.360	55.30	<0.000 1	29.63	100.94	<0.000 1
X₁	2.67	13.99	0.007 3	1.090	166.16	<0.000 1	83.39	284.05	<0.000 1
X₂	298.10	1 562.69	<0.000 1	0.470	70.95	<0.000 1	54.40	185.32	<0.000 1
X₁X₂	5.06	26.54	0.001 3	0.160	24.35	0.001 7	7.56	25.76	0.001 4
X₁²	0.01	0.07	0.799 2	0.012	1.80	0.221 5	0.47	1.59	0.248 2
X₂²	71.51	374.88	<0.000 1	0.094	14.30	0.006 9	2.58	8.78	0.021 0
	R²=0.997， $R A d j 2$ =0.994 CV=2.20%			R²=0.975， $R A d j 2$ =0.958 CV=0.94%			R²=0.986， $R A d j 2$ =0.977 CV=1.12%

Fig.2 Measured and predicted values of setting time， 1 d flexural strength， and 1 d compressive strength

Fig.3 Interaction effect of B content and DM content on setting time

Fig.4 Interaction effect of B content and DM content on 1 d flexural strength

Fig.5 Interaction effect of B content and DM content on 1 d compressive strength

Table 6 Mix proportion and predicted performance values of MPC

Group	B content/%	DM content/%	Setting time/min	1 d flexural strength/MPa	1 d compressive strength/MPa
BD	2.50	4.78	30.3	8.4	46.8

Table 7 Comparison of predicted values and measured values

Item	Setting time/min	1 d flexural strength/MPa	1 d compressive strength/MPa
Predicted value	30.3	8.4	46.8
Measured value	29.0	8.6	48.1
Relative error/%	4.5	2.3	2.7

Table 8 Mix proportion and performance test results for BD and B1

Group	M/P	W/C	B content/%	DM content/%	Setting time/min	1 d flexural strength/MPa	1 d compressive strength/MPa	28 d flexural strength/MPa	28 d compressive strength/MPa
BD	2	0.18	2.50	4.78	29.0	8.6	48.1	9.5	60.7
B1	2	0.18	7.28	0	10.5	7.5	41.8	8.6	52.6

Fig.6 Evolution of liquid phase pH value over time of BD and B1

Fig.7 XRD patterns of BD and B1 hydrated for 1 d

Fig.8 TG-DTG curves of BD and B1 hydrated for 1 d

Fig.9 SEM images of BD and B1 hydrated for 1 d

Fig.10 Schematic diagram of mechanism by which compound of DM and B affects MPC reaction

References 30

[1]	ZHANG G S， WANG Q， LI Y， et al. Microstructure and micromechanical properties of magnesium phosphate cement［J］. Cement and Concrete Research， 2023， 172： 107227.
[2]	石亚文，陈兵. 偏高岭土改性磷酸镁水泥［J］. 硅酸盐学报， 2018， 46（8）： 1111-1116.
	SHI Y W， CHEN B. Magnesium phosphate cement modified by metakaolin［J］. Journal of the Chinese Ceramic Society， 2018， 46（8）： 1111-1116 （in Chinese）.
[3]	秦继辉，钱觉时，宋庆，等. 磷酸镁水泥的研究进展与应用［J］. 硅酸盐学报， 2022， 50（6）： 1592-1606.
	QIN J H， QIAN J S， SONG Q， et al. Research progress on magnesium phosphate cement［J］. Journal of the Chinese Ceramic Society， 2022， 50（6）： 1592-1606 （in Chinese）.
[4]	李悦，龙世儒，王子赓，等. 磷镁物质的量比对天然水镁石制备的磷酸镁水泥性能的影响［J］. 材料导报， 2024， 38（17）： 1-6.
	LI Y， LONG S R， WANG Z G， et al. Effect of molar ratio of phosphate to magnesium on properties of magnesium phosphate cement prepared by natural brucite［J］. Materials Reports， 2024， 38（17）： 1-6 （in Chinese）.
[5]	QIN J H， DAI F M， MA H Y， et al. Development and characterization of magnesium phosphate cement based ultra-high performance concrete［J］. Composites Part B： Engineering， 2022， 234： 109694.
[6]	尤超. 磷酸镁水泥水化硬化及水化产物稳定性［D］. 重庆：重庆大学， 2017.
	YOU C. Hydration and hardening of magnesium phosphate cement and stability of hydration products［D］. Chongqing： Chongqing University， 2017 （in Chinese）.
[7]	许碧莞，张秋实，赵翊舟. 铝盐对磷酸钾镁水泥性能和水化的影响［J］. 硅酸盐学报， 2025， 53（5）： 1205-1213.
	XU B W， ZHANG Q S， ZHAO Y Z. Influence of aluminum salts on properties and hydration of magnesium potassium phosphate cements［J］. Journal of the Chinese Ceramic Society， 2025， 53（5）： 1205-1213 （in Chinese）.
[8]	范英儒，秦继辉，汪宏涛，等. 磷酸盐对磷酸镁水泥粘结性能的影响［J］. 硅酸盐学报， 2016， 44（2）： 218-225.
	FAN Y R， QIN J H， WANG H T， et al. Effect of phosphates on bond performance of magnesium phosphate cement［J］. Journal of the Chinese Ceramic Society， 2016， 44（2）： 218-225 （in Chinese）.
[9]	刘源，辛力，马思敏，等. 界面处理对磷酸镁水泥砂浆-普通混凝土界面黏结性能的影响［J］. 工业建筑， 2025， 55（5）： 270-279.
	LIU Y， XIN L， MA S M， et al. Effects of interfacial treatment on bonding properties of MPC mortar-NC interface［J］. Industrial Construction， 2025， 55（5）： 270-279 （in Chinese）.
[10]	杨楠. 磷酸镁水泥基材料粘结性能研究［D］. 长沙：湖南大学， 2014.
	YANG N. Research on the bonding properties of magnesium phosphate cement based materials［D］. Changsha： Hunan University， 2014 （in Chinese）.
[11]	王永康. 高延性磷酸镁水泥基纤维复合材料与既有混凝土黏结性能试验研究［D］. 郑州：郑州大学， 2021.
	WANG Y K. Experimental study on bonding properties of high ductility magnesium phosphate cement based fiber composites to existing concrete［D］. Zhengzhou： Zhengzhou University， 2021 （in Chinese）.
[12]	柴丽娟，岳中华，郭丽萍，等. 高延性磷酸镁水泥基快速修补材料的制备及性能优化设计［J］. 复合材料学报， 2025， 42（2）： 961-974.
	CHAI L J， YUE Z H， GUO L P， et al. Preparation and performance optimization design of high ductility magnesium phosphate cementitious rapid repair material［J］. Acta Materiae Compositae Sinica， 2025， 42（2）： 961-974 （in Chinese）.
[13]	蒋晚晚，朱金才，孟晓凯，等. 响应面法优化石灰石粉改性磷酸镁水泥抗压强度及机理［J］. 新型建筑材料， 2025， 52（1）： 51-55+128.
	JIANG W W， ZHU J C， MENG X K， et al. Optimization of compressive strength and mechanism of limestone powder modified MPC using response surface methodology［J］. New Building Materials， 2025， 52（1）： 51-55+128 （in Chinese）.
[14]	段新勇，吕淑珍，赖振宇，等. 多元复合缓凝剂制备及其对磷酸镁水泥性能的影响［J］. 武汉理工大学学报， 2014， 36（10）： 20-25.
	DUAN X Y， LYU S Z， LAI Z Y， et al. Preparation of a multi-composite retarder and its effect on properties of magnesium phosphate cement［J］. Journal of Wuhan University of Technology， 2014， 36（10）： 20-25 （in Chinese）.
[15]	杨建明，史才军，常远，等. 掺复合缓凝剂的磷酸钾镁水泥浆体的水化硬化特性［J］. 建筑材料学报， 2013， 16（1）： 43-49.
	YANG J M， SHI C J， CHANG Y， et al. Hydration and hardening characteristics of magnesium potassium phosphate cement paste containing composite retarders［J］. Journal of Building Materials， 2013， 16（1）： 43-49 （in Chinese）.
[16]	LU X， CHEN B. Experimental study of magnesium phosphate cements modified by metakaolin［J］. Construction and Building Materials， 2016， 123： 719-726.
[17]	李悦，谢梦洋，林辉，等. 钢渣对磷酸镁水泥性能的影响研究［J］. 硅酸盐通报， 2018， 37（11）： 3373-3378.
	LI Y， XIE M Y， LIN H， et al. Effect of steel slag on the properties of magnesium phosphate cement［J］. Bulletin of the Chinese Ceramic Society， 2018， 37（11）： 3373-3378 （in Chinese）.
[18]	DIT COUMES CCAU， ROUSSELET A， XU B W， et al. Investigation of aluminum nitrate as a set retarder of magnesium potassium phosphate cement： mechanisms involved in diluted suspension［J］. Cement and Concrete Research， 2021， 150： 106608.
[19]	LIU R Q， FANG B D， ZHANG G H， et al. Investigation of sodium alginate as a candidate retarder of magnesium phosphate cement： hydration properties and its retarding mechanism［J］. Ceramics International， 2022， 48（20）： 30846-30852.
[20]	戴俊，钱春香，陈竞，等. 无水乙酸钠对磷酸钾镁水泥水化性能和微观形貌的影响［J］. 材料导报， 2020， 34（6）： 66-74.
	DAI J， QIAN C X， CHEN J， et al. Effect of sodium acetate anhydrous on hydration and microstructure of potassium magnesium phosphate cement［J］. Materials Review， 2020， 34（6）： 66-74 （in Chinese）.
[21]	戴俊. 磷酸钾镁水泥缓凝剂及修补砂浆配合比设计方法研究［D］. 南京：东南大学， 2019.
	DAI J. Research on rertarder of potassium magnesium phosphate cement and mix design method for repair mortar［D］. Nanjing： Southeast University， 2019 （in Chinese）.
[22]	ZHAO J H， WANG D M， WANG X G， et al. Ultrafine grinding of fly ash with grinding aids： Impact on particle characteristics of ultrafine fly ash and properties of blended cement containing ultrafine fly ash［J］. Construction and Building Materials， 2015， 78： 250-259.
[23]	LAHALLE H， DIT COUMES CCAU， MESBAH A， et al. Investigation of magnesium phosphate cement hydration in diluted suspension and its retardation by boric acid［J］. Cement and Concrete Research， 2016， 87： 77-86.
[24]	MA H Y， XU B W， LI Z J. Magnesium potassium phosphate cement paste： degree of reaction， porosity and pore structure［J］. Cement and Concrete Research， 2014， 65： 96-104.
[25]	LI Y， LIN H. Experimental study on the effect of different dispersed degrees carbon nanotubes on the modification of magnesium phosphate cement［J］. Construction and Building Materials， 2019， 200： 240-247.
[26]	YANG J M， QIAN C X. Effect of borax on hydration and hardening properties of magnesium and pottassium phosphate cement pastes［J］. Journal of Wuhan University of Technology-Mater Sci Ed， 2010， 25（4）： 613-618.
[27]	李舒阳，陈正，韦京利，等. 三乙醇胺助磨剂对蔗渣灰颗粒特性及蔗渣灰砂浆性能的影响［J］. 硅酸盐通报， 2022， 41（3）： 931-939.
	LI S Y， CHEN Z， WEI J L， et al. Effect of triethanolamine grinding aid on bagasse ash particle characteristics and properties of bagasse ash mortar［J］. Bulletin of the Chinese Ceramic Society， 2022， 41（3）： 931-939 （in Chinese）.
[28]	王剑锋. 水泥助磨剂构效关系研究与分子结构设计［D］. 北京：中国矿业大学（北京）， 2013.
	WANG J F. Study on structure-performance relationship and molecular structure design of cement grinding aids［D］. Beijing： China University of Mining and Technology Beijing， 2013 （in Chinese）.
[29]	SUGAMA T， KUKACKA L E. Characteristics of magnesium polyphosphate cements derived from ammonium polyphosphate solutions［J］. Cement and Concrete Research， 1983， 13（4）： 499-506.
[30]	SONG T， GAO F F， DU X， et al. Removal of boron in aqueous solution by magnesium oxide with the hydration process［J］. Colloids and Surfaces A： Physicochemical and Engineering Aspects， 2023， 665： 131211.